Get documents about "Telephone survey of advertisers
Document Sample
APPENDIX 5.4
(referred to in paragraphs 2.22 to 2.26, 2.96, 2.97, 2.99, 2.107, 2.109, 5.110, 5.115, 5.150 to 5.155)
Telephone survey of advertisers
Introduction and summary
1. The CC commissioned ORC International (ORC) to conduct a questionnaire survey of business
advertisers that use the local press in the three broad areas of interest in this inquiry—in and around
Derby, Northampton and Peterborough. The objective of the survey, which represented the views of
higher-spending advertisers in the titles owned by Trinity Mirror and Johnston in the areas, was to
identify the nature of these advertisers and to understand their behaviour with respect to advertising
spend in local newspapers and other media.
2. The typical respondent—a higher-spending advertiser in the titles owned by Trinity Mirror and
Johnston—reported using local newspapers as its principal advertising medium. Other printed media
were relatively less important for these advertisers, and local radio appeared to be the most popular non-
print medium. Half of the respondents, when asked to compare free weeklies and paid-for weeklies, saw
them as equally effective for advertising purposes compared with a third who thought they were not, and
the balance of opinion was that advertising-only publications (as distinct from free weekly newspapers)
were not equally effective. When these advertisers negotiated they tended to obtain discounts off the rate
card, and those that had larger advertising budgets tended to obtain higher discounts. The majority said
that they would respond to a 10 per cent increase in rates for advertising in all local weekly papers, for
example by switching spend, cutting back or negotiating.
3. However, 30 per cent said that they would do nothing in the face of a 10 per cent rate increase.
These tended to be advertisers that received rate-card discounts of less than 25 per cent. Moreover, 8 per
cent of these advertisers said that they did not get discounts and about two-fifths said that they did not
know what discount they obtained. (This may be because they focused on the price they currently paid
when negotiating for the next campaign or annual plan, rather than how the price compared with the rate
card.) Even so, the proportion of different types of advertiser that reported discounts above or below
25 per cent was indicative of those who fared better or worse. Recruitment agencies, local government,
educational institutions, tradesmen and other local businesses appear to fare relatively worse based on
this analysis.
4. Contