Hunter Douglas / 247 Home Furnishings
Main Submission and Response to Phase I Decision
and Issues Statement
NON-CONFIDENTIAL VERSION
[] is confidential information
NON-CONFIDENTIAL
VERSION
2
INDEX
Section
Title
Page
1
Introduction and Summary
3
2
Jurisdiction
8
3
Counterfactual
9
4
Market Context
22
5
The Parties
30
6
Competitive Analysis
35
7
Barriers to Entry and Expansion
72
8
Neither the 2013 nor the 2019 Transactions lead to an SLC
99
NON-CONFIDENTIAL
VERSION
3
1. Introduction and Summary
1.1 The CMA’s investigation rightly focusses on questions of jurisdiction, counterfactual, a
comprehensive analysis of the competitive conditions, and barriers to entry and expansion.
When properly considered, it is clear that there is no basis for a finding of a substantial lessening
of competition (“SLC”) in relation to either the 2013 Transaction or the 2019 Transaction.
Jurisdiction
1.2 At the time of the 2013 Transaction, neither the turnover test nor the share of supply test were
met on any basis. For that reason alone, the CMA should conclude that no relevant merger
situation exists and that it therefore does not have jurisdiction over the 2013 Transaction. The
position is amplified by the fact that, even if shares of supply in 2019 are considered (as the
CMA has suggested in its Phase I Decision), the share of supply test is not met as a result of
the transaction (as is required under section 23 of the Enterprise Act 2002 (the “Act”)) but rather
as a result of an entirely unrelated transaction, namely the 2016 acquisition of Blinds2Go.
Despite being announced publicly, the CMA chose not to investigate that transaction at the time
and hence there is also no relevant merger situation in respect of that acquisition in accordance
with the time limits set out in section 24 of the Act.
Counterfactual
1.3 Even if the CMA does not agree with the jurisdictional position above as regards the 2013
Transaction, the question of the relevant counterfactual for each transaction is of key
importance. The Parties consider that the correct counterfactual is:
(i) In the case of the 2013 Transaction, the facts at that time, i.e. a situation in which Hunter
Douglas held minimal retail interests and no direct overlap with 247; and
(ii) In the case of the 2019 Transaction, the prevailing conditions at that time, i.e. a situation
in which Hunter Douglas already held certain interests and rights over the activities of
247.
1.4 When those counterfactuals are applied, it becomes clear that neither Transaction gives rise to
a lessening of competition, still less the SLC that is required under section 35 of the Act for the
CMA to find concerns at the end of the Phase 2 process. In the case of the 2013 Transaction,
there is no meaningful overlap between the activities of 247 and Hunter Douglas. In the case
of the 2019 Transaction, the change in the nature of control has no negative effect on Hunter
Douglas’ incentives regarding the activities of the 247 (as discussed further below).
NON-CONFIDENTIAL
VERSION
4
Competitive effects and barriers to entry/expansion
There is no evidence that 247 constrains Blinds2Go
1.5 If, in the alternative, the CMA considers that it is appropriate to treat the 2013 Transaction as if
it had occurred in 2019 by applying a counterfactual that includes Hunter Douglas’s unrelated
acquisition of Blinds2Go in 2016, there are no grounds to find an SLC in respect of either the
2013 or the 2019 Transactions.
1.6 In particular, the CMA has over 3 years’ worth of evidence since the acquisition of Blinds2Go,
a period in which there is no evidence of a lessening of competition, let alone a substantial one.
Despite a large increase in its sales since Hunter Douglas acquired a stake in June 2016,
Blinds2Go’s gross profit margins have remained relatively stable, and the most obvious
conclusion to draw is that rivalry from 247 has not been, and is not, an important factor affecting
Blinds2Go’s strategic decisions (including the prices it sets).
1.7 There is no evidence that the competitive constraints on Blinds2Go have changed, or will further
change, as a result of any change in the nature of Hunter Douglas’s relationship with 247. In
particular, 247 is many times smaller than Blinds2Go in the online retailing of M2M blinds, and
the CMA’s market share estimates (which the Parties believe overstate their shares) indicate
that 247’s market share has fallen from 10-20% in 2016 to 5-10% in 2019, with Interior Goods
Direct leap frogging 247 to become the second largest player. Accordingly, the competitive
pressure 247 exerts on Blinds2Go is at most limited and declining, and readily replicated by the
array of other online M2M competitors that exist, and other, important, “out of market”
constraints.
1.8 As regards out of market constraints, Blinds2Go’s growth has not been based on winning share
off other online M2M blinds retailers, but persuading customers to choose online M2M blinds
instead of cheaper ready-made blinds (which are suitable for a large part of the Parties’ sales),
in-store M2M blinds and alternative window coverings. The BDRC Survey, which was
commissioned to understand the purchase journey and choices of the Parties’ customers,
confirms that M2M blinds consumers also consider purchasing ready-made blinds and in-store
M2M blinds, as well as curtains and shutters.
1.9 247’s prominence in online search does not mean it is a close competitor to Blinds2Go
1.10 The Phase 1 Decision focused on Google pence per click (PPC) marketing and argued that
visibility in Google searches might mean that 247 was a closer competitor than suggested by
its market share. This is misplaced for several reasons:
(i) First, the BDRC Survey found that some three quarters of the Parties’ consumers spend
over an hour researching their purchase (and half spend longer than a few hours), and
NON-CONFIDENTIAL
VERSION
5
Google Trends results indicate that users searching for “blinds” are also searching in
the same search session for major retailers (such as Dunelm, Ikea, and Argos), with
247 being outside the top 20 most searched term;
(ii) Second, omni-channel retailers have a natural competitive advantage in that
consumers will frequently visit their websites, and they can run large scale email
marketing campaigns;
(iii) Third, rivals can use organic search, social media, referral or display sources for
customers and traffic (these sources collectively account for [] of both Blind2Go’s
and 247’s traffic), and in 2020 the Google Shopping tab will consist primarily of free
listings; and
(iv) Fourth, 247’s Google prominence does not translate into sales as otherwise its share
would not be 5-10% and declining. This is because consumers shop around. Indeed,
both Parties’ per session conversion rates are below [].
The market is highly competitive and 247 is not a particularly close rival
1.11 The Parties consider that the online sale of blinds is highly competitive, including a very large
number of online specialist retailers that sell directly, omni-channel retailers (such as Dunelm,
Next, and John Lewis – with Dunelm and Next entering the online sector in March 2018 and
Autumn 2019 respectively), and online platforms. The fact that Interior Goods Direct and others
have won share off 247 demonstrates that they can compete effectively for 247’s customers.
Amazon has recently introduced improved functionality to attract M2M retailers and levies
charges which are less than the Parties’ own costs for comparable services (including per click
costs, card fees and website costs), and consumer search increasingly starts on Amazon. []
1.12 The BDRC Survey indicated that very few of Blinds2Go’s customers (just []%), would have
switched to 247 in the event that Blinds2Go ceased selling blinds. More of 247’s customers
say they would switch to Blinds2Go if 247 were to cease selling blinds, but much fewer than
would be expected given Blinds2Go’s market share. It should also be noted that, whilst the
consumer survey was carefully designed, consumers might have been less inclined to consider
in-store retailers as an alternative due to the Covid-19 situation, while Dunelm’s online offering
was temporarily withdrawn in November 2019 and only recently replaced in a new, hybrid form.
1.13 In addition, []. Accordingly, there is no evidence that 247 is a particular source of price
competition to Blinds2Go.
NON-CONFIDENTIAL
VERSION
6
Barriers to entry and expansion are low
1.14 Given that there are already many competitors, the substantive question for the CMA is whether
timely expansion by existing rivals can be expected to replicate the limited competition from
247.
1.15 In this regard, the proof of the pudding is that in the UK since 2016 there has been extensive
competitor entry/expansion by Interior Goods Direct, Dunelm, Next, Makemyblinds and others
which already exceeds the CMA’s estimate of 247’s market share of only 5-10% (which the
Parties believe is in any event overstated). The ease of entry/expansion in the UK mirrors the
Parties’ experience of launching M2M online blinds companies in other countries. Going
forwards, the Parties expect significantly greater entry/expansion, in particular as omni-channel
retailers are alive to the growing trend towards online purchases and the Covid-19 crisis is likely
to accelerate the growth of their online presence.
1.16 It is clear that there are no material barriers to entry or expansion. There are many suppliers,
both in the UK and overseas, all of whom can deliver direct to the customer, with no shortage
of logistics providers to support the direct to customer model. Existing players and new entrants
can therefore operate entirely on a “drop ship” basis whereby they need hold no stock and
require no warehousing. Off the shelf websites can also be used. Indeed, this is how 247
operates.
1.17 The only barrier to expansion identified in the Phase 1 Decision is the cost of pay per click
(PPC) marketing and the CMA’s assertion that there may be economies of scale in such
advertising, notwithstanding that it is paid for on a per click basis. However, this is not the case.
There is no evidence that [].
Conclusions
1.18 It is clear that the constraint exercised by 247 on Blinds2Go is small, and that constraint is
reducing as the threat from larger omni-channel brands and platforms grows.
1.19 In short, 247 is easily replicable and there are a whole host of competitors ready to take the
place of 247 if Hunter Douglas chose to dictate 247’s commercial policy to the benefit of
Blinds2Go. Hunter Douglas’ incentives to act in such a fashion are further reduced given that
[].
1.20 Taken together, even if the CMA does not agree with a plain reading of the statutory provisions
on jurisdiction as regards the 2013 Transaction, and even if the CMA continues to apply a
counterfactual which assumes that 247 is an independent competitor to Hunter Douglas
(including Blinds2Go), the evidence shows that there can be no question of an SLC in respect
of either transaction. In respect of the 2013 Transaction, the competitive pressure exercised
NON-CONFIDENTIAL
VERSION
7
by 247 is limited and easily replicable. In the case of the 2019 Transaction, the change in the
nature of control exercised by Hunter Douglas over 247 and its interest in Blinds2Go give few
incentives to change 247’s strategy. Even if they did, there can be no question of an SLC for
the same reasons applicable to the 2013 Transaction.
NON-CONFIDENTIAL
VERSION
8
2. Jurisdiction
2.1 At paragraph 21 of the Issues Statement, the CMA rightly identifies the need to consider the
question of whether the CMA has jurisdiction to investigate the 2013 Transaction and the 2019
Transaction. As the CMA is aware, the Parties do not consider that the approach adopted in
the Phase I Decision is correct in asserting jurisdiction over the 2013 Transaction. In the view
of the Parties, that approach fails to take account of the statutory scheme and leads to perverse
outcomes that Parliament cannot have intended.
2.2 On 14 April 2020, the Parties provided the CMA with more detail on the question of jurisdiction,
setting out their view that the 2013 Transaction does not fall within the jurisdiction of the CMA.
At the time of the transaction, neither the share of supply nor the turnover test were met.
Further, to the extent that the share of supply test can be considered to be met on the basis of
the factual position in 2019
1
, this was not as a result of the transaction.
2.3 The Phase I Decision is wrong in law as it fails to take into account the requirements of section
23(2A) of the Enterprise Act 2002 (the “Act”) which states in terms that the share of supply test
is only met where the combined share of supply test is met “as a result of the enterprises
ceasing to be distinct enterprises” (section 23(2A)(a) of the Act). Consequently, no relevant
merger situation exists in 2013 and the CMA does not have jurisdiction to review the 2013
Transaction. It follows that the only question for the CMA to assess arises in respect of the
2019 Transaction (i.e. a change in the quality of control to a controlling interest) and the relevant
counterfactual in 2019 is Hunter Douglas holding at least the voting and other rights it acquired
in 2013
2
(see further Chapter 3).
2.4 The Parties do not propose to revisit the arguments set out in the paper of 14 April 2020 in this
submission but would be happy to explain their position further to the panel if appropriate.
1
The turnover test is not met on any basis.
2
As amended in 2016 and 2017.
NON-CONFIDENTIAL
VERSION
9
3. Counterfactual
3.1 The question of the appropriate counterfactual is key in defining the scope of the questions
which the CMA’s investigation will need to address. That counterfactual will differ according to
the transaction under consideration.
2013 Transaction
3.2 As the CMA is aware, at the time of the 2013 Transaction Hunter Douglas had only a very
limited, in-home, retail offer in the UK through its Thomas Sanderson brand and did not have
any online presence. 247 had turnover of £9 million in the UK and an estimated share of supply
of [] for the retail supply of M2M blinds in the UK
3
. To the extent that there was any overlap
between the activities of Hunter Douglas and 247 at that time, this would have been minimal at
best and, in the context of the retail supply of window coverings online (or any narrower subset
thereof), there was no overlap at all.
Phase I Decision
3.3 As noted above, the Parties consider that the CMA has no jurisdiction in law over the 2013
Transaction. For completeness, the Parties set out below why the Phase I Decision was also
incorrect in its description of the relevant counterfactual to the 2013 Transaction.
3.4 Clearly, the establishment of the relevant counterfactual is of critical importance in assessing
whether or not the transaction gives rise to an SLC. The Phase I Decision sets out the relevant
counterfactual as follows:
“Accordingly, the CMA considers that the counterfactual for the 2013 Transaction
should reflect the conditions of competition absent the 2013 Transaction (more
specifically, that there is a realistic prospect that 247 would have continued as an
independent market participant from Hunter Douglas) and take account of the
developments in the market (including Hunter Douglas’s own expansion) since the
2013 Transaction.”
4
3.5 The effect of this paragraph is to treat the relevant counterfactual for the 2013 Transaction as
being a situation in which Hunter Douglas is a group including Blinds2Go (which was acquired
in 2016) and Hillarys (which was acquired in 2017) and where 247 is a wholly independent
competitor to those companies. In order to take account of subsequent transactions for the
3
Supplementary Submission of 12 February 2020 in Phase I; Response to Request for Information (247)
of 20 April 2020 at paragraph 5.1.
4
Phase I Decision at paragraph 84.
NON-CONFIDENTIAL
VERSION
10
purposes of the counterfactual, the CMA relied on its own statements in Tesco/Brian Ford
5
and
Ryanair
6
and those of the CAT in BSkyB
7
to reach the conclusion that the CMA is entitled to
take account of market developments in the period 2013 – 2019 in determining the
counterfactual (i.e. including both the subsequent growth of 247 and the subsequent, unrelated,
acquisitions made by Hunter Douglas).
3.6 The role of the counterfactual is, as set out in the CMA’s Merger Assessment Guidelines
8
, to
provide “a comparison of the prospects for competition with the merger against the competitive
situation without the merger.” The Phase I Decision, in seeking to look forward to the factual
position in 2019, considers that the competitive situation without the merger is one in which:
• 247 has grown from a company with a turnover of £9m in 2013 to one with a UK
turnover of £18.7m (of which [] related to M2M blinds) for the last complete financial
year before the 2019 Transaction
9
;
• 247’s share of supply of the retail of M2M blinds online in the UK has grown to
approximately [] in 2019;
• Hunter Douglas has acquired 60% of Blinds2Go in 2016; and
• Blinds2Go has grown from a business with a UK turnover of £10.7m in 2013
10
to one
with UK turnover of £88.8m ([] of which related to M2M blinds) in the last completed
financial year before the 2019 Transaction
11
.
3.7 The Merger Assessment Guidelines make clear that the assessment of the counterfactual may
be affected by “the extent to which events or circumstances and their consequences are
foreseeable”
12
. This is the approach followed in the precedents cited in the Phase I Decision
insofar as a forward looking approach to the counterfactual is adopted, that is to say, absent
the transaction, how might competitive conditions progress in the short term, taking account of
actual data available at the time of the assessment. The Merger Assessment Guidelines
5
Completed acquisition by Tesco Stores Limited of Brian Ford Discount Store Limited (ME/3827/08), OFT
Decision of 22 December 2008.
6
Completed acquisition by Ryanair Holdings plc of a minority interest in Aer Lingus Group plc (ME/4694/10),
OFT Decision of 15 June 2012.
7
British Sky Broadcasting Group PLC v Competition Commission 2208 CAT 25
8
Merger Assessment Guidelines (CC2/OFT 1254), September 2010, at paragraph 4.3.1.
9
Year ended 28 February 2018.
10
Year ended 31 March 2013.
11
Year ended 31 December 2018.
12
Merger Assessment Guidelines, at paragraph 4.3.2.
NON-CONFIDENTIAL
VERSION
11
provide, in particular, regulation and market liberalisation as relevant factors to be taken into
account.
3.8 What the precedents cited in the Phase I Decision do not do, however, is take into account
events which were entirely unforeseeable at the time of the transaction. The unrelated
acquisition of Blinds2Go in 2016 is precisely such an unforeseeable event. In 2013, Hunter
Douglas was aware neither of Blinds2Go as an attractive investment opportunity
13
, nor of the
possibility of the later acquisition of Hillarys. Both of those acquisitions only became apparent
as possible and likely at the time that Hunter Douglas formulated its bids for those companies,
three and four years subsequent to the 2013 Transaction respectively.
3.9 The effect of the approach to the counterfactual in the Phase I Decision as regards the 2013
Transaction is to allow the CMA to look forward and consider the effect on competition resulting
from a combination of 247 and Blinds2Go in 2016
14
. However, the proper time at which to
conduct such an assessment was in 2016 in the context of the Blinds2Go acquisition. The
Blinds2Go acquisition was made public by Hunter Douglas, yet the CMA chose not to review
the transaction at the time. The CMA is not entitled to use either the 2013 or the 2019
Transactions as a proxy to review the acquisition of Blinds2Go.
3.10 Finally, the CMA relies on the OFT’s statement in Tesco/Brian Ford that “whilst Tesco was fully
entitled not to notify […], it effectively took the risk of developments in the analytical framework
applicable to this merger occurring between the time of the transaction completing and the time
of it subsequently becoming public”
15
. Yet this ignores the very crucial difference between the
2013 Transaction and Tesco/Brian Ford, namely that whilst the former was not a relevant
merger situation in 2013, meeting neither the turnover test, nor, on any basis, the share of
supply test, there was no question over jurisdiction at the time of Tesco’s completion of its
acquisition of Brian Ford. Tesco already had a share of supply of UK groceries in excess of
25% at the time and hence any accretion, however small, arising from the acquisition of Brian
Ford would satisfy the share of supply test. Tesco’s argument was more narrow – that even if
there was a relevant merger situation, the OFT would not have found any substantive concerns
based on the analytical framework adopted by it at the time.
3.11 By contrast, Hunter Douglas was not taking any merger risk by not notifying the 2013
Transaction as there was simply no relevant merger situation to notify
16
. The CMA is not
13
Hunter Douglas only became aware of the possibility of investment in Blinds2Go when it received an
information memorandum in summer 2015.
14
Note that at the time of the acquisition of Blinds2Go in 2016, Hunter Douglas acquired a 60% equity stake,
whilst it held 49% of the voting and economic rights in 247.
15
Phase I Decision, paragraph 83, citing Tesco/Brian Ford, paragraph 10.
16
See further the Parties’ Response to the Issues Letter at Phase I, section 5.
NON-CONFIDENTIAL
VERSION
12
entitled to rely on the passage of time simply to ‘convert’ a non relevant merger situation into a
relevant merger situation for the reasons set out above and in the separate submission on
jurisdiction
17
.
The Issues Statement
3.12 The Issues Statement rightly identifies the need to define a separate counterfactual for each of
the 2013 Transaction and the 2019 Transaction. The approach also correctly states that the
appropriate counterfactual should be based upon the facts available and “the extent to which
events or circumstances and their consequences are foreseeable.”
18
3.13 The starting point for the assessment is, as set out in the Merger Assessment Guidelines, “the
pre-merger situation in the case of completed mergers”
19
. In the context of the 2013
Transaction, this is, as set out above, a counterfactual in which 247 is a small, independent
online retailer of M2M blinds in the UK and Hunter Douglas’s only UK retail interest is Thomas
Sanderson, a small, high-end, traditional in-home retailer of window coverings
20
. Adopting this
counterfactual, there can be no basis for identifying an SLC (leaving aside the jurisdictional
questions highlighted in section 2 above). Depending upon the market definition retained, there
would either be no overlap between the Parties’ activities (if the relevant market were the online
retail of M2M blinds in the UK) or such overlap would be minimal (if the relevant market is that
for the retail of window coverings in the UK or any subset thereof).
3.14 The Merger Assessment Guidelines envisage a more in-depth approach to be taken at
Phase 2, with the panel examining several possible scenarios in order to determine the most
likely scenario. In each case, however, the assessment of the counterfactual is an assessment
against what would or would not have been likely at the time of the transaction. The CMA may
use subsequent events to inform itself of how the market has developed but it is not appropriate
to take account of developments which would not have been foreseeable at the time of the
transaction.
3.15 Indeed, the approach set out in the Merger Assessment Guidelines is entirely consistent with
this approach. The three examples of situations
21
in which an alternative counterfactual may
be used to the pre-merger conditions of competition are all situations which require the CMA to
conduct an assessment of the target’s competitive position. None of those scenarios require
the CMA to conduct an assessment of the likelihood of a strengthening of the market position
17
Submitted on 14 April 2020.
18
Issues Statement, paragraph 27
19
Paragraph 4.3.5
20
Thomas Sanderson was principally an in-home retailer of conservatory blinds that overlapped with 247
only to a very minor extent in respect of sales of vertical blinds on a putative wider UK blinds retail market.
21
Paragraph 4.3.7
NON-CONFIDENTIAL
VERSION
13
of the acquirer subsequent to completion of the transaction under review as a result of entirely
separate and unrelated acquisitions.
3.16 In the view of the Parties, such an approach is entirely appropriate in this case. Whilst it is
open to the CMA to take account of developments in the market since 2013 which were
foreseeable at the time, the CMA should not consider acquisitions which have occurred since
that time which were not foreseeable at the time of the 2013 Transaction. In the context of the
2013 Transaction, this means that the relevant counterfactual is a situation in which 247
continues as an independent competitor in the retail of M2M blinds online in the UK and Hunter
Douglas is a retailer of window coverings in the UK only through its interest in Thomas
Sanderson.
3.17 If the panel does not agree with that assessment and considers it necessary to include
subsequent Hunter Douglas acquisitions in its assessment of the counterfactual, it will however
need to consider 247 as an independent competitor from 2013 to 2019. To do so, it would
ordinarily be necessary to conduct an assessment of the likely market position of 247 as if the
2013 Transaction had not occurred. If not, the substantive assessment would be considering
the effect on competition of a merger between 247 (as it was in 2013) with Hunter Douglas (as
it was in 2019) which would clearly make little sense.
3.18 In this regard, the Phase I Decision speculates that the existence of veto rights might have
reduced competition from 247 even if never exercised, and that there might have been some
coordination due to some information sharing from 247 to Hunter Douglas.
“The CMA notes, in particular, that Hunter Douglas’s veto rights regarding offering
products with a profit margin below 15% and approving 247’s budget may have
deterred 247 from adopting decisions that would have made 247 a closer/stronger
competitor to HD in the supply of M2M blinds. The 2013 Transaction may therefore
have had the effect of substantially lessening competition, even if these veto rights
were not actually exercised in practice. The CMA also considers that the information
that was disclosed by 247 to Hunter Douglas via monthly management accounts may
also have dampened the competition between Blinds2Go and 247 by facilitating some
coordination of the commercial policies of two companies.”
22
3.19 However, the relationship between Hunter Douglas and 247 has been such that it is not
necessary for the CMA to engage in such speculation. The passage of time enables the CMA
to look at what actually happened in the period between 2013-2019, rather than relying on a
hypothetical forward-looking hypothesis which would ordinarily be required in a merger review.
22
Paragraph 222 of the Decision.
NON-CONFIDENTIAL
VERSION
14
3.20 As previously explained to the CMA
23
, Hunter Douglas typically looks to acquire successful
businesses and then to afford those businesses considerable autonomy in respect of day-to-
day and longer-term strategic matters. In the period 2013 to 2019, 247 was left to run its own
affairs with limited, if any, interference. The 247 founding shareholders were incentivised to
run the business as effectively as possible (the sale value contained in the put and call options
being dependent upon profit in the last two financial years preceding the exercise of the
options). The interaction between 247 and Hunter Douglas was therefore predominantly limited
to the provision of management reports containing basic financial and market information, and
to advice with regards to European expansion. None of those interactions affected the
development of 247’s market position in the UK (whether positively or negatively).
3.21 Insofar as the Phase I Decision identifies internal documents which purport to show that Hunter
Douglas exercised its material influence over 247 to substantially lessen competition between
the Parties
24
, it is not clear as to how this would be the case, as follows:
(i) The single instance of a co-ordinated price increase took place in the context of the fall
in sterling following the Brexit vote. Both Parties costs of supply therefore increased
as a result of the prevalence of imports of blinds and blind components. Such a price
rise was, therefore, inevitable and no further rises of this sort took place. In such
circumstances, it is difficult to see how any effect on competition could be considered
substantial;
(ii) [] The Phase I Decision advances no basis on which such agreements would have
an adverse effect on competition, still less that such an effect would be substantial;
(iii) In respect of TV advertising, Hunter Douglas’ concerns were twofold. First, they were
concerned that 247 was not of sufficient scale at the time for such a marketing expense
to be warranted. Secondly, []
25
. Hunter Douglas was concerned that the timing of
the TV advertising campaign would []. In short, the decision not to run a TV
advertising campaign had nothing to do with competition between 247 and B2G and
there is no evidence to suggest otherwise;
(iv) The email cited in footnote 182 in fact suggests, contrary to the position adopted in the
Phase I Decision, that []
3.22 In no way does the evidence relied upon in the Phase I Decision demonstrate an SLC. Indeed,
much of the evidence put forward does not suggest any lessening of competition, let alone the
23
See, for example, Hunter Douglas’ response to Question 5 of the Request for Information of 20 April 2020
24
Paragraph 225 of the Phase I Decision.
25
Clause 2 of each of the Put and Call Options of 3 May 2013, as explained in paragraph 7.2 of the response
to the Sixth Section 109 Notice in Phase 1.
NON-CONFIDENTIAL
VERSION
15
substantial lessening of competition which would be required under section 35 of the Act to
justify an adverse finding as regards the 2013 Transaction.
3.23 In fact, Blinds2Go and 247 have continued to compete as independent rivals from 2016 to the
present date. This includes the period between the acquisition of a 60% interest by Hunter
Douglas in 2016
26
to the 2019 Transaction (i.e. where Hunter Douglas had only a minority
interest in 247) and the period from completion of the 2019 Transaction to date. Indeed, even
if only looking at the period from completion of the 2019 Transaction to imposition of the Initial
Enforcement Orders on 21 November 2019, this still gives the CMA some nine months to
assess the actual impact on competition arising from the 2019 Transaction.
3.24 It is clear from the evidence that the 2013 Transaction did not give rise to an SLC. If there had
been a loss of rivalry in 2016, with Blinds2Go facing less competitive pressure, one would
expect that Blinds2Go would have raised its prices (and gross margin) and that such a rise in
prices would have been profitable. However, there is no evidence that Blinds2Go (or indeed
247) raised its prices after 2016
27
as gross profit margins have remained largely constant in
both the period pre- and post-2016 and that following the 2019 Transaction, as shown in Figure
3.1.
Figure 3.1: Blinds2Go and 247 monthly gross margin (January 2015 – March 2020)
[]
3.25 The Phase I Decision considers that the relative stability of Blinds2Go’s profit margins despite
a large growth in sales cannot support robust conclusions for two reasons
28
:
• Firstly, a number of other factors are likely to drive Blinds2Go’s gross margins, and
• Secondly, that Hunter Douglas exercised material influence over 247, which (for the
reasons set out elsewhere in the Phase I Decision) would otherwise have been
Blinds2Go’s most significant competitors, for the entire period covered by the data.
3.26 In respect of the first statement, a more obvious interpretation is that competition from 247 is
not a key factor driving Blinds2Go’s price setting. This is unsurprising given that Blinds2Go’s
overall competitive strategy in the UK (including how it sets prices) has not changed since
Hunter Douglas acquired a 60% stake, as set out in Chapter 5. The Issues Statement
emphasises that its theory of harm is that, due to the Transactions, there was a loss of rivalry
26
Hunter Douglas acquired a further 5% interest in Blinds2Go in 2019.
27
As the CMA is aware, there was a one-off price increase of 2% agreed between 247 and Blinds2Go to
respond to an increase in supply costs as a result of the fall of the pound following the Brexit vote.
28
Paragraph 178 of the Decision
NON-CONFIDENTIAL
VERSION
16
between 247 and Blinds2Go and this would lead to higher prices or lower quality etc.
29
Such
an outcome would be readily observable in Blinds2Go’s gross profit margins increasing if
competition from 247 reduced after Hunter Douglas acquired a 60% stake in Blinds2Go in June
2016 or full control of 247 in February 2019 (many months before the CMA contacted Hunter
Douglas).
3.27 With regards to the CMA’s conclusion in the Phase I decision that Hunter Douglas held material
influence over 247 during the period and did not therefore compete with Blinds2Go as
effectively as it otherwise would have, this makes little sense. The data in question covers the
period from 2013 to 2016, a time at which Blinds2Go was entirely independent of Hunter
Douglas. In the context of the 2013 Transaction, this shows that there was in fact no loss of
competition either at the time of acquisition or subsequently and no SLC can be demonstrated.
The relevant question as regards the 2019 Transaction is whether the change in the nature of
control further reduces competition beyond that observed in respect of the 2013 Transaction,
therefore the evidence relating to the period from 2013 is entirely relevant.
2019 Transaction
3.28 The identification of the counterfactual for the 2019 Transaction is a more straightforward
exercise. The pre-merger conditions of competition are those in which Hunter Douglas holds
de jure control over the activities of Blinds2Go through its shareholding
30
and where the 2013
Transaction “may have conferred on Hunter Douglas the ability to exercise material influence
over 247”
31
. The key question for the substantive assessment is therefore whether the change
of influence alone of Hunter Douglas over the activities of 247 as a result of the 2019
Transaction results in an SLC.
Phase I Decision
3.29 In the Phase I Decision, the CMA considered that “247 was capable of exercising some degree
of competitive constraint on Hunter Douglas absence the 2019 Transaction.”
32
This conclusion
is reached on the basis of a somewhat theoretical approach to the rights of the 247 founding
shareholders and the extent to which they could oppose action proposed by Hunter Douglas.
The Phase I Decision does not, however, identify how 247 constrained Blinds2Go in the period
2016 to 2019 in practice.
3.30 In reality, Blinds2Go was more focussed on a number of other factors, including changes to its
own cost price during the period and, as set out further below, the need to ensure that its prices
29
Paragraph 29 of the Issues Statement
30
Initially 60%, subsequently increased to 65%.
31
Phase I Decision, paragraph 46.
32
Paragraphs 86 and 227 to 230 of the Phase I Decision.
NON-CONFIDENTIAL
VERSION
17
were attractive to consumers trading up from ready-made blinds, as well as to consumers
trading down from M2M offers from in-home providers and high street stores, at the same time
scraping prices from a whole host of online competitors, not simply 247. As explained, at the
virtual site visit, this was extended to include Next from Autumn 2019 and only did not include
Dunelm due to the practical difficulty of scraping its website (and when Blinds2Go manually
checked Dunelm’s website its prices were the same as Blinds2Go’s).
3.31 In reaching the decision to refer the 2019 Transaction, the Phase I Decision therefore relies on
a theoretical premise that (a) “the 2019 Transaction removed any constraint exercised by the
founding shareholders on Hunter Douglas in relation to decisions relevant to the behaviour of
247’s business in the market and, therefore, eliminated any ability that 247 might have to
compete with Hunter Douglas to any extent”
33
and (b) “Hunter Douglas’s incentives to increase
prices/reduce quality will increase further as it will get all the profits resulting from 247’s
business”
34
.
3.32 The 2019 Transaction completed in February 2019 and the Interim Enforcement Order was
made on 21 November 2019. During that period, Hunter Douglas was free to pursue whatever
strategy it wished as regards 247. The CMA therefore had nine months of actual data before
it which might be used to demonstrate an effect on competition arising from the 2019
Transaction. It is notable that the Phase I Decision fails to advance any evidence that 247
followed a different course of action pre- and post- the 2019 Transaction. Indeed, the evidence
shows that 247 continued to develop and compete as aggressively following the 2019
Transaction. For example, 247 launched its first television advertisements in March 2019,
whilst the data shows no reductions in its other marketing activities, including PPC and organic
search spend. Further, in order to respond to the needs of its customers, 247 []. As a result,
247 continued to grow strongly in the period between completion of the 2019 Transaction and
21 November 2019, with monthly growth in M2M blind sales of [] (on a 12-month moving
average basis) in the 9 months to November 2019, as shown in Figure 3.2.
Figure 3.2: 247 M2M blind monthly sales at the time of the 2019 transaction
[]
3.33 Similarly, the Phase I Decision’s statement that Hunter Douglas has an incentive to increase
prices and reduce quality as a result of getting all the profits from 247’s business is not backed
up by any evidence at all. There is no evidence of Hunter Douglas having increased prices
and/or reduced quality following the 2019 Transaction. Indeed, it is entirely unclear that Hunter
33
Paragraph 230 of the Phase I Decision.
34
Paragraph 231 of the Phase I Decision.
NON-CONFIDENTIAL
VERSION
18
Douglas would have an incentive to do so and the Phase I Decision provides no analysis to
support its position.
3.34 In fact, Hunter Douglas’ business model is one which recognises that in dynamic consumer
markets, management teams need to be empowered to run their businesses as they see fit in
order to maintain incentives to grow and compete. This is precisely how Hunter Douglas has
left 247 to function following the 2013 Transaction and the 2019 Transaction. Hunter Douglas
has only limited knowledge of local markets and little in the way of central business capabilities,
and 247 is a very small business in the context of the wider Hunter Douglas group. The
management teams of its local businesses are therefore afforded significant leverage and
incentive to run their own business in the most profitable manner possible. It is entirely
implausible that Hunter Douglas could impose any particular commercial strategy on its
subsidiaries in those circumstances.
Issues Statement
3.35 The Issues Statement sets out the need to consider a separate counterfactual for the 2019
Transaction. The Parties agree that this is the correct approach.
3.36 In the view of the Parties, the correct counterfactual for the 2019 Transaction is that set out in
the panel’s summary of the Phase I Decision
35
to the effect that “the appropriate counterfactual
for the 2019 Transaction is that 247 would have continued to exercise the degree of competitive
constraint it had done absent the 2019 Transaction.” The question, therefore, is the degree of
competitive constraint which 247 exercised prior to the 2019 Transaction. As set out in more
detail in the following sections, prior to the 2019 Transaction 247 exercised only a limited
constraint on the activities of Blinds2Go. A change in the nature of Hunter Douglas’ control
over the activities of 247 has little effect on the conditions of competition within the market. In
the view of the Parties, the evidence shows that there is no risk of an SLC arising and the panel
should not seek to rely on a purely theoretical approach in relation to the 2019 Transaction, as
the CMA did in the Phase I Decision.
Conclusion
3.37 In summary, and irrespective of the issue of jurisdiction, the Parties are of the view that the
relevant counterfactuals are as follows:
• For the 2013 Transaction, the conditions of competition prevailing immediately before
the 2013 Transaction took place, with 247 as a small online retailer of M2M blinds in
35
Issues Statement, Paragraph 27
NON-CONFIDENTIAL
VERSION
19
the UK and Hunter Douglas having only a small interest in the retail of window
coverings in the UK and no online presence;
• For the 2019 Transaction, the conditions of competition prevailing immediately before
the 2019 Transaction, a situation in which, according to the Phase I Decision, Hunter
Douglas held material influence over the activities of 247
36
.
3.38 In neither case can there be said to be any risk of an SLC. In respect of the 2013 Transaction,
there was only limited overlap with 247 having a turnover of just £9m and a share of supply of
just [] on the narrowest market where there was an overlap with Hunter Douglas’s existing
business. In respect of the 2019 Transaction, the acquisition of full control has no impact on
the competitive structure of the market, let alone a substantial impact.
3.39 The Phase I Decision adopts a differing approach, however, in respect of the 2013 Transaction
and proceeds to make an assessment of the acquisition of material interest over the activities
of 247 against a counterfactual where 247 is a completely independent rival of Hunter Douglas
in 2019. Whilst the Parties disagree with this approach, in their view there is no evidence of an
SLC even if this route is followed, as set out in more detail in Chapters 6 and 7.
36
Phase I Decision, paragraph 46;
NON-CONFIDENTIAL
VERSION
20
4. Market context
4.1 As with many industries, the internet has profoundly changed the way in which the window
coverings sector operates and continues to do so.
4.2 Before the advent of the internet, consumers had two options for the purchase of M2M window
coverings; either in-store or in-home. There are a number of steps in the purchase and supply
process in both cases, all of which have been affected by online services:
• Design - a customer chooses the type of window covering, material and design, either
in-home or by visiting a specialist local store or larger retailer (such as John Lewis or
Dunelm);
• Measuring – once material and design is selected, a fitter will measure the window,
either at the time (in-home) or by subsequent appointment (in-store);
• Ordering – once measurements have been taken, a final quote is provided to the
customer. If the customer wishes to proceed, payment is taken and an order placed
with a supplier;
• Supply – once the product has been finished, it is supplied to the retailer; and
• Fitting – upon receipt of the finished product, the retailer arranges with the customer
for fitting.
4.3 The internet has affected all of these stages.
Consumers use of the internet for purchasing window coverings
4.4 At the design phase, consumers start their purchase journey by considering what options and
alternatives they find appealing, across blinds, curtains and shutters. Consumers’ choices are
likely to be based on a combination of factors, such as what they see in friends’ and family
members’ homes, online on social media and home furnishing magazines and websites, visiting
local home furnishing/window covering retailers and through online searches. Customers will
conduct extensive research to decide upon the product that they consider will best suit their
window, often visiting large home stores such as Next, John Lewis and Dunelm, high street
blind and curtain specialists and online blind and curtain sites. There is no single factor which
will drive consumer choice in this respect with each individual customer having a differing
budget and design idea.
NON-CONFIDENTIAL
VERSION
21
4.5 The internet is, however, almost always part of the customer experience and affects all
distribution channels. Customers may browse the internet for design ideas or for shopping
around or to make a specific purchase. As recognised in the Global Data report
37
:
• “Researching is widespread across all window dressing product types, as consumers
commit from a few hours to a few days to track down the best prices and find items of
the right quality, design and functionality to improve their homes
• The internet is at the forefront of much of this research, with 64.2% of shoppers stating
that it is the first place that they visit. Online is a particularly popular channel for looking
at readymade products, with consumers consulting reviews, comparing prices and
scrutinising pictures and videos of products.”.
4.6 This research phase occurs in respect of all customers and involves multiple channels for such
research with 54% of customers visiting stores before making a purchase, 56% of customers
browsing websites and 28.3% reading reviews online
38
. There will be a significant overlap
between these customers, with the design phase for most customers taking a day or more to
complete.
4.7 Hunter Douglas’ own data supports this. As shown Figure 4.1, Hillarys data shows that, in
2019, [] of its in-home customers began their search online, up from [] in 2007. There is
no reason to suspect that the same is not equally true for other channels, with the internet
providing an easy research tool to explore design ideas, products and prices.
Figure 4.1: Share of Hillarys In-Home Sales Leads generated Online
[]
Source: Hunter Douglas (Hillarys)
4.8 The low conversion rates experienced by online only retailers further support this. 247’s
conversion rate in 2019 was [] and that of Blinds2Go is around []
39
meaning over [] of
visitors to Blinds2Go’s and 247’s websites do not purchase when visiting their sites. Rather,
they are likely to be browsing styles and prices to inform their design choices and their final
purchase decisions. The Parties consider it likely that other online retailers have similar
conversion rates. Consequently, whilst advertising may be successful in driving customers to
37
Global Data Report: Window Dressings, November 2018, provided in Phase I at Annex 22.2, at page 40
38
Ibid. page 43
39
Number of orders divided by the number of unique users in 2019. See further Chapter 7.
NON-CONFIDENTIAL
VERSION
22
use the Parties’ sites as a source during the research phase, customers rarely purchase when
they arrive at the site.
4.9 The impact of online affects all distribution channels, with traditional, in-store retailers such as
John Lewis, Dunelm and Next becoming multi-channel retailers with an increasing online
presence. Smaller in-store retailers have turned to the internet either as a marketing tool or as
a primary source of revenue.
4.10 Dunelm initially entered the online M2M blinds space in March 2018, before temporarily
suspending the click-to-order functionality in November 2019 as a result of issues with a new
web platform. The Parties understand that Dunelm has now reinstated a click-to-order online
option for customers of its M2M blinds through a hybrid, virtual consultation route. The Dunelm
website sets out the four steps to order as follows:
Screenshot of Dunelm Made-to-Measure website
4.11 As is clear from the above, both the measuring and the fitting aspects of the service are
intended to be carried out by the customer, just as they are with 247, Blinds2Go and all other
online only retailers. The browsing journey to choose designs, blind styles etc is also clearly
conducted entirely online, with Dunelm offering a point of differentiation by offering live chat
appointments for further consultation prior to order.
NON-CONFIDENTIAL
VERSION
23
4.12 In-home providers, such as Hillarys, have also needed to adapt. Hillarys has begun to offer
more extensive online offerings, such as its web visualisation service. It is likely that [] The
advent of the measure-yourself retailers has also led Hillarys to [].
The effect of online on the value chain
4.13 The online model has had a profound effect on the industry. The efficiencies provided by online
players have reduced prices and improved delivery times significantly, with in-store and in-
home providers forced to reconsider their pricing and service offering. Neither is this effect
limited to retailers. The explosion of online retailers has led to new relationships being forged
with suppliers across the world and wholesalers/assemblers of M2M products competing more
fiercely for retailer customers.
4.14 Sales of M2M blinds online have grown dramatically in the last 10 years and continue to do so.
By way of illustration, Hunter Douglas estimates that sales online of M2M blinds have grown by
over 25% in the last two years from £199.9m in 2017 to £256.3m in 2019
40
. The Parties are
not aware of a single exit from the online M2M segment of the window coverings market in the
last ten years and there continues to be significant scope for market expansion.
4.15 Indeed, the growth of online sales may accelerate following the Covid-19 outbreak. In the first
four months of 2020, Blinds2Go and 247 saw even greater growth, with year on year sales
increases of M2M blinds in April 2020 of []. As the market grows, there will continue to be
opportunities both for existing retailers and new entrants.
4.16 The growth of the retail of M2M blinds online has come primarily at the expense of sales of
ready-made blinds as well as M2M blinds through other distribution channels. As set out in the
AMA Report 2018
41
, sales of ready-made blinds have decreased from 28% of UK blinds sales
in 2012 to 23% in 2017. Hunter Douglas estimates that since 2017, ready-made sales have
fallen still further
42
, to around 18%
43
.
4.17 Online sales of M2M blinds have also grown at the expense of offline channels, whether in-
store or in-home
44
. Hunter Douglas estimates that, during the period 2017 to 2019, online sales
40
Annex 0002, Tab 3.1
41
Annex A3
42
See Annex 0002, Tab 3.6
43
Based on figures in Annex 0002, Tabs 3.1 and 3.6
44
In reality, in-store and in-home sales do not differ significantly. As measuring and fitting is conducted by
the retailer, in-store sales also require a home visit.
NON-CONFIDENTIAL
VERSION
24
of M2M blinds grew by approximately £50m
45
. In the same period, sales in-home/in-store fell
by £17m
46
.
4.18 The disruption caused by the recent rapid growth in sales of M2M blinds online, which has
come largely at the expense of retailers of ready-made products and in-store/in-home
providers, has required these retailers to make a major change to their strategy. Such an
approach is necessary both because of the negative impact on their existing businesses and
because of the opportunities provided by the online MTM blinds sector. The process has
already begun and is likely to accelerate in the near future, particularly as a response to the
COVID-19 crisis. Indeed, Sharon White, CEO of John Lewis has recognised this, stating that
“Online is going to be a bigger deal. When we look at Asia and Europe, even without social
distancing, people are not flocking back to stores. They are worried about being with other
people.”, (The Guardian, 29 April 2020).
47
As with many sectors, the traditional retail model is
increasingly moving online and window coverings is no different.
4.19 As set out in Chapter 6, retailers of ready-made blinds are increasingly turning to offer M2M
products. Online retailers, particularly through platforms such as eBay and Amazon are offering
“made to measure” or “bespoke” blinds through a cut down service where the customer orders
the next largest ready made size and specifies the exact size to which the seller should cut
down the blind.
4.20 Traditional retail models are also moving online with in-store providers expanding their online
offering.
4.21 Next has a strong retail brand in home furnishings and one of the most visited websites in the
UK. It has been offering ready-made blinds online for many years and now offers a range of
895 SKUs online. In addition, in Autumn 2019, Next began offering a range of M2M blinds
online, with a current range of 821 SKUs.
4.22 Dunelm is another multi-channel retailer that has recognised the challenge from online-only
M2M retailers. In 2016, Dunelm was one of those expressing an interest in purchasing
Blinds2Go. Having lost out to Hunter Douglas, Dunelm has now begun to expand its online
M2M offering in earnest. Hunter Douglas estimates that, in 2018, Dunelm accounted for 21.2%
of retail sales of ready-made window coverings in the United Kingdom
48
. In order to capitalise
on the growth of online M2M and maintain its market position, Dunelm has recently relaunched
45
Annex 0002, Tab 3.1
46
Annex 0002, Tab 3.1
47
(https://www.theguardian.com/business/2020/apr/29/some-john-lewis-stores-may-not-reopen-after-
lockdown-admits-boss).
48
Annex 0002, Tab 3.8
NON-CONFIDENTIAL
VERSION
25
its online M2M sales, offering online consultations where a customer reserves an appointment
for 30 minutes (single window) or 1 hour (two or more windows).
4.23 The online model has also presented challenges to manufacturers. In order to stay competitive,
online retailers must not only be able to compete on price but must also be able to ensure that
a sufficient range of products is available, and that delivery can be achieved within a short
space of time. Online retailers are therefore more demanding in their relationships with
suppliers and are willing to shop around for the best prices, products and service, sourcing
products from multiple suppliers both within and outside the UK.
4.24 Manufacturers are therefore under greater pressure to perform as customer expectations rise
and they are put under pressure from online retailers. This has increased the interest of
manufacturers in creating vertically integrated businesses which allow them greater control of
the supply chain. Interior Goods Direct and Swift Direct Blinds are examples of vertically
integrated online retailers of M2M blinds and the Parties expect the number of vertically
integrated retailers to increase. As the CMA will be aware, []. Hunter Douglas has no doubt
[].
4.25 Indeed, entry and acquisition are unlikely to be limited to UK manufacturers. Nien Made, one
of the largest blinds’ wholesalers in the world, purchased a 51% stake in Veneta, one of the
leading online window coverings retailers in the Netherlands
49
and Australia
50
. There is no
reason to suspect that other large international manufacturers will not follow suit.
4.26 It is not just traditional retailers and suppliers that have come under pressure. As business
models develop, online retailers themselves are facing significant pressure from the emergence
of eBay and Amazon as serious competitors.
4.27 The Parties estimate that eBay accounts for around 13.6% of online sales of M2M blinds with
nearly 100 sellers of M2M blinds identified on eBay
51
. Retailers on eBay face none of the costs
associated with a full service online retailer (in particular, website and design) and are thus able
to offer product at the lowest possible price. This is even more the case as retailers from across
the world are able to offer products to UK customers with the same delivery times as the UK
online retailers. As UK online retailers continue to diversify their wholesale supply options to
keep costs down, and delivery cost and speed improves, sourcing product from suppliers in the
Far East becomes increasingly attractive and there is nothing which prevents retailers around
the world from doing the same to service customers in the UK. Similarly, wholesalers in the
49
www.veneta.com
50
www.venetablinds.com.au
51
Annex 0002, Tabs 3.2 and 3.4
NON-CONFIDENTIAL
VERSION
26
Far East wanting direct access to UK customers can act as a retailer directly on eBay and
Amazon without any capital investment.
4.28 Amazon currently accounts for around 3.2% of online M2M blinds sales
52
. As with eBay,
Amazon offers retailers from around the world low cost access to UK consumers for the sale of
M2M blinds online. Amazon has recently introduced a customization function to allow
customers to input the precise measurements of their blinds without having to include this as
an additional message (as happens on eBay). The service offered by Amazon is therefore
fundamentally little different to the establishment of a website save in respect of the fact that it
requires no capital investment.
4.29 The Amazon and eBay models present serious challenges to retailers operating their own
websites. As customers become more accustomed to Amazon and eBay for M2M blinds, it is
reasonable to expect that consumers will gravitate towards them given the strength of those
brands. The next big challenge for the existing online M2M retailers is to work out how to
respond, given that they cannot hope to compete with either eBay or Amazon in terms of brand
value and marketing spend.
4.30 The Phase I Decision considered that Amazon and eBay commission levels were around 20%
and PPC advertising amounted to a margin reduction of 12-13%. It therefore considered that
a difference of 7 to 8% in the retail margin was significant enough such that retailers on Amazon
and eBay could not be considered to be serious competitors
53
. Further, it noted that neither
Hunter Douglas nor 247 sell through marketplace platforms, but obviously that does not reveal
whether retailers can sell effectively via these platforms. However, Amazon’s customization
function has changed this situation significantly. As set out in 247’s response to the Request
for Information
54
, []. []. [] therefore shows that Amazon commissions are at a level which
is comparable or lower than PPC charges whilst avoiding incurring website costs, contrary to
the position asserted in the Phase I Decision.
4.31 The rise of online M2M blind sales has led to increased consumer choice, with lower prices and
higher quality products the result of intense competition. As explained to the CMA at the site
visit on 27 April 2020, the cost of a M2M blind today is around the same price that Blinds2Go
were offering their original ready-made blinds when they launched their site in 2000. This has
arisen despite the 2013 and 2019 Transactions and considerable inflation in cost prices
52
Annex 0002, Tab 3.2
53
Phase I Decision at paragraphs 201 and 202.
54
247 Response to Request for Information at paragraph 8.4
NON-CONFIDENTIAL
VERSION
27
resulting from the significant fall in sterling since 2016
55
. There is every reason to believe that
this competition will be maintained if not increased in the coming years.
55
Most fabrics and blind components are imported, particularly from Asia.
NON-CONFIDENTIAL
VERSION
28
5. The Parties
5.1 Both Blinds2Go and 247 have seen growth in their online sales of M2M blinds, albeit at a
different rate. Blinds2Go has far outstripped its rival in recent years and is now over 5 times
the size in terms of revenue. That they have achieved this without any form of capital
investment underlines the ease of entry and the significant risks that they now face from the
emergence of new business models and new entry.
Blinds2Go
5.2 In 2011/12, Blinds2Go was the number 3 online retailer of M2M blinds in the UK, with sales of
£[]million. At the time, 247 was the largest retailer, with sales of £[] million. Since then,
Blinds2Go has grown its sales of M2M blinds online in the UK [] to over [] in 2018 (the last
financial year before the 2019 Transaction) whilst 247 has only grown its sales by [] to around
[] over the same period. Blinds2Go is now, by some distance, the largest of the pure online
retailers of M2M blinds in the UK, a position which it has held since before Hunter Douglas’
investment in 2016.
5.3 There are a number of factors which Blinds2Go considers are responsible for its rapid growth
from 2011/12. Until 2008/9, there had been tensions between the original partners which had
prevented expansion. Those tensions were resolved with the exit of []. This allowed the
restructuring of the business to include [] and [] as shareholders and to develop a new
business plan. The particular skills of the shareholders of Blinds2Go at that time provided (and
continues to provide) Blinds2Go with a range of expertise in product development, supply chain,
digital marketing and website development which is largely unmatched in other online
providers. Those shareholders are incentivised to continue to grow the business aggressively
through their continued share ownership. Unlike many other online retailers, Blinds2Go’s plan
was, and is, therefore focussed on delivering quality products at the lowest possible price and
began to bear fruit from 2010/11.
5.4 Blinds2Go’s focus on quality has marked it out against its rivals. This understanding of the
customer has enabled it to meet demand to a far greater extent than its rivals. In 2009/10, the
focus for online retailers was price. All of the online retailers therefore offered the most basic
products at the lowest prices in order to attract customers.
5.5 Blinds2Go, however, recognised that []. Blinds2Go therefore ensured that its products were
of the highest quality and its approach to supply relationships meant that it was able to ensure
that its prices remained competitive. This is highlighted in the relationship with []. Blinds2Go
[].
NON-CONFIDENTIAL
VERSION
29
5.6 Blinds2Go also recognises the importance of its website in attracting customers and converting
visitors to sales. Blinds2Go was the first of the online retailers to [] to ensure an enhanced
customer experience and it continues to review the design of its site to ensure it is current.
5.7 This understanding of customer requirements has allowed Blinds2Go to persuade consumers
both to shop online rather than other purchase options (ready-made, in-home, in-store, other
forms of window covering) and that Blinds2Go should be the retailer of choice when shopping
online. Blinds2Go specifically set its prices at a level which []. Whilst Blinds2Go does [],
it does not []. Broadly speaking, bearing in mind Blinds2Go’s higher focus on quality, it has
sought to ensure that []. In fact, as the CMA will be aware, [] as set out in more detail in
Chapter 6.
5.8 Blinds2Go is only able to achieve this balance [] through careful management of its supply
relationships. In 2008/9, online retailers in the UK focussed on existing products made
available by UK suppliers. They did not develop their own products in conjunction with the
manufacturers and paid little attention to product quality. The focus for those providers was
price as this was the only element on which they considered they competed.
5.9 As noted above, in [], Blinds2Go [] in 2009. Blinds2Go []. As it looked to [], Blinds2Go
approached []. []. It is important to note that scale was no barrier to [].
5.10 [] but Blinds2Go has developed a diversified supply network of [] wholesalers/assemblers.
This diverse network allows Blinds2Go to review its pricing and quality regularly, switching with
ease between suppliers where one or other factor drops below that which Blinds2Go expects.
Nothing is holding its competitors back as there are many more workrooms and actual or
potential suppliers of blinds whether in the Far East or closer to home. Indeed, trade shows
such as R+T Stuttgart
56
and R+T Asia
57
bring together retailers (including online retailers) from
across the world (including the UK) with hundreds of low labour cost assemblers
58
.
5.11 Blinds2Go’s identification of a broad range of competitors has also enabled it to develop
different strategies to address different customer groups. At Annex 0049, Blinds2Go submitted
[]. Since May 2016, []. As previously indicated to the CMA, Blinds2Go also [].
5.12 In Blinds2Go’s view, this approach has allowed it to carefully position itself as []. The M2M
blinds offered by Blinds2Go are invariably []. Equally, Blinds2Go offers a range of products
56
https://www.messe-stuttgart.de/r-t/en/
57
https://en.rtasia.net/
58
The list of exhibitors at R+T Stuttgart 2018 (see https://www.messe-stuttgart.de/r-t/en/visitors/index-of-
exhibitors/) includes 176 exhibitors in the Internal Sun category and 89 in the Blinds Textiles category.
The list of exhibitors at R+T Asia 2019 (see https://en.rtasia.net/exhibitors-at-rt-asia.shtml) includes 489
exhibitors in the Sun Protection category.
NON-CONFIDENTIAL
VERSION
30
that compete []. When coupled with Blinds2Go’s particular approach to design, with [],
Blinds2Go has managed to position itself as a more premium online retailer.
5.13 Blinds2Go’s marketing approach since 2011 has been to recognise its evolution and adapt
accordingly. Blinds2Go []. In the beginning, Blinds2Go’s marketing strategy was focussed
on search engine optimization since organic search was the primary means of accessing
customers. That strategy has changed over time to focus more on [] although there are a
number of reasons to believe that this too will change over time.
5.14 Firstly, the rise of Facebook and Instagram as ways in which to generate customers is
increasingly changing Blinds2Go’s approach to marketing. The value in such channels as a
free means of advertising is highlighted by the example of Makemyblinds. Hunter Douglas
estimates Makemyblinds 2019 sales at £2.3 million, a rapid increase from its launch in 2015.
Blinds2Go attributes this increase in part to an effective online marketing strategy that has
focussed not just on PPC advertising but on social media, in particular Instagram.
Makemyblinds has nearly 35,000 followers on Instagram, compared with under 5,000 for 247
and under 10,000 for Blinds Direct. Blinds2Go considers that social media is likely to become
an increasingly important marketing tool. With around 60,000 followers on each of Facebook
and Instagram, Bilnds2Go continues to recognise this phenomenon and is working hard to
make sure that it is not taken by surprise.
5.15 The other reason that PPC advertising may become less important in the near future is because
of Google’s decision to change its approach to Google Shopping, with listings becoming free
by the end of the year
59
. In Blinds2Go’s opinion, this is likely to lead to search engine
optimization becoming more important in comparison with PPC marketing as Google Shopping
becomes a largely free platform.
5.16 Overall, Blinds2Go believes that it has been able to grow its business as a result of a greater
understanding of how to attract customers to purchase online. Blinds2Go’s []
60
shows that
its customers are []. Rather, they are customers that []. In-home and in-store products
which might have appealed to such customers are considered over-priced when faced with a
comparable product at significantly lower price which can be measured and fitted with little
effort. Similarly, Blinds2Go positions its M2M products [], ensuring that it continues to be
well-placed both to grow the M2M blinds market and to be well-placed to benefit from that
increase in consumer demand.
59
See paragraph 8.6, Hunter Douglas response to Request for Information of 20 April 2020
60
Annex 2.2 of the response to the Fourth Section 109 Notice in Phase 1.
NON-CONFIDENTIAL
VERSION
31
247
5.17 In 2011/12, 247 was the largest online retailer of M2M blinds in the UK with total sales of
£9.2 million. Since then, it has seen both Blinds2Go and Interior Goods Direct exceed 247’s
sales. With online M2M sales of c£[] in 2019
61
, 247 is now the third largest online only retailer
of M2M blinds in the UK
62
.
5.18 When compared to Blinds2Go, it is clear that 247 is focussed on providing consumers with a
[] than Blinds2Go. In particular, 247 outsources many of its operations in terms of website,
online advertising and design. 247’s website has been developed using the Magento open
source platform which offers a full service, low cost e-commerce solution incorporating site
development and a range of back office functions. 247 uses an external agency (Journey
Further) as digital marketing agents and has no product designers employed in-house.
5.19 Between 2010 and 2020, 247’s sales grew by []%
63
. Most of the growth can, however, be
attributed to a period of significant expansion between calendar year 2015 and 2017 (i.e. after
the 2013 Transaction), during which sales grew by []. Between 2010 and 2015, growth was
more modest, with 247 losing share to its rivals, Blinds2Go and Blinds Direct.
5.20 Prior to the 2013 Transaction, 247 fell from a position of market leader to second in 2017 to
third as regards online only M2M blind retailing. This position is reflected in the fact that 247
now monitors [].
5.21 In any event, the evidence shows that 247’s prices []. 247 is keen to ensure that its prices
are amongst the lowest for the products it offers. 247 therefore prices as low as it can whilst
maintaining an appropriate gross margin, as set out in more detail in Chapter 6.
61
Latest full calendar year figures. Please note that this is different from the £[] figure provided for the
financial year ending in February 2019.
62
See Annex 0002, Tab 3.2
63
Paragraph 6.2 response to 247 Request for Information of 20 April 2020.
NON-CONFIDENTIAL
VERSION
32
6. Competition analysis
Introduction
6.1 The CMA’s Issues Statement indicates that it is “assessing a horizontal unilateral effects theory
of harm with respect to each Transaction in relation to the online retail supply of M2M blinds in
the UK”.
64
Accordingly, the CMA is exploring the extent to which important rivalry between 247
and Blinds2Go has been lost, and whether, as a consequence of this loss of rivalry, it would be
profitable for the merged business to appreciably worsen its offer as regards price, quality,
range or innovation.
6.2 The focus of this Chapter is on whether this is likely given the prevailing competitive conditions,
with Chapter 7 addressing why the Parties consider that barriers to entry and expansion are
low, such that new and growing rivals can be expected to replicate any loss of competition
between the Parties in a timely fashion (i.e. within two years).
6.3 This Chapter is structured as follows:
(i) Section (b) identifies the three key elements that must all apply for this theory of harm
to generate an expectation that the Transactions lead to an SLC. This provides
important context to the evidence set out in subsequent sections;
(ii) Section (c) describes how firms compete to meet consumer demand, developing
themes set out in Chapter 4 on the market context;
(iii) Section (d) describes why small competitors are effective rivals given the absence of
any economies of scale and their ability to ensure that their offerings are visible to
consumers. The Parties fundamentally disagree with the implicit premise in the Phase
I Decision that smaller online M2M competitors, including pure online players and major
omni-channel rivals, are not a competitive alternative;
(iv) Section (e) considers the competitive constraints from ready-made blinds and in-store
M2M blinds, and why the Parties consider that these alternatives should at least be
treated as important “out of market” constraints;
(v) Section (f) considers the evidence as regards price competition between the Parties
and other competitors, with this indicating that Blinds2Go faces active price competition
from many competitors and Blinds2Go’s prices do not closely track 247’s. This
64
Paragraph 29 of the Issues Statement.
NON-CONFIDENTIAL
VERSION
33
evidence contradicts any view that the Parties are particularly close competitors or that
the only price competition is limited to Blinds2Go, Interior Goods Direct and 247;
(vi) Section (g) describes the dynamically competitive nature of the market, with 247 being
displaced from being market leader to the third largest online retailer of M2M blinds
behind Interior Goods Direct.
The key elements of the CMA’s theory of harm
6.4 The CMA is exploring whether there was important rivalry between 247 and Blinds2Go which
has been lost as a result of the Transactions, and whether, as consequence of this loss of
rivalry, it would be profitable for the merged business to appreciably worsen its offer as regards
price, quantity (which is presumably simply the mirror of price as neither Blinds2Go nor 247
compete on manufacturing capacity), quality, range or innovation. There are three key
elements to this theory of harm, which all need to be satisfied for there to be an SLC.
6.5 The first element of the CMA’s theory of harm relies on the 2013 and 2019 Transactions having
actually substantially reduced rivalry between 247 and Blind2Go. This is addressed in Chapter
3 and this discussion is not repeated here. However, it is appropriate to emphasise that, unlike
most mergers, a material period of time has elapsed post-completion for each Transaction prior
to the CMA opening an investigation. As set out in Chapter 3, the Parties do not consider that
there is any evidence over this time of any loss of rivalry between them leading to price, quality,
range or innovation being worsened.
6.6 The second element of this theory of harm requires that, as a result of the Transactions, it would
be profitable for Hunter Douglas if Blinds2Go and 247 increased their prices (or otherwise
worsened their offer). This assessment has multiple dimensions, which collectively determine
whether there would be a SLC, absent countervailing factors. There are a number of key
points that must be taken into account in this second element of the theory of harm and these
are addressed in the following paragraphs.
6.7 First, whilst the CMA’s theory of harm refers to the merged business worsening its offer,
Blinds2Go is the market leader and many times the size of 247. When Hunter Douglas acquired
Blinds2Go in June 2016, Blinds2Go was already the market leader in the sale of M2M blinds
online and this remained the case in 2019: in the CMA’s decision in Hunter Douglas/Bellotto
(2017)
65
, the CMA estimated that in 2016 Blinds2Go and Web Blinds had a combined share of
50-60% with there being a market share increment of 5-10% as a result of the addition of Web
Blinds). Similarly, in the Phase I Decision, the CMA estimated that Blinds2Go’s market share
65
Paragraphs 145 and 146, Hunter Douglas/Bellotto (2017).
NON-CONFIDENTIAL
VERSION
34
in 2019 was 50-60% (i.e. similar to the combined share of Blinds2Go and Web Blinds in 2016).
66
The Parties consider that the CMA’s estimates overstate their market shares, however these
estimates are referred to in this Chapter to highlight the large differences in the Parties’ market
shares and trends in market shares over time. As a consequence, the core theory of harm being
considered by the CMA requires 247 to be a material competitive constraint on Blinds2Go’s
prices and offering, such that Blinds2Go would otherwise find it profitable to increase its prices
and/or otherwise worsen its offer in the absence of that constraint.
6.8 Second, any SLC assessment must also take into account that 247 is a relatively weak
competitor and there are multiple other rivals. 247 is a much smaller rival than Blinds2Go, with
its market share being estimated by the CMA in the Phase I Decision to be 5-10%
67
. Moreover,
247 has lost share of the M2M online blinds sector consistently over time - it had been displaced
from being the market leader prior to Hunter Douglas’ acquisition of a 60% stake in Blinds2Go
in June 2016. The CMA market share estimates also indicate that 247’s share has fallen from
10-20% in 2016 to 5-10% in 2019, whereas Interior Goods Direct/Blinds Direct’s market share
increased from 5-10% to 10-20%. Accordingly, 247 is now the third largest competitor, or the
fourth if eBay were to be treated as a single rival. In addition, there are many other pure online
and omni-channel M2M rivals who are growing their online sales - and who will have a
competitive imperative to do so due to the strong growth of online sales, which have been turbo
charged by Covid-19 and social distancing rules that are likely to apply for the remainder of
2020 even when shops re-open. Finally there are competitively significant “out of market”
constraints from ready-made blinds and in-store M2M blinds;
6.9 Third, the CMA should have regard to the fact that Hunter Douglas has a strong interest in
ensuring that 247 maximises its own profits and maximises its own cash generation. Although
normal commercial considerations make this the case, the nature of its shareholding in
Blinds2Go and the accompanying management incentives make this more likely. Hunter
Douglas has a 65% stake in Blinds2Go, such that it does not fully benefit from increases in
Blinds2Go’s sales and profits.
6.10 []
6.11 []
6.12 []
• []
66
Table 1 of the Phase I Decision.
67
Ibid.
NON-CONFIDENTIAL
VERSION
35
• []
• []
6.13 []
6.14 In any event, if 247’s competitiveness were to deteriorate, its customers are likely to be lost to
an array of rivals rather than only to Blinds2Go. Indeed, since 2016, 247 has been losing
market share to Interior Goods Direct and other rivals and not just to Blinds2Go, whose market
share has been relatively stable (as noted above). It is clear that 247 is facing substantial
competitive pressure from multiple sources.
6.15 Finally, []. As a consequence, at present there is nothing that differentiates 247 from a host
of other online M2M blinds retailers, and it lacks the natural brand advantages of the major
omni-channel retailers. In contrast to 247, []. These observations follow from the descriptions
of 247 and Blinds2Go set out in Chapter 4, with further detail being set out in the following
paragraphs.
6.16 To be clear, 247 is a profitable business, but simply getting consumers to visit its website does
not maximise its sales as [] of users visiting 247’s website do not purchase anything.
Accordingly, []. To put this in context, if 247 and Blinds2Go each spend £1,000 attracting
consumers to their respective websites, then there is, on average, a [] chance of those
consumers purchasing from 247 but a [] per cent higher chance of those consumers
purchasing from Blinds2Go (Blinds2Go’s user conversion rate is [], compared to [] for 247).
[]
6.17 Hunter Douglas’ view is that the key principles of a holistic user experience can be summarised
succinctly, but before doing so it is important to emphasise that none of this requires material
capital investment. The core elements of a holistic consumer strategy are as follows:
(i) []
(ii) []
(iii) []
(iv) []
(v) []
(vi) []
6.18 The third and final required element of the CMA’s theory of harm is that there would not be
even the modest entry and expansion that is required to offset the limited loss of rivalry from
NON-CONFIDENTIAL
VERSION
36
247 as a small M2M blinds competitor and one which has been losing market share. Chapter
7 considers further the extent of any economies of scale and whether entry and expansion
would be timely likely and sufficient to offset the loss of rivalry from 247, with Hunter Douglas
considering that any competition from 247 can be readily and rapidly replicated.
How firms compete in the market
6.19 A key driver of competition in the market for online M2M blinds is how customers go about
choosing their preferred window covering supplier. This section considers the customer
journey, building on the key themes set out in Chapter 4, including what consumers value and
how they consider alternative products and suppliers.
6.20 As noted in Chapter 4, consumers start their purchase journey by considering a range of
window covering alternatives they find appealing across blinds, curtains and shutters. How
customers choose the product and supplier is likely to be influenced by a range of factors
including price, design, product quality/range, and their research based on what they see in
friends’ and families’ homes (and their recommendations), online on social media and on home
furnishing magazines and websites, visiting local home furnishing/window covering retailers,
and online searches.
6.21 In order to explore this issue further, the Parties commissioned BVA-BDRC to survey 37,773
customers who placed an order in early 2020 (the BDRC Survey).
68
Full details of this survey
including the questionnaire
69
, methodology and results, customer emails, and raw data are
provided at Annexes 0092, 0093, 0094 and 0095. This reveals that many consumers search
online, including comparing prices, reading reviews and discussing with family and friends.
Patterns of online search are systematically greater for 247 customers, suggesting that these
customers shop around more.
Figure 6.1: Consumer search behaviour prior to purchase
[]
Source: BDRC Survey
6.22 The BDRC Survey also asked customers how much time they spent researching before
deciding to purchase. These results are shown in Figure 6.2 below. This shows that [] of
the Parties’ customers are spending more than an hour researching / comparing products
before purchasing, which should be judged in the context that online comparisons are quick
and easy. Indeed, about half of consumers are spending a considerable amount of time
68
This included 30,841 Blinds2Go customers and 6,932 247 Blinds customers.
69
A draft version of the survey questionnaire was shared with the CMA on 17 April 2020.
NON-CONFIDENTIAL
VERSION
37
considering their options (in excess of a few hours), showing that consumers are likely to
consider a range of alternatives prior to purchase. This evidence is consistent with the Google
Analytics data relating to customers’ browsing behaviour, which shows that the average
customer makes its purchase ten days after the first website interaction and after having visited
the website on three different occasions.
70
Even for those customers only spending a few hours
or less, given the speed of online search, many products/retailers can be compared within a
short space of time.
Figure 6.2: Time spent researching or comparing products before purchase
[]
Source: BDRC Survey
6.23 In terms of what drives consumers’ choice of retailer, [] were the [] for choosing one retailer
over another. Very few were ultimately influenced purely by online search advertising or social
media. Consistent with the other evidence presented in this submission, 247 customers were
[]. These results are presented in Figure 6.3 below.
Figure 6.3: Reasons for choosing a particular retailer
[]
Source: BDRC Survey
6.24 The BDRC Survey data is supported by recent third-party evidence. As highlighted in
the AMA report, the “[window coverings] market remain[s] highly competitive, with greater
transparency of prices on the internet enabling consumers to easily compare product offers and
with widespread discounting taking place between suppliers and retailers”. Similarly, findings
from GlobalData suggest that quality and price are the most important factors driving
consumers’ choice of window covering (see Figure 6.4 below). In contrast, brand is not
considered important by consumers, mentioned by just [] of consumers, suggesting that
products are highly commoditised and that consumers are highly likely to switch in response to
an increase in prices or any deterioration in quality.
70
See Annexes 0106 – 0108. Note that these estimates likely understate the actual purchase delay for the
reasons set out in the email from Dilip Roy to the CMA on 20 May 2020.
NON-CONFIDENTIAL
VERSION
38
Figure 6.4: Factors of importance in choice of window dressing
Source: GlobalData, ‘Home Retail Series, Window Dressings November 2018’, page 67.
6.25 Figure 6.5 below shows the most common reasons for consumers’ choice of retailer for
window dressings. While the figures may not be fully representative of the online channel –
e.g. ease of access would not be a consideration – it highlights the importance of competitive
pricing, quality and product range, which is again consistent with the BDRC Survey.
Figure 6.5: Drivers of window dressings retailer choice
Source: GlobalData, ‘Home Retail Series, Window Dressings November 2018’, page 64.
NON-CONFIDENTIAL
VERSION
39
6.26 Consumers also use a variety of ways to find the right retailer online. Search engines clearly
provide a valuable resource for consumers to find and compare alternative products and
retailers to find what they need. Evidence from Google Trends suggests that consumers were
approximately 10 times’ more likely to search for “blinds” than to search for one of the leading
M2M suppliers by name.
71
6.27 The Parties agree that advertising in search engines is important for their businesses, since,
as small on-line mono-line retailers, they do not have strong brand names, with this particularly
being the case for 247. The Parties also agree that they appear frequently and rank prominently
in paid-for search, organic search and Google shopping results. Where the Parties strongly
disagree is the CMA’s statement that: “The CMA considered that the frequency and prominence
of the Parties suggested that their competitive significance is greater than their shares of supply
migh[t] suggest.”
72
6.28 First, consumers’ purchasing journeys are not limited to focusing on paid Google ads for generic
search terms such as blinds. This is clear from the Google Trends results which reveal what
else users searching for “blinds” are also searching for within the same search session.
73
This
is shown in Table 6.1 below.
71
Phase 1 response to Issues Letter – Figure 3.3.
72
Paragraph 161 of the Phase I Decision.
73
Google refers to these as related queries, defined as “terms that are most frequently searched with the
term you entered in the same search session, within the chosen category, country, or region”
(https://support.google.com/trends/answer/4355000) .
NON-CONFIDENTIAL
VERSION
40
Table 6.1: Google Trends related searches for “blinds” (May 2019 – April 2020).
#
Search term
Relative
score
#
Search term
Relative
score
1
roller blinds
100
11
wooden blinds
42
2
blinds to go
99
12
blinds direct
40
3
window blinds
63
13
Curtains
35
4
blinds uk
62
14
blackout blinds
35
5
venetian blinds
55
15
argos blinds
35
6
vertical blinds
54
16
Hillarys
26
7
roman blinds
47
17
white blinds
26
8
ikea blinds
43
18
hillarys blinds
26
9
Dunelm
42
19
blinds to measure
22
10
dunelm blinds
42
20
grey blinds
22
Source: Google Trends (Annex 0096)
Notes: 1) Scoring is on a relative scale where a value of 100 is the most commonly searched query, 50
is a query searched half as often as the most popular query, and so on.
6.29 This shows that while “blinds to go” is the most popular branded search term related to “blinds”,
users are also searching for Dunelm (combining the two separate terms would put the score
on 84), Ikea, Argos, and Hillarys - with 247 outside the top 20. Specific product search terms
(such as roller blinds and venetian blinds) also suggest that after searching for “blinds”,
consumers consider a range of alternative blind types (exposing themselves further to a range
of different suppliers).
6.30 This is consistent with the BDRC Survey results, which show that []% of Blinds2Go’s
customers, and []% of 247 customers are searching for “blinds” or other similar search terms.
Equally brand is relatively important for Blinds2Go with []% of its customers being aware of
its brand and either searching for Blinds2Go or going straight to the website. 247 customers
were less aware of the 247 brand prior to purchase (only []%) and were consequently less
likely to search for the brand or go directly to the website.
Figure 6.6: Brand awareness and customer search journey
[]
Source: BDRC Survey
6.31 Second, it is also important to appreciate that omni-channel retailers have a natural competitive
advantage in that consumers will frequently visit their websites to research and purchase a
wide range of products. These retailers also have the option of email marketing campaigns to
their millions of registered customers. In this regard, the Google Trends data suggest that
NON-CONFIDENTIAL
VERSION
41
consumers are performing a variety of searches in one search session (which Google Analytics
defaults to defining as being for no more than 30 minutes, or any session after midnight)
74
.
6.32 Third, whilst paid search is an important source of traffic to the Parties’ websites, this should
not be exaggerated or over interpreted. The breadth of ways customers find even pure on-line
mono-brand retailers is also highlighted by the source of traffic to the Parties’ websites. This
includes generic paid search (e.g. where consumers search for words such as blinds, rather
than the company’s own brand name, either in the paid ads or via Google Shopping), branded
paid search (e.g. consumers searching for “Blinds to go” or “247 blinds”), organic search (where
consumers visit the organic results instead) and direct (where consumers go directly to the
website). Figure 6.7 below shows the number of sessions split by different channels from 2016
- 2019. A range of points should be noted as regards this data, including the differences
between Blinds2Go and 247 traffic sources.
Figure 6.7: Sources of monthly sessions by channel – Blinds2Go and 247 (2016-2019)
[]
Source: Google analytics; AlixPartners analysis. (Annex 0097)
6.33 This highlights that although paid search makes up just over half of user sessions for the Parties
([] for Blinds2Go v [] for 247), consumers use a mixture of organic search results, email,
display and referral sources to find the Parties’ e-commerce websites and the importance of
these other sources is increasing modestly over time.
6.34 In addition, the Parties’ use of PPC advertising should not be over interpreted and it should be
borne in mind that this reflects their specific market strategies. Organic search results remain
a cost-effective way to grow a business, and other e-commerce sites may also rely more on
social media and other referral or display referral sources. Further, Google has announced that
during the course of 2020 “search results on the Google Shopping tab will consist primarily of
free listings, helping merchants better connect with consumers, regardless of whether they
advertise on Google”
75
. Accordingly, Google Shopping is now moving to an organic model,
which will further provide avenues for smaller online M2M blind suppliers to reach customers
at low cost.
6.35 Fourth, the CMA’s statement that prominence in search suggests that the Parties’ competitive
significance – and presumably specifically 247, which the CMA has found only has a 5-10%
share - “is greater than their share of supply migh[t] suggest”. This is based on three mistaken
74
https://support.google.com/analytics/answer/2731565?hl=en
75
See https://www.blog.google/products/shopping/its-now-free-to-sell-on-google/
NON-CONFIDENTIAL
VERSION
42
premises – that consumers’ researching is limited to online generic word traffic search (which
is contradicted by the Google Trends analysis set out above), that 247’s Google prominence
somehow automatically renders 247 better at attracting and winning customers (which it self-
evidently does not as its share would not be 5-10% and declining) and that visits to 247’s
website translates directly into sales.
6.36 As regards both 247 and Blinds2Go, they have session conversion rates (i.e. consumers
visiting a website and then making a purchase) ranging from [] over the last six years,
indicating that []% of sessions do not translate into sales. These rates reflect that consumers
shop around before buying, and ultimately are not making purchase decisions based on Google
prominence. User conversion, which aggregates sessions per user, is higher as consumers
may purchase after multiple visits (reflecting their further research) but were still only [] for
Blinds2Go in 2019 and [] for 247. This suggests that consumers are willing to make multiple
visits to websites and undertake multiple sessions (likely considering a range of different
products and retailers in the process) before deciding on their choice of product.
6.37 It is also insightful to consider the differences between the Blinds2Go and 247 session and user
conversion rates over time, as set out in Figure 6.8 below. The Parties’ session conversion
rates are similarly low, but Blinds2Go’s user conversion rate is [] higher ([] % versus [])
because consumers visiting Blinds2Go’s website are more likely to return and ultimately make
a purchase. In contrast, []. Moreover, 247’s user conversion rate has fallen materially over
time from [] in 2014 to [] in 2019. This is consistent with its decline in market share, [].
These facts do not suggest that 247 is a close competitor to Blinds2Go.
Figure 6.8: Session and user conversion rates – Blinds2Go and 247 (2014-2019)
[]
Source: Google analytics; AlixPartners analysis (Annex 0098)
The market for M2M online blinds is highly competitive
6.38 At paragraph 183 of the Phase I Decision, the CMA indicates that Interior Goods Direct is
considered to be the main constraint on the Parties and that other, smaller, online M2M blind
retailers constrain the Parties to a less significant degree. Further, at paragraph 198, the CMA
disregards the competitive constraint exercised by omni-channel retailers on the basis that they
are more distant competitors because of differences in business model and commercial focus,
with less focus on their M2M blinds products, and different functionality of webstores.
6.39 Accordingly, the CMA’s approach disregards the competitive constraint exercised by most of
the pure online retailers and omni-channel retailers, either individually or in combination. In
reality, competitors of both types exercise a significant constraint on the Parties, even more so
NON-CONFIDENTIAL
VERSION
43
when taken into account in combination. In some markets, small competitors cannot compete
as they lack economies of scale or suffer from capacity constraints. However, as set out fully
in Chapter 7, this is not a market where economies of scale are material and smaller retailers
can compete effective across multiple dimensions (namely price, quality, customer service, and
ranges are not a decisive differentiator), nor do they face any type of capacity constraints as
the products come directly from their blinds suppliers. There is therefore no evidence that
smaller competitors face any constraint competing with larger rivals.
6.40 Evidence from the BDRC Survey suggests that the Parties’ customers compare products/prices
across many websites. For Blinds2Go’s customers, 247 is the [] most used source of price
comparisons, behind [].
Figure 6.9: Websites visited before orders placed
[]
Source: BDRC Survey
6.41 As emphasised in Chapter 4, competitors to the Parties include traditional home decoration
retailers (who have developed their own online channels), online home furnishings specialists
and online blinds specialists, all of whom compete actively and for whom there is no indication
that their competitive threat will be reduced post-merger. This is reflected in the survey results
for Dunelm (which only entered the online M2M sector in March 2018), John Lewis and Next
(which only materially entered the online M2M sector in Autumn 2019).
6.42 Online competitors, including omni-channel competitors, are highly comparable in terms of
price, product, range and customer service. At Annex 0099, the CMA will find a selection of
screenshots of the roller blind sections of the websites of Blinds2Go, 247 and a selection of
online only competitors. With the exception of Concept Blinds, all providers use an almost
identical format for the presentation of products, with graphic design the only evident difference.
Omni-channel retailers also follow the same format, as set out in Annex 0100 containing
screenshots of the roller blind pages of the omni-channel retailers. In all cases, customers are
able to select from a range of fabrics before inputting their own measurements to obtain a quote
for a M2M blind. Objectively, there is little difference between the offerings regardless of the
scale of the provider.
6.43 The BDRC Survey also indicated that [] of 247’s customers visited physical stores before
placing orders and this was also the case for [] of Blinds2Go’s customers. Figure 6.10 shows
the physical stores visited by these customers.
NON-CONFIDENTIAL
VERSION
44
Figure 6.10: Stores visited before orders placed
[]
Source: BDRC Survey.
6.44 This demonstrates that a significant number of customers are using physical stores to compare
products/prices before making their purchase online, suggesting online retailers such as the
Parties are constrained by the window covering offers available in-store.
6.45 The next-sub-section supplements some of the competitor descriptions set out at Chapter 4. It
should be noted at the outset that all of the competitors set out below in Table 6.2 offer the
same core product range as Blinds2Go and 247, such as Venetian Blinds, Roller Blinds,
Vertical Blinds and Roman Blinds.
Table 6.2: Product Offerings of selected competitors
Company
Roman
Roller
Venetian
Wooden
Venetian
Vertical
Blinds2Go
982
947
108
225
191
247
475
615
169
245
279
Interior Goods Direct
2,603
623
81
41
254
Swift Direct
455
652
159
386
410
Blinds UK
1,180
309
50
117
470
Unbeatable Blinds
62
209
156
58
136
Wilsons Blinds
173
499
231
168
416
Blinds4uk
237
388
257
146
328
English Blinds
387
344
74
280
317
Terrys Fabrics
638
314
169
92
69
So Easy Blinds
258
851
122
131
747
Concept Blinds
120
253
90
132
272
Make My Blinds
62
260
39
134
135
Dunelm
66
178
24
13
4
John Lewis
438
56
1
2
5
Next
532
267
6
-
39
Source: Retailer websites – 16 April 2020
6.46 Interior Goods Direct (Blinds Direct) is rightly acknowledged by the CMA as a serious
competitor to both Blinds2Go and 247. Interior Goods Direct launched its Blinds Direct website
in 2017 to consolidate several different window covering websites and has expanded rapidly
since to overtake 247 as the number two supplier of online M2M blinds. Its M2M retail operation
was established in 2004 and is vertically integrated into blinds manufacture, with an estimated
NON-CONFIDENTIAL
VERSION
45
UK turnover of M2M blinds in 2019 of £18.7m.
76
This vertical integration allows IDG to stock
over 1,000 M2M products offering next day delivery. Interior Goods Direct have also recently
acquired Wilsons Blinds, an online M2M blinds supplier established in 2005. Wilsons Blinds
has a similar business model to that of 247 and Blinds2Go but is expected to benefit from
accelerated growth through the benefits of Interior Good Direct’s vertical integration. The
estimated turnover for Wilson Blinds in 2018 was £2.1 million.
77
6.47 Blinds4UK has over 35 years’ experience in window coverings. On its website, it states that it
is the first choice for designers and project managers due to its easy-to-use website and fast
response times on sample orders and customer service. Blinds4UK has had its blinds featured
on BBC Grand Designs and In-House Beautiful magazine, and they are rated excellent on
Trustpilot with over 3,000 reviews. Evidence of their competitive constraint comes through
strongly in the BDRC survey with approximately a quarter of customers visiting the retailer
before making their purchase with the Parties (see Figure 6.10 above).
6.48 Make My Blinds entered in 2015 and has experienced rapid growth due to its high-quality
website and customer experience. Make My Blinds has a similar business model to 247 and
has developed good supplier relationships in China and Poland.
6.49 Swift Direct Blinds (formerly Direct Blinds) was established as a supplier of blinds in 1969,
recently moving online in 2012. Their confidence in the quality of their blinds is such that they
offer a five-year guarantee on all products. Like Interior Goods Direct, Swift Direct Blinds is
also vertically integrated.
6.50 Terrys Fabrics was established 46 years ago and specialises in fabrics and M2M blinds. It
has a retail warehouse premises in Staffordshire, but primarily carries out its M2M business
online. Its website is well designed and offers payment through Klarna, a credit provider. Terrys
has been featured on ITV’s This Morning and in the Good Housekeeping magazine. In August
2019, Terrys released an augmented reality window dressing app called Window Planner, to
help customers visualise their blinds in their home before placing an order.
6.51 Concept Blinds is rated 4.9 out of 5 on Trustpilot and have been online for 14 years. It has
sold over 2 million blinds to date and claim to have the lowest UK prices. It is open to trade
enquiries and offers a three-year warranty on all of its products. It offers a price promise, doing
their best to match any listed price for an equivalent product on another online site 48 hours
after purchase.
78
76
Annex 0002, Tab 3.1.
77
Annex 0002, Tab 3.1.
78
https://www.conceptblindsdirect.co.uk/pages/Price_Promise.html
NON-CONFIDENTIAL
VERSION
46
6.52 At the Issues Meeting, the Parties listed out over 50 online blinds specialists (including providing
details of their websites/product offerings) and there are many hundreds of registered urls
containing the word blinds.
Omni-channel retailers
Dunelm offers a range of M2M blinds with the added service of online consultation with one of
their “Made to Measure” consultants, entering the online M2M sector for the first time in March
2018. This new virtual consultation service enriches the online experience by offering guidance
on measuring and suitable styles and fabrics. It has a wide range of fabrics which it sets out for
customer browsing in its online brochure, and it also offers a wide range of ready-made blinds.
79
Dunelm []
6.53 John Lewis has a very strong home furnishing brand and offers a wide range of M2M blinds
online. It has significant potential for expansion, offering a selection of plain blinds but also
offering patterns from influential designers such as Orla Kiely. Its website allows specific
measurements and customisations to be specified as per the Blinds2Go and 247 online user
experience.
80
John Lewis is also continually shifting more business online, especially in light
of the recent Covid-19 outbreak. It reported that its online sales were up 84% year on year
from mid-March to mid-April 2020.
81
6.54 Next began selling M2M blinds online materially in Autumn 2019, with a current range of 821
SKUs. It has a very strong brand both in-store and online and is well renowned for its home
furnishings business unit, “Next Home”. It has a significant presence on the high street and
retail parks with 500 stores across the UK, giving it heightened brand awareness over online-
only retailers. This also allows it to provide better customer service through in-store delivery
(“click and collect”). In the financial year ending January 2020, Next’s online revenue exceeded
in-store revenue for the first time
82
.
79
https://dunelm-prod.cdn.prismic.io/dunelm-prod/e3e31705-3474-4d17-9295-
c624e8fa20a5_M2M_Catalogue_09_2020_v1.pdf
80
https://www.johnlewis.com/orla-kiely-multi-stem-made-to-measure-roman-blind-
multi/p3153307?roman=true
81
https://www.drapersonline.com/news/john-lewis-online-sales-and-profits-up/7040193.article
82
https://www.nextplc.co.uk/~/media/Files/N/Next-PLC-V2/documents/2020/annual-report-and-accounts-
jan20.pdf
NON-CONFIDENTIAL
VERSION
47
Marketplaces
6.55 As mentioned in Chapter 4, eBay and Amazon marketplaces are becoming growing threats to
the Parties’ business model and, in the case of Amazon, becoming a distinct competitor in its
own right.
6.56 eBay has a specific made to measure category on its marketplace platform. Sellers such as
“theblindsoutlet”, with 183,951 positive reviews, have a wide range of colours on various blinds,
stocking Wooden Venetian Blinds, Roller Blinds, Roman Blinds and more. The ordering
process involves the user selecting their size “up to” a certain height and width, paying a fixed
price, then emailing the seller with the precise measurements. The blind will then be cut
precisely to the customer’s exact measurements. Hunter Douglas estimates that sellers on the
eBay platform sold £30.9m of online M2M blinds in 2018.
6.57 Amazon is already growing its blinds business through both ready-made and M2M products
on its marketplace service and through its own ready-made products. Amazon’s entry into the
M2M blind market is considered a significant threat to 247 and Blinds2Go’s business model.
83
Swift Direct Blinds already sell through Amazon’s marketplace, as well as marketplace seller
“First blinds” and [].
Diversion
6.58 The BDRC Survey asked a forced diversion questions in the form of “Imagine that before
starting your shop on [Order date] you knew that [Online Brand] had stopped selling blinds.
Thinking of all the options open to you, what would you be most likely to have done instead?”
The results show that the majority of customers would divert to another online M2M blind
retailer, with []% of customers preferring ready-made blinds, []% in-store M2M, []% in-
home M2M blinds, or []% an alternative window covering such as curtains or shutters. These
results are shown in Figure 6.11.
Figure 6.11: Forced diversion to alternative options
[]
Source: BDRC Survey
6.59 Customers responding to each of the first four categories in Figure 6.11 were then presented
with a tailored list of retailers. Considering first Blinds2Go’s customers, []% of them would
83
As highlighted in 247‘s site visit deck.
NON-CONFIDENTIAL
VERSION
48
have switched to 247, behind Dunelm (at []%), Blinds Direct ([]%) and John Lewis ([]%).
This is shown in Figure 6.12 below.
Figure 6.12: Forced diversion to alternative options – Blinds2Go customers
[]
Source: BDRC Survey
Notes: Hunter Douglas includes Hillarys and Web Blinds
6.60 Turning to 247 customers, []% stated that []. Other material diversion sources include
[]%.
Figure 6.13: Forced diversion to alternative options – 247 customers
[]
Source: BDRC Survey
Notes: Hunter Douglas includes Blinds2Go ([
]%), Hillarys ([
]%), and Web Blinds ([
]%)
6.61 It should also be noted that:
(i) These responses are likely to understate in-store diversion. Whilst the survey was
carefully drafted to explore consumers’ recent purchase journeys pre-Covid-19, store
closures may have affected the answers given;
(ii) Interior Goods Direct has proved very successful at winning 247’s customers and has
displaced 247 as the second largest competitor, with the CMA’s market share
estimates indicating that 247’s share has fallen from 10-20% in 2016 to 5-10% in 2019
and Interior Goods Direct’s share has correspondingly increased from 5-10% to 10-
20%. Blinds2Go/Web Blinds’ market share has remained the same at 50-60% over
this period;
(iii) Dunelm only entered the online sector in March 2018 (and ceased operating
temporarily online in November 2019 due to a website change and has only recently
re-opened), and Next only materially entered the online sector in Autumn 2019.
6.62 In addition, Hunter Douglas does not have a financial interest in 247 ceding sales or profits to
Blinds2Go, [].
Out of market constraints on made to measure blinds
6.63 In the Phase I Decision, the CMA adopts a narrow approach to the relevant Candidate Market,
limiting it to the online retail of M2M blinds and asserts that there is no material “out of market”
constraint. In doing so, it excludes from the relevant market the constraint exercised by:
NON-CONFIDENTIAL
VERSION
49
(i) In-store retailers of M2M blinds;
(ii) In-home retailers of M2M blinds;
(iii) The retail of ready-made blinds; and
(iv) The retail of other forms of window covering (curtains and shutters).
6.64 Even if they are not considered as part of the relevant market, it is highly relevant to consider
how such out-of-market factors exercise a constraint on the activities of the Parties. However,
the Decision fails to do so to any meaningful extent.
6.65 The BDRC Survey asked directly what other window coverings they would consider purchasing
instead. The results are presented in Figure 6.14 below which shows that almost half would
consider alternatives to M2M blinds. This includes []% and []% of Blinds2Go’s and 247’s
customers respectively considering ready-made blinds, []% and []% considering Shutters,
and []% and []% considering Curtains.
Figure 6.14: Consideration of alternative window covering products
[]
Source: BDRC Survey
6.66 As discussed above, Blinds2Go’s strategy has focused on competing directly with ready-made
blinds and offline M2M spend. Its growth has not been achieved by winning sales from other
online M2M retailers and it has not focused on competing head on with other online M2M blinds
suppliers.
6.67 This strategy is commercially justified by the high proportion of the Parties’ products with a
ready-made equivalent. In early May 2020, there were over 440 different ready-made size
combinations available online among the top ready-made blinds retailers. Table 6.3 below
shows the number of product combinations (i.e. type, size, colour, and price) sold by the 10 top
ready-made blind retailers (representing approximately 70% of the market in 2018)
84
. These
products include 352 different size combinations, with an additional 93 combinations available
from eBay. This suggests consumers have a large range of ready-made options for their
window covering needs.
84
Annex 002, Tab 3.8.
NON-CONFIDENTIAL
VERSION
50
Table 6.3: Ready-made blind product combinations available online – Top 10
competitors (May 2020)
Product
Amazon
Argos
B&M
B&Q
Dunelm
Harry
Corry
Home-
base
John
Lewis
The
Range
Wilko
Day Night
134
9
0
35
0
0
0
5
0
0
Pleated
468
0
0
0
39
0
0
0
0
0
Roller
281
84
42
135
399
150
76
235
188
39
Roman
31
5
0
8
142
0
0
57
0
0
Venetian
290
103
0
62
263
34
37
19
93
26
Vertical
16
12
0
1
22
0
0
0
17
0
Total
1,220
372
42
241
865
184
113
316
298
65
Source: Retailer websites in May 2020. See Annex 0101).
Notes 1: Top 10 is based on market share in 2018. eBay is excluded as product
colour combinations were not easily accessible.
Notes 2: A product is based on a unique type, size, colour and price.
Notes 3: Amazon is based on taking a sample of the products listed in the top “best
sellers” list for each category.
Notes 4: No prices were used to distinguish between products for Amazon.
Notes 5: Colours have been standardised.
6.68 Further, Table 6.4 shows the number of these products that overlap with the Parties’ M2M
products, based separately on size and type.
Table 6.4: Percentage of online M2M blind sales for which there is a ready-made
equivalent (May 2020)
Blinds2Go
247 – All recess
2
247 – Modelled
Recess
3
By dimension
1
47.5%
34.9%
54.0%
By dimension and product
type
39.8%
25.6%
44.9%
Source: Retailer websites in May 2020. See Annex 0102.
Notes 1: Includes all retailers listed in Table 6.3, including eBay. The methodology for these
percentages is set out in Annex 0090.
Notes 2: Modelled on the assumption that all 247 blinds are inside recess measurements and
therefore have more strict tolerances.
Notes 3: Modelled using the uplift factor from Blinds2Go between inside recess vs outside
recess.
NON-CONFIDENTIAL
VERSION
51
6.69 This evidence clearly shows that if consumers were only interested in blind dimensions, 48%
of Blinds2Go’s and up to 54% 247’s products would be available in a ready-made equivalent.
Even assuming a consumer wanted a particular type of blinds (e.g. roller or roman), then almost
40% of Blinds2Go’s and 45% of 247’s would have a ready-made equivalent
6.70 In any event, as set out in Hunter Douglas’s response to Question 3 of the first section 109
Notice of 20 April 2020, these percentages are likely to significantly understate the competitive
constraint imposed by ready-made blind retailers due to the analysis only including c.70% of
the market, the ability for customers to modify ready-made blinds to suit their needs, and the
ability for blinds to be fitted outside the recess, where tolerances for width and drop dimensions
are much wider.
6.71 As at May 2020, ready-made blind alternatives are also priced very competitively compared
with M2M. The analysis below shows the average price of the equivalent ready-made
alternative(s) for each of Blinds2Go’s M2M blinds sales in March 2020.
85
For a ready-made
blind to be considered equivalent, it must have the same blind type, and have dimensions within
the tolerances described in response to Question 3 of the 20 April S109.
6.72 For example, the average ready-made roller was £23.72, while Blinds2Go’s equivalent sized
M2M blinds were £[]. Similarly, for venetian blinds, the average ready-made was £32.57,
while Blinds2Go’s equivalents were £[]. This is shown in Figure 6.15 below.
Figure 6.15: Price comparison by product type – Blinds2Go vs. Ready-made retailers
(March 2020)
[]
Source: Retailer websites; Blinds2Go sales data; AlixPartners analysis (Annex 0103).
Notes 1: This excludes Amazon and eBay as the Parties could not scrape pricing information
from these competitors in the time available.
Notes 2: Day Night only represents 0.1% of sales (March 2020) and is therefore not
representative.
6.73 While there are differences in quality of some ready-made products, even for products sold by
John Lewis and Dunelm (who in the Parties’ view have products with comparable quality),
ready-made blinds are still priced to undercut Blinds2Go’s products and compete directly with
M2M. For example, Dunelm’s vertical ready-made blinds are on average £[] cheaper than
85
These sales only include those for which a full retail price was available (excluding c.5% of sales), and for
which a ready-made equivalent could not be found (c. 70%) – i.e. 25% of March 2020 online M2M blind
sales are included in Figure 6.15 and 6.16.
NON-CONFIDENTIAL
VERSION
52
Blinds2Go, while John Lewis’s ready-made venetian blinds are only £[] cheaper. This
analysis is shown in Figure 6.16 below.
Figure 6.16: Price comparison by product type and retailer – Blinds2Go vs. Ready-made
retailers (March 2020)
[]
Source: Retailer websites; Blinds2Go sales data (March 2020); AlixPartners analysis (Annex
0103).
Notes: This excludes Amazon and eBay as the Parties could not scrape pricing information
from these competitors in the time available.
6.74 It is clear therefore that consumers have a wide variety of ready-made blind alternatives
available should any worsening of the Parties’ offer arise post-merger.
6.75 In addition to ready-made blinds, the Parties currently compete directly with retailers of offline
M2M blinds, both in-store and in-home. As presented in Annex 0002, Hunter Douglas
estimates that from 2017 to 2019 the offline M2M market fell from approximately £331m to
£314m (a fall of c.5%). Over the same period, Hunter Douglas estimates that the online M2M
market grew by £56m (an increase of c.28%). While Blinds2Go and other online M2M retailers
have been successful in growing the sales of that particular channel from outside of M2M blinds
(likely from ready-made blinds online), there has clearly been some substitution away from the
offline channel, implying customers are increasingly seeing these channels are readily
interchangeable.
6.76 Finally, the CMA should not lose sight of the constraint provided by different window coverings
including curtains and shutters. The Phase 2 Issues Statement carries forward the narrow
frame of reference adopted at Phase 1. However, the Parties contend that the market is wider
than M2M blinds, and also wider than blinds more generally.
6.77 As set out in the Response to the Issues Letter at Phase 1, the AMA report highlights shifts in
demand between blinds, curtains and shutters, with blinds and shutter purchases increasing at
the expense of curtains.
86
In addition, as mentioned above, the Customer Survey highlights
that []% of customers consider curtains and []% consider shutters before purchasing M2M
blinds online, suggesting that the competitive constraints imposed on the Parties are not just
from the retail of online M2M blinds. Any competitive analysis must therefore consider the
extent to which retailers of other window covering products exercise a competitive constraint
on the activities of the Parties.
86
Response to the Issues Statement – Phase 1, paragraph 2.3.
NON-CONFIDENTIAL
VERSION
53
Price competition
6.78 Since [], Blinds2Go has []. As explained at the virtual site visit on 27 April 2020, this did
not include []. In Autumn 2019, []. Blinds2Go also [].
6.79 Figure 6.17 sets out the yearly average weekly prices across the various scraped products.
Notwithstanding its efforts to offer higher quality prices and its market leading position in the
niche sector of online M2M blinds, Blinds2Go’s prices are not generally higher than its
competitors and in certain categories its prices are lower than a number of its scraped rivals
(such as White Venetian, White Faux Wood and Standard White Roman). A noteworthy
exception to this is [].
Figure 6.17: Blinds2Go prices relative to scraped competitors – by product (2017 –
2019)
[]
Source: Blinds2Go; AlixPartners analysis (Annex 0104)
6.80 This weekly pricing data also permits a detailed analysis of price trends over time, and the
extent to which weekly price changes by Blinds2Go correspond with these various rivals moving
their prices in the same direction across the various product categories. This does not suggest
that Blinds2Go and 247 track each other’s prices any more closely than other competitors (and
certainly not in 2018 and 2019), which is inconsistent with the theory in the Phase I Decision
that they are particularly close rivals. This is shown in Figure 6.18 below.
Figure 6.18: Blinds2Go prices relative to scraped competitors (April 2016 – March 2020)
[]
Source: Blinds2Go; AlixPartners analysis (Annex 0105)
6.81 This analysis is presented in a slightly different way in Figure 6.19 and aggregated over last
15 months to highlight the most recent period of price competition. Similarly, this does not
suggest that Blinds2Go’s prices move in line with 247’s relative to other competitors.
Figure 6.19: Blinds2Go prices relative to scraped competitors (January 2019 – March
2020)
[]
Source: Blinds2Go; AlixPartners analysis (Annex 0105)
NON-CONFIDENTIAL
VERSION
54
The market is dynamically competitive
6.82 The above analysis largely focuses on the current competitive conditions. However, as
emphasised at the site visit on 27 April 2020, Blinds2Go is mindful of the dynamically
competitive history of the niche sector in which it is active. As emphasised in Chapter 5, in
2011/12, Blinds2Go was the number 3 online retailer of M2M blinds in the UK, with sales of just
[]. At the time, 247 was the largest retailer, with sales of £[]million. Since then, Blinds2Go
has grown its sales of M2M blinds online in the UK almost [] to over £[]million in 2018 (the
last financial year before the 2019 Transaction) whilst 247 has only grown its sales by []% to
around £[]million over the same period.
6.83 Moreover, the CMA estimates indicate that 247’s market share has fallen from 10-20% in 2016
to 5-10% in 2019, whereas Interior Goods Direct/Blinds Direct’s market share increased from
5-10% to 10-20%.
87
Accordingly, 247 is now the third largest competitor, or the fourth if eBay
were to be treated as a single rival. In short, 247 was the market leader, but has been leap
frogged by both Blinds2Go and Interior Good Direct, who were able to expand their sales
rapidly, principally through improving their websites and getting the right range and pricing of
products to appeal to consumers.
6.84 The question for the CMA is not whether Blinds2Go can be replicated so as to restore pre-
merger rivalry, but whether rivals collectively have the ability and incentives to replicate any
loss of rivalry from 247 and in the context of its market share declining. As discussed above,
Hunter Douglas considers that there is an array of competitors capable of replicating 247’s
business model.
87
Table 1 of the Phase I Decision and paragraphs 145-146 of Hunter Douglas/Bellotto (2017).
NON-CONFIDENTIAL
VERSION
55
7. Barriers to entry and expansion
Overview of section
7.1 There is no question that Blinds2Go is the number one player in the online M2M blinds sector
with the CMA estimating that its share is 50-60% in Phase 1. Blinds2Go has developed a
compelling consumer proposition that has enabled it to grow rapidly, taking considerable share
from ready-made blinds retailers and in-store/in-home M2M retailers. The merger does not
change that situation. Post-merger Blinds2Go will continue to be the number one player and
the direct competitive constraint that Blinds2Go imposes on other online M2M blinds retailers
will be unaffected.
7.2 Chapter 6 set out the evidence that 247 is not a material competitive constraint on Blinds2Go,
such that there cannot be any expectation that a reduction or removal of that constraint might
enable Blinds2Go to increase its prices or worsen its offering. In addition, even if there was
some reduction in rivalry, reality shows that there has already been material entry and
expansion since 2016, which exceeds 247’s estimated market share of 5-10% (which in any
event the Parties consider overstates 247’s market share). This has come from a combination
of the expansion of IGD, Make My Blinds, Dunelm, Next, and other as discussed below.
7.3 To emphasise the important points made in Chapters 5 and 6 above, 247 is a relatively small
player in the online M2M blinds sector with a 5-10% market share based on the CMA’s
estimates from Phase 1 (the Parties consider that these estimates overstate their market
shares, and they are mainly referred to so as to highlight the differences in the Parties’ shares
and trends in market shares over time). The fact that 247’s offering is relatively undifferentiated
with extensive use of off-the-shelf technical solutions, means it has been easily replicated by
rivals, leading to a steady decline in its market share for a number of years. Entry or expansion
simply needs to replicate 247’s degree of rivalry.
7.4 The remainder of this Chapter is structured as follows:
(i) Section (b) explains that the elements required to enter the online M2M blinds segment
are readily available, and there are no material economies of scale that would
constitute a significant barrier to an entrant or smaller rival that made a determined
effort to increase its sales rapidly. The Parties’ own recent experience of entering
markets outside the UK provides further evidence of this.
(ii) Section (c) sets out the evidence that such entry and expansion is likely, is expected
to happen soon (and certainly within two years), and will be of significant scale to
replicate the limited rivalry from 247.
NON-CONFIDENTIAL
VERSION
56
The elements required to successfully enter the online M2M blinds market are readily available
7.5 As set out in the Parties’ RFI responses, there are no material barriers to entry or expansion in
the market for the retail supply (online) of M2M blinds. The absence of any significant upfront
costs, economies of scale or other incumbency advantages mean that the market is highly
dynamically competitive. This is demonstrated by various examples of successful recent entry
both in the UK and in other countries.
Upfront costs are modest
7.6 As explained in the Parties’ RFI responses, there are very limited upfront costs that prospective
entrants in the online retail market for M2M blinds would have to absorb. Both Blinds2Go and
247 started their businesses without any significant initial capital or experience and have grown
organically since.
88
While the market has matured since the Parties started their respective
businesses, there is nothing that would prevent prospective entrants from replicating the
Parties’ approach.
7.7 To develop a successful online M2M business, new entrants need to (a) develop a website (or
use an alternative third party sales platform, particularly Amazon, eBay or Wayfair), (b) find a
supplier of blinds, and (c) market their products to consumers. We discuss these three
elements and associated costs in the remainder of this section. Before doing so, it is worth
pointing out that – contrary to many other e-commerce markets – there are several upfront
investments that entrants do not have to make:
(i) Office premises: dedicated office premises are not required at the initial stage of the
development – the Blinds2Go website was developed in the founders’ bedrooms.
(ii) Distribution centres: suppliers of M2M blinds can send the products direct to the final
customers. The only logistics functions carried out by 247 relates to the dispatch of
samples and dealing with returns, which require very limited warehousing space.
[].
89
(iii) Inventory financing: as a result of the supply chain arrangements, online M2M retailers
do not carry any inventory. Further, as the Parties are paid by the customer before
they are invoiced by their suppliers, the businesses do not require any working capital
and are cash generative.
88
See paragraph 8.1 of 247’s RFI response, and paragraph 8.2 of Hunter Douglas’ RFI response.
89
See paragraph 13.1 of Hunter Douglas’ RFI response.
NON-CONFIDENTIAL
VERSION
57
7.8 As regards the three types of upfront costs that entrants would have to incur, the Parties note
that these are by no means prohibitive or insurmountable.
Website development costs
7.9 The costs associated with website development are low or, as explored below, can be avoided
altogether by selling through online marketplaces such as Amazon, eBay or Wayfair. While
there would be costs associated with replicating the Blinds2Go website, which was developed
in-house over many years, the same cannot be said of 247’s site. As noted in 247’s RFI
response, the website is based on the Magento open source platform and its original
development costs amounted to just £[], with a further £[] spent on a mobile friendly
redesign in 2016.
90
New entrants could adopt a similar approach by picking one of a range of
readily available e-commerce platforms, such as Shopify (as discussed at the virtual site visit).
91
7.10 The costs related to the continuous improvement and development of the website would
account for only a small share of revenue for any successful retailers. As explained in 247’s
RFI response, website and new product development costs accounted for [] of its total costs
in 2019 (or less than £[] across all of its websites).
92
The information submitted in response
to question 13 of the first section 109 notice of 20 April 2020 supports the view that the costs
of developing and hosting a website of comparable quality to that of 247 are not prohibitive.
Website and visual imagery development accounted for an average of £[] and £[]per year
over the period 2017 - 2019.
93
In addition, 247 []. Other website related costs, listed in
Annex 0010, are insignificant.
94
The situation is not dissimilar for Blinds2Go, which estimates
that its total website development costs amounted to [].
95
7.11 New entrants could also choose to avoid website development costs altogether by selling
through established marketplaces such as Amazon, eBay or Wayfair. As set out in both Parties’
RFI responses, Amazon has recently made significant improvements to its website that allow
90
247 RFI response, paragraph 8.4.
91
Various e-commerce platforms other than Magento are readily available. These include Shopify,
Bigcommerce, Volusion, etc. See e.g. https://www.techradar.com/uk/news/the-best-ecommerce-platform.
92
247 RFI response, paragraph 13.4.
93
247 S109 response, annex 0008 and 0009.
94
Annex 0010 to 247’ S109 response.
95
Annex 0055 to Hunter Douglas’ S109 response.
NON-CONFIDENTIAL
VERSION
58
for a more user-friendly customisation of products (see Figure 7.1 below).
96
In this regard it is
worth noting that, in January 2020, [].
7.12 As explained in 247’s RFI response, the commission charged by Amazon does not place
retailers who choose this route at a competitive disadvantage vis-à-vis incumbents. Amazon’s
commission, which currently stands at 15% (and is negotiable
97
), covers various cost elements
that incumbents who have their own websites need to cover through their gross margin. These
include not only website development and hosting costs, but also payment processing fees as
well as digital advertising costs. Regarding the latter, it is worth noting that almost 90% of UK
shoppers use Amazon and 70% of these use the site as their first point of call.
98
This generates
substantial free traffic for Amazon marketplace retailers.
Figure 7.1: Swift Direct Blinds Amazon customisation
Source: https://www.amazon.co.uk/Swift-Direct-Blinds-Stockholm-
individually/dp/B086G9M6KF?th=1.
Establishing supplier relationships
7.13 There are no direct costs associated with the establishment of supplier relationships, apart from
the occasional phone calls and management time. Finding such suppliers is relatively easy.
There is a wide range of suppliers, both in the UK and overseas, all of which can arrange for
direct shipment to customers. Some large suppliers, such as Decora, have a sufficiently large
96
247 RFI response, paragraph 8.4 and Hunter Douglas RFI response paragraph 9.5.
97
See 247 RFI response, paragraph 8.4.
98
See https://www.theguardian.com/business/2019/mar/07/almost-90-of-uk-shoppers-use-amazon-
research-reveals.
NON-CONFIDENTIAL
VERSION
59
range of products to allow new entrants to compete with just a single supplier relationship.
Further, there are many smaller suppliers who actively try to break into the UK market. Indeed,
247 is contacted regularly by suppliers – mostly from East Asia – who wish to sell their products
through its website.
7.14 As far as Hunter Douglas is aware, existing retailers’ agreements with suppliers do not contain
any formal exclusivity clauses that would prevent a new entrant from obtaining supply. 247 in
particular [].
99
As []accounted for []of 247’s total blinds purchases in 2019
100
and does
not sell products that could not be sourced from other suppliers, there is nothing preventing a
new entrant from replicating the range of products offered by 247.
7.15 Finally, as discussed in further detail below, there are various existing omni-channel retailers
who already have established supplier relationships who do not have to overcome this hurdle.
The same is true for blinds manufacturers who already operate in the UK and Ireland, such as
Decora. The entry of Swift Direct Blinds and Interior Goods Direct is a good example of
manufacturers who saw the opportunity offered by online retail and began selling products
directly to consumers.
Marketing costs
7.16 Prospective online retailers of M2M blinds would need to invest in marketing to increase the
visibility of their website either through PPC advertising or other means. As with many online
markets, a significant proportion of marketing budgets is allocated to PPC, mainly with Google.
While new entrants have a variety of options for engaging with consumers, there is nothing that
prevents them from pursuing a similar strategy to Blinds2Go and 247 and competing directly
with them, even if they have limited working capital. This is for the following reasons:
(i) First, PPC costs are entirely variable and not subject to any economies of scale. While
the first element is obvious – Google charges on a Cost Per Click (CPC) basis – the
second point requires some further discussion. The issue is explored further below.
(ii) Second, the marginal cost for PPC advertising charged by Google – the CPC – only
depends on the quality of the ad and landing page and not on historical traffic volumes
or other factors that may favour incumbents. A more detailed explanation is provided
below. It is noteworthy that Google’s ad quality measure – as summarised by the quality
score – can be built up very quickly by new entrants. As explained below, Blinds2Go
[].
99
Response to CMA’s seventh Phase 1 S109 notice dated 30 January 2020, paragraph 4.8.
100
Calculated based on information provided in Annex 0002 and 0024 of 247’s S109 response.
NON-CONFIDENTIAL
VERSION
60
(iii) Third, CPCs can be controlled by bidding for less common keywords where auctions
are less competitive. For example, [].
101
Alternatively, bids can be focused on certain
geographical locations, audiences or times of the day to reduce or/ limit expenditure.
This can provide a viable route to entry for retailers who wish to “test the water” before
increasing their PPC spend more significantly. The fact that [] suggests that such a
strategy could be viable.
(iv) Finally, PPC advertising does not require any significant working capital. As set out in
Hunter Douglas’ RFI response
102
, PPC advertising is cash generative as the retailer
receives cash from the customer at the point of order, weeks or months before PPC
charges or supplier invoices fall due. It is therefore easy for a new entrant to begin
generating cash before any costs at all are incurred.
7.17 While paid search advertising via Google Ads is a viable route to market, it is by no means the
only way to reach customers online. There are other channels such as Google Shopping,
organic search, referral, email, or social media. Further, established omni-channel retailers
and manufacturers could leverage the strength of their existing brands to attract free traffic and
cross-sell blinds from other parts of their home furnishings business.
7.18 Google Shopping currently operates on a PPC model in line with Google Ads, which means
that any marketing strategy based on Google Shopping would be the same. However, as
discussed in Chapter 6, Google has decided to move away from this model and to allow for
free product listings alongside paid ads. This will provide a new free route to market for online
retailers.
7.19 Organic search, which accounts for a significant share of the Parties’ website traffic ([] for
247 and [] for Blinds2Go in 2019), already provides a free route to market, though some
investment in SEO is required. As explained in Hunter Douglas’ RFI response, Swift Direct
Blinds concentrated its efforts on this channel and regularly featured in the number one position
for organic search.
103
Retailers who wish to follow this route, but do not want to pay for the
SEO cost upfront, could replicate Swift Direct Blinds’ approach. As set out in its RFI
response
104
, 247 believes that Swift Direct Blinds entered into a profit-sharing agreement with
its SEO service provider, which meant that these costs became variable. Finally, it is worth
noting that neither Blinds2Go nor 247 spend any significant amount on SEO themselves. As
101
See Annex 0058 of Hunter Douglas’ S109 response.
102
Hunter Douglas RFI response, paragraph 11.5.
103
Hunter Douglas RFI response, paragraph 10.5.
104
247 RFI response, paragraph 10.6.
NON-CONFIDENTIAL
VERSION
61
discussed in Chapter 6, Blinds 2Go []. 247 outsources its SEO activities to an external
agency, Journey Further, at a cost of []in 2019.
7.20 Social media channels such as Facebook and Instagram provide an attractive alternative route
to market. As explained at paragraph 5.14, competitors such as Make My Blinds have made
extensive use of this channel to reach sales of £2.3 million by 2019. Instagram, a network on
which 247 is particularly weak (only 5,000 followers vs nearly 35,000 for Make My Blinds), is a
free and highly visible channel.
7.21 Finally, established omni-channel retailers or manufacturers can use the strength of their brand
names to attract traffic from branded searches. Next is already one of the most visited websites
in the UK with a strong brand recognition, allowing it to attract consumers at limited or no costs.
Dunelm similarly benefits from its established brand name, allowing it to circumvent the PPC
route, as well as reducing the effective cost of PPC advertising (as a stronger brand is expected
to lead to higher click through rates and user conversion rates). As noted in Chapter 6, the
Google Trends analysis shows that consumers who search for blinds often search for brands
such as Dunelm as well.
There are limited economies of scale and the costs to expand are modest
7.22 The data provided in the Parties’ respective RFI responses
105
clearly shows that there are no
economies of scale with regards to three of the Parities’ four major cost centres (together the
four cost centres accounted for []% of total costs for both Parties):
(i) Cost of Goods Sold (including distribution), which accounted for []% and []% of
247’s and Blinds2Go’s total costs in 2019 respectively, have not decreased as a
percentage of revenue despite significant growth of both businesses. Scale clearly
does not therefore confer any significant buyer power, either with regards to blinds or
distribution costs.
(ii) Staff costs, which accounted for []% and []% of 247’s and Blinds2Go’s total costs
in 2019 respectively, have shown no clear trend over time. For both businesses, staff
costs initially fell as a share of revenue but have since risen again to higher levels. The
lack of economies of scale can be explained by the fact that most employees work in
the customer service and samples departments – areas that need to grow in line with
sales. Further, as retailers become larger (as in the case of Blinds2Go) they need to
hire more staff in management positions (e.g. in HR), leading to additional costs that
smaller companies can avoid.
105
Hunter Douglas RFI response to Q13 and 247 RFI response to Q13.
NON-CONFIDENTIAL
VERSION
62
(iii) Bank charges, which accounted for []% and []% of 247’s and Blinds2Go’s total
costs in 2019 respectively, also do not show a clear trend over time.
7.23 Similarly, there is no evidence of any economies of scale in respect of marketing and
advertising costs. Advertising and marketing costs of both Blinds2Go and 247 have increased
broadly in line with the growth in their sales.
7.24 Both 247 and Blinds2Go spend [] on PPC advertising ([]% for 247
106
and []% for
Blinds2Go
107
in 2019), [] of which is allocated to Google Ads or Google Shopping. To
understand why PPC advertising is not subject to any material economies of scale, the
fundamentals of PPC advertising pricing by Google are set out, views regarding the analysis
that was undertaken at Phase 1 are provided, before drawing conclusions.
The fundamentals of Google PPC advertising pricing
7.25 Retailers advertise on Google via Google Ads or Google Shopping campaigns, of which the
former accounts for most of the Parties’ PPC spending. As both Google Ads and Google
Shopping pricing works in a similar way, no separate explanation is provided for the latter.
Whether and where an ad appears on Google’s general search results page is determined by
an ad auction, which is run every time a search term is entered by a prospective consumer.
108
Ads whose “Ad Rank” does not pass the “Ad Rank threshold” (set independently by Google)
are excluded. The remaining ads are ordered based on their respective Ad Rank.
7.26 According to Google, the Ad Rank is determined by six “high level factors”, including the
retailer’s bid, the competitiveness of the auction, as well as the quality of the ad and landing
page (the “auction time ad-quality”).
109
Only the last factor is specific to the retailers – all other
factors are equal across all auction participants. In other words, if retailer A and B were to put
in the same bid for a keyword, only the auction time ad-quality would determine the relative
position of their ads.
7.27 Google is not entirely transparent about how it determines the auction time ad-quality during
an action. However, it does provide an estimate for this metric in the form of the “quality
106
247 RFI response, paragraph 9.6.
107
Hunter Douglas RFI response, paragraph 13.3.
108
See https://support.google.com/google-ads/answer/1704431.
109
See https://support.google.com/google-ads/answer/1722122.
NON-CONFIDENTIAL
VERSION
63
score”.
110
The factors that affect the quality score are in line with those that Google uses to
determine the auction time ad-quality for the purpose of the Ad Rank:
111
(i) The ad's expected click-through rate (CTR): this is based on historical clicks and
impressions and is adjusting for factors such as ad position (e.g. if an ad that always
appears in the top position has the same historical CTR as an ad that always appears
at the bottom of the page, then the latter gets a higher quality score);
(ii) The ad’s relevance to the search: whether the text in the ad and its description matches
the keyword; and
(iii) The quality of the landing page: this is driven by relevance, transparency and
navigability of the page to which the consumer is directed when clicking on the ad.
7.28 None of these factors are affected by a website’s turnover or its total Google PPC spend. The
latter two factors are driven only by best practice drafting of ads and design of landing pages.
The expected CTR is calculated based on historic CTRs – this is done in real time and does
not require a long track-record. Historical CTRs are affected by a variety of factors, including
the usefulness of the ad to the consumers who entered the related search term. It is therefore
relatively easy to achieve a maximum quality score within a short period of time. Indeed, as
noted below, Blinds2Go []. It should also be noted that there are a number of agencies that
help retailers maximise their returns to PPC marketing.
7.29 It is therefore clear that two websites with the same quality score – independently of their
historic performance or overall spend on Google Ads – would pay the same amount to achieve
a given Ad Rank. As the quality score does not depend on historic traffic or overall Google
PPC spend, it follows that larger retailers have no cost advantage against smaller players.
7.30 Another important measure which needs to be understood when considering Google Ads is the
CPC. The CPC, which is closely linked to the Ad Rank, is the amount that websites are charged
when a consumer clicks on an ad on Google’s search results page.
112
7.31 Before the auction takes place, all participants need to set a maximum amount they are willing
to pay for their ad to be shown (the maximum CPC). As Google’s auction works on a second
price basis, the actual CPC is thus typically lower than the maximum CPC set. The winning
110
See https://support.google.com/google-ads/answer/140351?hl=en-GB and
https://ads.google.com/intl/en_uk/home/resources/improve-quality-score/.
111
See https://support.google.com/google-ads/answer/1722122.
112
See https://support.google.com/google-ads/answer/6297.
NON-CONFIDENTIAL
VERSION
64
bidder only pays what is minimally required to clear the Ad Rank thresholds and beat the Ad
Rank of the next best competitor (i.e. the next highest quality adjusted bid).
113
Average CPCs
for a retailer can thus be affected by the following factors:
(i) The choice of keywords on which to bid (i.e. in which auctions to participate);
(ii) The maximum CPC set by the retailer;
(iii) The maximum CPC set by the other auction participants (i.e. the competitiveness of
the auction); and
(iv) The relative auction-time ad quality of the bidders (as approximated by the quality
score).
7.32 This is important because it means that CPCs do not necessarily reflect differences between
auction-time ad quality (as approximated by the quality score), across retailers, or over time.
Retailers can bid on common and highly competitive keywords or opt for niche keywords where
they face limited competition. The competitiveness of auctions for specific key words can also
vary over time as the number or aggressiveness of bidders changes. Finally, retailers can set
higher maximum CPCs to achieve higher Ad Ranks, which increases visibility at a higher cost.
The Phase I Decision’s assessment of PPC economies of scale
7.33 The Phase I Decision noted that “[…] evidence on the Parties’ spend on PPC is consistent with
the existence of economies of scale in PPC spending […]”. In particular, the Decision relied on
the finding that “[…] Blinds2Go had achieved much higher growth in revenue than 247, while
spending slightly less on PPC as a percentage of revenue.”
114
The analysis consisted of a
comparison of the Parties‘ ratio of PPC spending to revenue in one year against growth rates
in the following year. The Phase I Decision observed that Blinds2Go had achieved higher
growth than 247 in 2018 and 2019 despite similar PPC spending to revenue ratios in the
preceding years, and that 247 did not outperformed Blinds2Go in 2017 despite exceeding its
PPC spending to revenue ratio by 50% in 2016.
7.34 The Parties do not believe that any meaningful conclusions can be drawn from such a high-
level analysis for a number of reasons. First, the CMA’s approach, which attempts to explain
growth in one year with PPC spending to revenue ratios in the preceding year, does not reflect
the realities on the digital advertising market. PPC spending has a more immediate return than
suggested by the CMA’s analysis. The gap between the PPC expenditure – incurred when the
113
See https://support.google.com/google-ads/answer/6297.
114
Phase I Decision, paragraph 240.
NON-CONFIDENTIAL
VERSION
65
consumer clicks on an ad – and the sale is on average just a couple of days.
115
A more
appropriate model would need compare sales and PPC spend over a short period of time (less
than one month).
7.35 Second, a simple comparison of revenue growth across the Parties without taking account of
other factors that may have affected their relative performance is simply not informative. While
PPC spending clearly is a factor that may drive growth, it is by no means the only or even the
most important one. PPC advertising can be used to increase website traffic – but it does not
guarantee any increase in sales if the overall product/service offering is not highly competitive.
As shown in Figure 6.3 in Chapter 6 above, whether traffic can be translated into sales growth
depends on a variety of factors such as the quality of the website, competitiveness of prices,
perceived quality of the products on offer, etc. These factors will in turn be reflected in
conversion rates, average order values and the share of returning customers.
7.36 Indeed, 247’s comparatively low sales growth in 2016, highlighted in the Phase I Decision, can
be attributed to a significant fall in user conversion rates, as set out in Figure 6.8 in Chapter 6.
A measure that combines both conversion rates and average order values is the revenue per
user, which is presented in Figure 7.2 below. This clearly shows that [].
Figure 7.2: Revenue per unique user (2014-2019)
[
]
Source: Source: Google Analytics accounts blinds-2go.co.uk, 247blinds.co.uk and web-
blinds.com for number of unique users.; Parties sales data for revenues.; AlixPartners analysis
(Annex 0098).
7.37 Finally, the effectiveness of PPC advertising depends on the way in which it is spent. [].
Contrary to the Phase I Decision there are no economies of scale in PPC advertising
7.38 It is worth setting out a clear definition for economies of scale in PPC advertising. For the
reasons set out above, a definition of economies of scale in PPC advertising based on sales
growth is not appropriate. Sales are driven by a variety of factors that are unrelated to PPC
spend (e.g. competitiveness of prices, quality of website and products and investment in other
types of advertising such as TV).
7.39 A definition based on website traffic is thus preferable. Economies of scale in PPC advertising
would exist if companies who spend more on PPC advertising can attract disproportionately
115
See Annexes 0106 - 0108
NON-CONFIDENTIAL
VERSION
66
more consumers to their website. In other words, economies of scale in PPC advertising exist
if a company that spends twice as much on PPC advertising can generate more than twice the
number of clicks. This is mathematically equivalent to asking whether CPCs decrease with total
PPC spend.
7.40 The description of the Google Ads pricing mechanism above suggest that there are unlikely to
be such economies of scale in PPC advertising. All else equal, the price that a retailer needs
to pay to reach a certain Ad Rank does not depend on its total PPC spend. Retailers can only
gain a cost advantage against their competitors if they increase their quality score – a measure
that does not depend on the overall ad spend or historical traffic volume. This is also
demonstrated by the data submitted in response to Question 19 of the Parties’ respective RFIs.
The average quality score for Blinds2Go across its top 100 key words (by impressions) over
the period 2017-2019 was []. 247’s average quality score across the same key words was
actually [].
116
7.41 To test whether the theoretical findings based on the mechanics of the Google Ads auction are
correct, data was extracted on total PPC costs and clicks from the Parties’ Google Ads
accounts. Figure 7.3 below shows the evolution of total Google Ads (excluding Shopping and
display) PPC spend and CPCs for the blinds-2go.co.uk, 247blinds.co.uk and web-blinds.com
websites.
Figure 7.3: Google Ads (search only) spend and CPC (2011-2019)
[]
Source: Google Ads accounts blinds-2go.co.uk, 247blinds.co.uk and web-blinds.com.
Campaign type: search only; AlixPartners analysis (Annex 0109).
7.42 The figure shows that there are no economies of scale in Google Ads PPC advertising – CPCs
do not decrease with total PPC spend. To the contrary, it suggests that there may be
diseconomies of scale. Blinds2Go’s CPCs [].
7.43 A similar picture emerges with regards to Google Shopping PPC advertising, as presented in
Figure 7.4 below.
Figure 7.4: Google Ads (Shopping) spend and CPC (2011-2019)
[]
116
Annex 0018 and 0058 of the Parties S109 responses. The quality score is not always provided for all key
words. For an explanation see here: https://support.google.com/google-ads/answer/7050591?hl=en-GB.
The average was thus calculated only across those keywords where the quality score was available for
the respective website.
NON-CONFIDENTIAL
VERSION
67
Source: Google Ads accounts blinds-2go.co.uk, 247blinds.co.uk and web-blinds.com.
Campaign type: Shopping; AlixPartners analysis (Annex 0109).
7.44 Blinds2Go’s and 247’s CPCs []. It is also worth noting that Google Shopping will operate on
a largely free basis going forward, as explained in Chapter 6, which could have a significant
impact on the ability of small firms to enter markets even more cheaply.
7.45 A similar analysis can be replicated at a more aggregated level. Figure 7.5 below shows the
Parties’ total PPC spend (across all search engines and social networks) per unique user and
per session.
Figure 7.5: PPC spend per unique users and sessions – Blinds2Go, 247 and Web-Blinds
(2014-2019)
[]
Source: Google Analytics accounts blinds-2go.co.uk, 247blinds.co.uk and web-blinds.com for
number of unique users and sessions; Parties sales data for total PPC advertising spending;
AlixPartners analysis (Annex 0098).
7.46 This shows that []
Successful entry by pure online M2M blinds suppliers provides further evidence that barriers to entry
are low
7.47 As discussed in Chapter 6, there are several examples of recent successful entry in the online
M2M blinds market in the UK. For example:
(i) Blinds Direct, which was established in 2004 and relaunched its website in 2018, grew
rapidly in recent years and overtook 247 as the second largest pure online M2M blinds
provider.
(ii) Swift Direct Blinds also registered significant growth since its launch in 2012, selling an
estimated £12m worth of blinds in 2019.
(iii) Makemyblinds was founded in 2015 and has exhibited rapid growth since then. It has
a similar business model to 247 and has developed good blinds supplier relationships
in China and Poland.
These examples provide proof of the limited costs of entry and the ability for firms to expand
relatively easily.
NON-CONFIDENTIAL
VERSION
68
7.48 There are similar examples of successful entry from other countries which corroborate this
conclusion. Both Blinds2Go and 247 have successfully launched M2M blinds websites in other
European countries that generated profits within a short period of time. The following
paragraphs summarise 247’s experience in the Netherlands and Ireland as well as Blinds2Go’s
more recent expansion into Ireland.
117
247’s entry in Ireland (247blinds.ie)
7.49 247 launched its Irish website (247blinds.ie) on 14 January 2014, selling around £5,300 worth
of blinds within the same month. As 247 had to spend []% of that amount on digital
advertising, the site generated negative profits after marketing costs of £0.11. By February
2014 – just one month after its launch – the site was returning a profit after advertising costs:
sales of around £[] at a gross profit margin of []% sufficed to pay for the digital advertising
costs, which had decreased to just []% of sales. Profits after marketing costs stayed positive
in every month since.
7.50 As illustrated in Figure 7.6 below, 247’s Irish business registered steady growth of []% in the
first two years of operation. This period was followed by rapid expansion in the following two
years, during which the business grew by []% and []% respectively, leading to sales of
more than £[]million in 2018. Given the relatively modest size of the Irish population (just
[]% of that of the UK in 2018
118
), these sales are quite significant. Equivalent per capita sales
in the UK would have amounted to more than £[]million in 2018. Indeed, [].
7.51 Year on year turnover growth for the first three months of 2019 had been []%. However, as
explained further below, the website was shut down towards the end of April 2019 as Hunter
Douglas thought that Blinds2Go would be better placed to compete in the Irish market.
Figure 7.6: 247Blinds.ie revenues and profit margins (2014-2019)
[]
Source: 247 Daily Performance Reports. AlixPartners analysis (Annex 0110).
Notes 1: Sales in 2019 are for January to April only as the website was subsequently closed.
117
247 Blinds also launched a website in France (stores247.fr), which was less successful than its
equivalents in the Netherlands and Ireland. 247 attributes this to insufficient knowledge of the local market,
including language barriers (French for blinds is stores, such that this captures many other retailers) and
different consumer preferences (including a desire to purchase from French companies).
118
Source: UN World Population Prospects, available here:
https://population.un.org/wpp/Download/Standard/Population/.
NON-CONFIDENTIAL
VERSION
69
Notes 2: Figures include sales of products other than blinds provided they were sold through
the website.
7.52 As illustrated in Figure 7.6 above, the website maintained its profitability throughout its period
of operation. By 2018 it was generating more than £[] in gross profits (margin of []%) and
£[] in profits after marketing costs (margin of 26.2%). 247 did not calculate Ireland specific
non-marketing overhead costs, but a net profit margin could be calculated using a revenue
attribution model. This would show that 247blinds.ie’ net profit margin in 2018 stood at []%.
119
247’s entry in the Netherlands (247raamdecoratie.nl)
7.53 247 entered the Dutch market with its 247raamdecoratie.nl website on 30 March 2015. As in
the case of Ireland, the website initially ran at a modest loss due to its marketing expenditure.
In April 2015, the first full month of operation, 247’s digital marketing expenditure accounted for
more than []% of sales, leading to net losses of £[] or []% of sales. However, 247 quickly
turned things around. By May 2015, marketing expenditure had fallen to []% of sales,
allowing 247 to generate gross profits after marketing costs of £[] or []% of sales.
7.54 As shown in Figure 7.7 below, 247’s sales growth trajectory in the Netherlands was very steep
initially, exceeding []% and []% in 2016 and 2017 respectively. Total sales peaked at
£[]million in 2017, which – adjusted for population size – would equate to £[]million worth
of sales in the UK.
120
As in the case of Ireland, 247 did not have to engage in a loss leading
strategy to achieve this rapid expansion. The website generated significant profits after
marketing costs in every year of operation, as explored in more detail below.
7.55 Growth came to a halt in 2018 and then increased year on year in the first 7 months of 2019
before the website was eventually closed in August 2019. 247 believes that this downturn can
be explained by the change to a new marketing strategy, which its agency had developed for
the UK and which turned out not to be appropriate for the Dutch market. As 247’s focus was
on its UK business, the marketing issues were only addressed with some delay, leading to
negative growth for one year. As in the case of Ireland, Hunter Douglas subsequently decided
that the Blinds2Go platform would be better suited for further expansion in the Netherlands. As
a result, the website was rebranded as raamdecoratievantuiss.nl and all traffic from the original
247 website (247raamdecoratie.nl) was redirected to the new site. [].
119
Non-marketing overheads accounted for []% of revenues across the 247 business in 2018. On average
they accounted for []% over the period 2015-2019.
120
The population of the Netherlands was just 25.5% of that of the UK in 2017. Source: UN World Population
Prospects, available here: https://population.un.org/wpp/Download/Standard/Population/.
NON-CONFIDENTIAL
VERSION
70
Figure 7.7: 247raamdecoratie.nl revenues and profit margins (2015-2019)
[]
Source: 247 Daily Performance Reports. AlixPartners analysis (Annex 0110).
Notes 1: Sales in 2019 are for January to August only as the website was subsequently
closed.
Notes 2: Figures include sales of products other than blinds provided they were sold through
the website.
7.56 During its period of operation the Dutch 247 website generated material profits for 247. At its
peak in 2017, 247 sold more than £[]million worth of blinds while generating £[]million in
gross profits (margin of []%) and £[]million in profits after marketing costs (margin of
[]%). As already explained for the Irish example, 247 did not calculate Netherlands specific
non-marketing overhead costs. A net profit margin can thus only be calculated using a revenue
attribution model. This would show that 247raamdecoratie.nl’ net profit margin in 2017 stood at
[]% and increased further to []% in 2018.
121
Blinds2Go’s entry in Ireland (blinds-2go.ie)
7.57 Blinds2Go replaced 247 in the Irish market as Hunter Douglas believed that website and
management team would be better suited to compete. The blinds-2go.ie website was launched
in April 2019 when the 247Blinds.ie website was closed as a result of the transfer of 247’s
overseas businesses to Blinds2Go as part of the 2019 Transaction.
7.58 As all customers who arrived at the 247blinds.ie website were automatically redirected to
blinds-2go.ie, the latter benefited from some free traffic. However, this traffic and the resulting
revenues were insignificant: only [] users were redirected from the 247 website, leading to
[] transactions worth c. £[] in total over the period April 2019 to March 2020.
122
These
figures are insignificant compared to Blinds2Go’s overall sales in Ireland and can thus be safely
ignored for the purpose of the analysis below.
7.59 As shown in Figure 7.8 below, Blinds2Go []. By November, which is generally the peak
month, sales had reached £[[]. [].
123
March 2020 sales stood at £[] despite the
beginning of the economic fall-out resulting from the Covid-19 pandemic.
121
Non-marketing overheads accounted for []% and []% of revenues across the 247 business in 2017
and 2018 respectively.
122
Based on Google Analytics data for blinds-2go.ie website.
123
247blinds.ie’s sales in November 2018 amounted to £[].
NON-CONFIDENTIAL
VERSION
71
Figure 7.8: blinds-2go.ie revenues and profit margins (May 2019 – March 2020)
[]
Source: Blinds2Go management accounts; AlixPartners analysis (Annex 0111). Notes:
Figures include sales of products other than blinds provided they were sold through the
website.
7.60 The fact that [] highlights that fact that PPC advertising pays for itself. Blinds2Go []. To
the contrary, []. Advertising expenditure [].
7.61 []. As explained above, the cost of a Google Ads bid for a key word depends only on the
bids submitted by other bidders as well as their relative quality scores. As Blinds2Go [].
124
7.62 This is also reflected in Blinds2Go overall marketing costs per visitor (measured as number of
sessions) and conversion (number of orders), which are shown in Figure 7.9 below.
Figure 7.9: blinds-2go.ie cost per visitor and conversions (May 2019 – March 2020)
[]
Source: Blinds2Go management accounts. AlixPartners analysis (Annex 0111).
Notes: Figures include sales of products other than blinds provided they were sold through
the website.
7.63 In May 2019, Blinds2Go spent []. These costs []. Blinds2Go’s []. The fact that
Blinds2Go was able to [].
Conclusions regarding international entry examples
7.64 The three examples above clearly demonstrate the absence of any notable barriers to entry for
the online retail supply of M2M blinds. Both 247 and Blinds2Go were able to enter new
geographic markets in which they had no brand recognition or PPC advertising track record.
They were able to scale their operations rapidly without any significant capital injection. To the
contrary, []. Further, as consumers pay for their purchases up-front, while the Parties pay
for the products and PPC advertising only after the sale is made, the websites were likely cash-
flow positive at all time. In other words, the expansion through PPC advertising paid for itself.
7.65 Whilst the Parties could be considered to have benefited their existing operations in the UK in
the sense that they had websites and suppliers, as explained above, these factors could be
easily overcome by new entrants. Website development costs are relatively limited and there
124
[].
NON-CONFIDENTIAL
VERSION
72
is a range of suppliers who could support entry in the UK. For international online retailers or
established omni-channel players these up-front costs should be of limited importance.
Further entry and expansion is timely, likely and of significant scale to replicate the rivalry from 247
7.66 The Parties consider that it is very likely that there will be entry and expansion in the retail of
online M2M blinds. This is because, not only are there strong pull factors that make retail of
M2M online blinds attractive (i.e. the rewards associated with successful entry/expansion
outweigh the limited costs of entry) but also, for a number of key market participants, because
there are important “push” factors (namely existing players have strategic interests in
entering/expanding in the retail of online M2M blinds to defend their existing sales of ready-
made blinds or offline M2M blinds).
7.67 The Parties believe that entry and expansion is likely by a variety of types of firms:
(i) Omni-channel retailers that already sell ready-made blinds or in-store and in-home
M2M blinds;
(ii) Manufacturers of blinds; and
(iii) Pure online M2M blinds retailers.
There will be further entry and expansion by large omni-channel retailers with strong consumer brands
and existing ready-made blinds sales
7.68 The rapid growth in the retail of online M2M blinds has come at the expense of retail sales of
ready-made blinds and offline M2M blinds. Consequently, sizeable players in these markets
that are losing sales to online M2M blinds retailers, have a strong incentive to develop their own
online M2M retail propositions in order to both defend their existing sales as well as for the
opportunities from online M2M blinds sales.
7.69 These incentives have become stronger in recent years as the sales of online M2M blinds have
increased considerably and the consequent negative impact on the ready-made and offline
segments has grown. As explained in Chapter 4, between 2017 and 2019, the retail M2M
online blinds segment increased by £56m (28% growth) at the same time as the offline M2M
blinds segment fell by approximately £16m (a 5% fall). Given this picture, it is of no surprise
that some of the existing large suppliers of ready-made blinds and in-home and in-store blinds
have entered the M2M online retail space in the last couple of years.
7.70 As described in Chapter 4 above, the parties understand that Dunelm has been interested in
the online M2M blinds space for some time, having expressed an interested in purchasing
Blinds2Go in 2015, before Hunter Douglas ultimately purchased the business. Dunelm
ultimately entered the space in March 2018. While Dunelm experienced some issues with its
NON-CONFIDENTIAL
VERSION
73
website, which led to the click-to-order functionality being temporarily suspended in November
2019, the Parties understand that Dunelm has now reinstated a click-to-order online option for
customers of its M2M blinds through a hybrid, virtual consultation route. As the M2M online
blinds segment rapidly increases, at the expense of Dunelm’s ready-made business (online
and offline), Dunelm will be strong incentivised to quickly develop its online M2M blinds
business.
7.71 In Autumn 2019, Next, which is also a sizeable retailer of ready-made blinds, also began
offering in a material way M2M blinds online, building on its strong retail brand.
7.72 While other major existing blinds suppliers, such as the Range, have not yet developed online
M2M blinds businesses, they also have a strong incentive to do so too.
7.73 Dunelm, Next and other existing blinds suppliers with strong retail brands have an advantage
over existing M2M online blinds suppliers because they are able to get online customers from
their websites and through organic search without having to pay for advertising. In addition,
their brands give them an advantage over existing M2M online blinds suppliers in the ease of
attracting online customers through PPC advertising and in the costs of doing so. Consumers
are more likely to click on recognisable brands
125
and so the click through rate for paid search
advertising would be expected to be materially higher. Moreover, as these are brands that
consumers trust, the user conversion rate for these firms would be expected to be higher. This
is important because firms pay for PPC advertising based upon the number of consumers who
click on their ads, which means that a higher conversion rate leads to an effectively lower cost
of online marketing per converted customer. In contrast, the consumer recognition and trust of
Blinds2Go’s and, in particular, 247’s brands is noticeably lower, as shown in Figure 7.10 below.
Figure 7.10: Prompted brand awareness made to measure window furnishings
(December 2018)
125
See for example https://www.digitalthirdcoast.com/blog/does-brand-awareness-affect-ppc-results.
NON-CONFIDENTIAL
VERSION
74
Source: Phase 1 Annex E2.1, Mediacom Brand Tracker Dec 18, page 7.
Note 1: Response to question: “Q: Which of the following brands who sell made to measure
window furnishings have you heard of?”
Note 2: Base: Jan-18: 503, Feb-18: 351, Mar-18: 372, Apr-18: 505, May-18: 362, Jun-18: 358,
Jul-18:359, Aug-18: 355, Sep-18: 506, Oct-18: 506, Nov-18: 353, Dec-18: 350.
7.74 The Covid-19 crisis is turbo charging existing trends of increased online shopping, creating a
step change in the incentives of existing ready-made blinds and offline suppliers to develop
their own M2M blinds retailers. The Parties understand that offline sales of ready-made and
M2M blinds in the last couple of months have been decimated. At the same time, []. This is
likely to continue for an extended period. Even as shops reopen, social distancing rules are
likely to apply until at least the end of 2020 and, in any case, consumers are likely to avoid
going to physical retail stores for some time. Moreover, many market observers are predicting
that the shift to increased online shopping will continue beyond the current crisis.
126
This shift
has a considerable impact on the incentives of executives at existing omni-channel suppliers
to rapidly develop their online M2M blinds plans – the opportunity from higher revenues and
profits from online M2M blinds is increased and the necessity to diversify away from offline
sales heightened (and note that the vast reductions in offline sales due to Covid-19 will dwarf
the previous reductions in offline M2M blinds sales due to consumer switching).
Manufacturers have an incentive to enter into online retail of M2M blinds
7.75 UK blinds retailers (both ready-made and M2M) have typically been supplied by UK blinds
manufacturers, such as []. However, the move to online has threatened their businesses as
online retailers have been more willing to source from multiple suppliers, including those outside
the UK. Indeed, [].
7.76 The threat of being substituted by international suppliers, has provided UK manufacturers with
a strong incentive to enter into retail or expand their existing retail operations. Indeed, both
Interior Goods Direct and Swift Direct Blinds are examples of vertically integrated online
retailers of M2M blinds.
7.77 The Parties expect that other UK manufacturers are likely to enter in the near future, either
through launching their own retail operations or through purchasing an existing retailer.
7.78 Vertically integrated providers that are managed as single businesses have an ability to be
particularly commercially aggressive through the elimination of double marginalisation (i.e.
focusing on maximising their total profits across both their retailing and manufacturing
activities).
126
See the quote from Sharon White, John Lewis CEO, at paragraph 4.18 above.
NON-CONFIDENTIAL
VERSION
75
7.79 International suppliers may also have an incentive to go direct to consumer. Nien Made
purchased Veneta, the number 1 retailer of online M2M blinds in the Netherlands. Somfy, a
manufacturer of motorised blinds, has started to offer blinds online through
poweredblinds.co.uk, which is advertised on the Somfy website.
Entry and expansion by pure online M2M blinds suppliers is likely
7.80 As discussed above, there is recent evidence of entry and growth in other countries. Similarly,
in the UK, the recent growth in the retail of online M2M blinds has attracted entrants and led to
expansion by existing firms. For example, Interior Goods Direct has grown rapidly since 2017
displacing 247 as the number 2 firm in the market and Make My Blinds has increased sales
significantly in recent years.
7.81 This shows that there are no material barriers to entry and expansion by pure online M2M blinds
suppliers. Consequently, while the Parties understandably have no knowledge of the specific
plans of rival online M2M blinds suppliers, as there are profitable opportunities for entry, which
have only increased due to the Covid-19 crisis, there is every reason to believe that there will
be further entry and expansion.
Further entry using marketplaces is likely
7.82 As described above, Amazon already offers a low-cost route to market for retailers. For a
retailer using Amazon as its only route to market, there is no need to build a website and the
retailer can benefit from Amazon’s delivery infrastructure, Amazon’s no quibble customer
service, and as many consumers search for products on Amazon. Moreover, Amazon’s recent
inclusion of a customisation option, has improved the user experience, providing a more
comparable ordering process to that on online M2M blind retailers’ own websites. The
effectiveness of the improved Amazon experience is shown through the Swift Direct Blinds
example above and [].
7.83 Firms can either use Amazon as their only route to market (avoiding considerable costs, as
mentioned above) or as an additional route to market, as Swift Direct Blinds has done and [].
In either case, using Amazon will increase the expected profitability of entry, thereby increasing
the incentives to enter.
7.84 Other marketplaces also offer attractive opportunities for online M2M blinds retailers to reach
consumers at limited costs. eBay is already an option and one that could be improved, similar
to Amazon, by including a customisation option.
7.85 There is the possibility that other marketplaces may also develop their offering to become an
attractive option for the online retail of M2M blinds. For example, Wayfair, one of the world’s
largest e-commerce providers of home furnishings, and already a retailer of ready-made blinds,
NON-CONFIDENTIAL
VERSION
76
has recently emphasised that it is aggressively investing.
127
As emphasised above, the recent
market developments, in particular the Covid-19 crisis, significantly increases the incentives to
make such investments.
127
https://www.bloomberg.com/news/articles/2020-05-05/wayfair-soars-to-record-after-retailer-s-positive-
margin-update
NON-CONFIDENTIAL
VERSION
77
8. Neither the 2013 nor 2019 Transactions lead to an SLC
Introduction
8.1 In the Issues Statement, the CMA set out that it is exploring whether there was important rivalry
between 247 and Blinds2Go which has been lost, and whether as a consequence of this loss
of rivalry it would be profitable for the merged business to appreciably worsen its offer as
regards price, quality, range or innovation to such an extent that an SLC can be identified.
8.2 As emphasised in Chapter 3, this case differs from most mergers as a material period of time
has elapsed post-completion for each Transaction. It is highly relevant that there is no evidence
over this time of any loss of rivalry between 247 and Blinds2Go leading to price, quality, range
or innovation being worsened and any assessment must take this into account. If a finding of
an SLC were to be made, it would need to clearly demonstrate, on the basis of the evidence,
that price, quality, range or innovation had been worsened over that period or that this was a
likely result of the 2019 Transaction. In respect of the former, the evidence indicates entirely
the opposite.
8.3 Given the CMA’s theory of harm, this Chapter focuses on whether, as a result of the
Transactions, it would be profitable for Hunter Douglas if Blinds2Go and 247 increased their
prices (or otherwise worsened their offer).
The market context to the CMA’s theory of harm
8.4 Blinds2Go is the main competitive force in the online retail of M2M blinds in the United Kingdom.
All other players look to what Blinds2Go is doing and often copy its website, designs, etc. That
was the case in 2016 and in 2019 and is unaffected by either Transaction.
8.5 As regards the issue of whether either transaction gives rise to an SLC, the key issues are
whether 247 imposes a material constraint on Blinds2Go and whether the Transactions has a
substantial impact on that constraint. As set out in Chapter 5, [] reflects the fact that any
rivalry it faces from 247 has not been a factor affecting its pricing decisions.
8.6 247 is not of any unique or special importance. It is a much smaller rival than Blinds2Go, with
the CMA estimating its market share to be 5-10%. In addition, the CMA estimates that 247’s
share has fallen from 10-20% in 2016 to 5-10% in 2019, whereas Interior Goods Direct’s market
share increased from 5-10% to 10-20%. Accordingly, 247 is now the third largest competitor,
or the fourth if eBay were to be treated as a single rival. In addition, there are many other pure
online and omni-channel M2M rivals who are growing their online sales.
8.7 Moreover, it is clear that online M2M blind retailers face material “out of market” constraints.
As discussed in Chapter 5, Blinds2Go’s strategy has focused on competing directly with ready-
NON-CONFIDENTIAL
VERSION
78
made blinds and offline M2M spend. Its growth has not been achieved by winning sales off
other online M2M retailers and it has not focused on competing head on with other online M2M
blinds suppliers. Indeed, the BDRC Survey asked both Parties’ customers directly what other
window coverings customers would consider purchasing, and almost 50% of both said they
would consider alternatives other than M2M blinds, including ready-made products and shutters
and curtains. That figure may in fact understate the extent to which alternatives are considered
since customers taking the survey may take a broad view of what constitutes an M2M blind.
Many retailers of ready-made blinds cut down those products to customer specifications and
sell their products as M2M.
8.8 It should also be noted that in early May 2020 there were over 440 different ready-made size
combinations available online among the top ready-made blinds retailers and these are priced
very competitively
128
. This is consistent with the evidence that []% of Blinds2Go’s sales and
up to []% 247’s sales have equivalent ready-made products. If consumers were only
interested in precise blind dimensions, then ready-made products are clearly substitutable for
many M2M sales. Consumers can also modify ready-made blinds to suit their needs, and blinds
can be fitted outside the recess with tolerances for width and drop dimensions in such a way
that most windows in a standard sized residence can be fitted with a ready-made blind.
Prominence in search does not mean that 247 is a particularly close competitor
8.9 In the Phase 1 Decision, “The CMA considered that the frequency and prominence of the
Parties suggested that their competitive significance is greater than their shares of supply
migh[t] suggest.”
129
The Parties strongly disagree with this statement for a number of reasons.
8.10 First, the BDRC found that some three quarters of the Parties’ consumers spend over an hour
researching their purchase (and half spend longer than a few hours). Google Trends results
indicate that users searching for “blinds” are also searching in the same search session for
Dunelm, Ikea, Argos, and Hillarys – with 247 not even in the top 20 most searched terms. The
BDRC Survey results also indicate that 247 customers were less aware of the 247 brand prior
to purchase and were consequently less likely to search for the brand or go directly to the
website (only []%).
8.11 Second, it is also important to appreciate that omni-channel retailers (such as Dunelm) have a
natural competitive advantage in that consumers will frequently visit their websites to research
and purchase a wide range of products and they can run large scale email marketing
campaigns to their millions of registered customers. For example, Next’s website combines
128
See further Chapter 6 and the response to question 3 of the response to the section 109 notices of 20
April 2020.
129
Paragraph 161 of the Phase 1 Decision.
NON-CONFIDENTIAL
VERSION
79
online shopping, pick up at stores, and fitting by Next personnel. This is a powerful offer that
the rest of the industry needs to take very seriously.
8.12 Third, whilst paid search is an important source of traffic to the Parties’ websites, this should
not be exaggerated or over interpreted. Whilst paid search makes up [], consumers use a
mixture of organic search results, email, display and referral sources to find the Parties’ e-
commerce websites. In addition, rivals can use organic search, social media and other referral
or display referral sources. Importantly, Google has also recently announced that search
results on the Google Shopping tab will consist primarily of free listings by the end of the year,
which is a major policy change that is expected to have a big impact for online retailers.
8.13 Fourth, not only is consumer research not limited to online generic word traffic search (as
highlighted by Google Trends analysis), 247’s Google prominence does not translate into sales
as otherwise its share would not be 5-10% and declining. This is because consumers shop
around. Indeed, for both Parties their per-session conversion rates are below [].
8.14 Moreover, Blinds2Go’s user conversion rate is []% higher than 247’s ([]% versus []%)
and 247’s user conversion rate []. This is consistent with 247’s declining market share, rather
than suggesting that 247 is somehow a close competitor to Blinds2Go.
The market for M2M online blinds is highly competitive and 247 is not an important competitor to
Blinds2Go
8.15 At paragraph 183 of the Phase 1 Decision, the CMA indicates that Interior Goods Direct is
considered to be the main constraint on the Parties and that other smaller online M2M blind
retailers constrain the Parties to a less significant degree. Further, at paragraph 198, the CMA
disregards the competitive constraint exercised by omni-channel retailers on the basis that they
are more distant competitors because of differences in business model and commercial focus,
with less focus on their M2M blinds products, and different functionality of webstores.
8.16 However, as set out in Chapter 7, this is not a market where economies of scale are material
or smaller competitors cannot compete, particularly omni-channel retailers.
8.17 Moreover, evidence from the BDRC Survey suggests that the Parties’ customers compare
products/prices across many websites. For Blinds2Go’s customers, 247 is []. For 247’s
customers, []% of its consumers make comparisons against Blinds2Go, but Blinds Direct and
Dunelm [] (at []% and []% respectively). Over []% of 247’s customers made
comparisons at Blinds-4UK ([]%), John Lewis ([]%), Amazon market place ([]%), eBay
([]%) and Next ([]%). In addition, the BDRC Survey also indicated that []% of 247’s
customers visited physical stores before placing orders and this was also the case for []% of
Blinds2Go’s customers.
NON-CONFIDENTIAL
VERSION
80
8.18 In the event that Blinds2Go were to stop selling blinds, the BDRC Survey indicates []. This
is further supporting evidence that 247 is not an important competitor to Blinds2Go.
8.19 An analogous question was also asked of 247 customers. It should be noted that one would
expect the responses to such questions to suggest relatively high diversion from small
competitors to market leaders, simply reflecting relative market shares. If 247 were to have a
[]% share of online M2M blinds and Blinds2Go []% and out of market constraints are very
limited, then based on relative market shares one would expect diversion of about []). In fact,
[].
8.20 It should also be noted that:
(i) These responses are likely to understate in-store diversion as it was commissioned
when various stores were closed;
(ii) Interior Goods Direct has proved very successful at winning 247’s customers and has
displaced 247 as the second largest competitor;
(iii) Dunelm only entered the online sector in March 2018 (and ceased operating
temporarily online in November 2019 due to a website change and has only recently
re-opened), and Next only materially entered the online sector in Autumn 2019.
8.21 In addition Hunter Douglas does not have a financial interest in 247 ceding sales or profits to
Blinds2Go, reflecting that [].
There is no evidence that 247 is an important source of price competition to Blinds2Go
8.22 The CMA has indicated that it intends to seek the views of third parties – i.e. competitors – as
to the closeness of competition between the Parties.
130
Competition authorities typically take
competitors’ view with a “pinch of salt”, since rivals should welcome mergers that lead to higher
prices and less competition for key inputs and this tends to influence the views they express.
In the present case, it would seem appropriate to use a larger measure than a pinch since
various market participants have already expressed an interest in acquiring 247 and a forced
divestment of 247 would provide a natural spring board to any entry plans (potentially at a low
cost). Accordingly, the Parties would welcome confirmation that the CMA will use its formal
information gathering powers under section 109 to obtain information (including as regards
management presentations, internal documents and entry plans), and focus on pre-existing
documents that have been created in the ordinary course of business. It is also far from obvious
130
Paragraph 31(c) of the Issues Statement.
NON-CONFIDENTIAL
VERSION
81
how competitors can assess the extent to which the Parties win/lose business from one
another.
8.23 The CMA also indicates in assessing the closeness of competition between the Parties that it
intends to review internal documents.
131
As noted above, all other online M2M players look to
what Blinds2Go is doing and often copy its website, designs, etc. and this is also the case for
247. 247 monitors Blinds2Go’s prices and those of Interior Goods Direct reflecting that this
enables it to cover market trends as a small competitor. This monitoring does not mean that
the Parties closely compete for customers in terms of their actual competitive conduct, and
specifically how they actually compete on price, which the BDRC survey highlights as the most
important choice factor for the Parties’ consumers.
8.24 Blinds2Go routinely []. As explained at the virtual site visit []. In Autumn 2019, [].
Blinds2Go also [].
8.25 It is striking that despite its higher quality, Blinds2Go’s prices are not generally higher than its
competitors and in certain categories its prices are lower. In addition, as set out in Chapter 6,
an analysis of weekly price changes does not suggest that Blinds2Go and 247 are particularly
closely tracking each other’s prices (particularly in 2018 and 2019). Accordingly, there is no
evidence that 247 is a close competitor on price to Blinds2Go.
Barriers to entry and expansion
8.26 As set out in Chapter 7, the Parties consider there are a number of credible sources of new
entry. It is also crystal clear that small-scale entry is viable given that there already a large
number of pure online blinds retailers and major omni-channel retailers.
8.27 The substantive SLC question is, if Blinds2Go were to face an appreciable reduction in rivalry
as a result of 247 ceasing to compete independently, would sufficient aggregate expansion
from rivals be expected in a timely manner (i.e. within two years) so as to replicate the
competition lost from 247. This issue should be judged in the context that 247 only has a
market share of 5-10%, online M2M blind sales of c£[] million last year, and its market share
has fallen consistently over time.
8.28 Before addressing any hypothetical barriers to expansion, it is important to be clear that there
has already been substantial entry/expansion from a range of rivals in recent years which is
substantially in excess of 247’s 5-10% market share. In particular:
131
Paragraph 31(c) of the Issues Statement.
NON-CONFIDENTIAL
VERSION
82
(i) Interior Goods Direct has increased its market share from 5-10% in 2016 to 10-20%
now;
(ii) Makemyblinds entered the market in only 2015 and has built up a multi-million pound
business from scratch. The dismissal in the Phase I Decision of this rival on the basis
of a passing comment at the issues meeting that the Parties “do not know much about
them” is unreasonable,
132
particularly since this comment was made in the context of
describing a privately owned unlisted business and then giving detail on the quality of
this business’ website, products and its supplier relationships (it has developed
relationships with blinds suppliers in China and Poland).
(iii) Dunelm entered the on-line M2M sector in only March 2018, and the BDRC Survey of
the Parties’ customers confirms that it is already a major competitor;
(iv) Next only materially entered the online M2M sector in Autumn 2019;
(v) Amazon has been an attractive route to market for some retailers and has recently
made significant improvements to its website that allow for a more user-friendly
customisation of products.
133
In January 2020, []. Contrary to the views expressed
by the CMA in the Phase 1 decision,
134
the commission charged by Amazon does not
place retailers who choose this route at a competitive disadvantage vis-à-vis
incumbents. Amazon’s commission, which currently stands at 15% (not 20%, and this
is negotiable
135
) is in line with the marketing costs incurred by the Parties, covers not
only website development and hosting costs, but also payment processing fees as well
as digital advertising costs. Moreover, almost 90% of UK shoppers use Amazon, and
70% of these use the site as their first point of search, indicating the marketing value
of being on Amazon.
136
8.29 Chapter 7 also provides multiple examples of the Parties entering successfully and at low cost
in a number of other European countries.
8.30 In short, the proof of the pudding is that barriers to both entry and expansion are low, and it
would take minimal entry and expansion to offset the loss of 247.
132
Paragraph 177 of the Phase 1 Decision.
133
247 RFI response, paragraph 8.4 and Hunter Douglas RFI response paragraph 9.5.
134
Paragraphs 201-202 of the Phase 1 Decision.
135
See 247 RFI response, paragraph 8.4.
136
See https://www.theguardian.com/business/2019/mar/07/almost-90-of-uk-shoppers-use-amazon-
research-reveals.
NON-CONFIDENTIAL
VERSION
83
8.31 Assessment of barriers to entry and expansion tends naturally to focus on the costs and risks
of entry, but it is important to highlight that in this market there are other “pull” and “push” factors
which substantially increase rivals’ incentives to expand. In particular, online M2M blinds are
growing strongly, which is a strong pull factor. Moreover, sizeable omni-channel players are
losing sales to online M2M blinds retailers, and therefore have a strong incentive to develop
their own online M2M retail propositions in order to both defend their existing sales as well as
to profit from the opportunities arising from online M2M blinds sales. Similarly, Amazon and
eBay have strong incentives to develop their businesses in this area to exploit their very large
customer bases. The Covid-19 crisis is also turbo charging existing trends of increasing online
shopping.
8.32 The only hypothetical barrier to expansion identified in the Phase 1 Decision is the cost of pay-
per-click (PPC) advertising, which by definition is paid on a pence per click basis. How this
operates is described in detail in Chapter 7, but this is self-evidently not a simple fixed cost.
8.33 Even if PPC advertising were a pure fixed cost (which is not the case), the Phase 1 Decision
does not explain why this represent a material barrier to expansion:
(i) 247’s PPC costs were only of []% of its UK sales in 2019;
(ii) Large omni-channel retailers can leverage their brands and large registered user base
to win customers, with a large number of consumers searching for and visiting their
websites;
(iii) Moreover, while this is a viable route to market, it is by no means the only way to reach
customers online. There are other channels such as Google Shopping (which Google
states will largely be free listings during the course of 2020), organic search, or social
media.
8.34 Turning to the Phase 1 Decision’s analysis of economies of scale in PPC advertising,
137
this is
based on an observation that Blinds2Go had achieved higher growth than 247 in 2018 and
2019 despite similar PPC spending to revenue ratios in the preceding years, and that 247 did
not outperform Blinds2Go in 2017 despite exceeding its PPC spending to revenue ratio by
[]% in 2016.
8.35 Unfortunately, this proposition was not put to the Parties until the Phase 1 Decision was
released. The analysis is, however, flawed as 247’s comparatively low sales growth in 2016
can be attributed to []. 247’s main problem is not attracting consumers to its website, but
persuading them to buy as discussed above and in Chapter 5. [].
137
CMA Phase 1 decision, paragraph 240.
NON-CONFIDENTIAL
VERSION
84
8.36 The fact that there are no economies of scale in Google Ads PPC advertising is clear by
comparing 247 and Blinds2Go’s costs per click (CPC) and over time, with Blinds2Go being
many times the size of 247. As set out in Chapter 7, this analysis suggests that there may in
fact be diseconomies of scale. Blinds2Go’s CPC [].
8.37 Analysis also shows that 247, while spending significantly less in absolute terms than
Blinds2Go on PPC advertising, managed to attract more unique users to its website for every
£1 in PPC spend (with the exception of 2015 and 2016 where, for reasons explained above,
247 []. PPC spending for session, on the other hand, was very similar for both Parties (again
except for 2015 and 2016). This can be explained by the fact that Blinds2Go’s users engage
in more sessions per year than 247’s.
8.38 To sum up, the Parties do not consider that there are any material barriers to expansion. Firms
that are likely to expand are large, have prominent brands and have an incentive to complete
strongly to protect their existing revenues. The evidence from history is that entry and
expansion can occur very quickly (well within two years), likely given the factors outlined above,
and sufficient (these large firms will provide a considerable constraint on Blinds2Go, which can
be expected to be greater than any constraint 247 provides).
8.39 For all these reasons, the Parties do not consider there to be any prospect on any basis of
either transaction giving rise to a lessening of competition, still less to a SLC.
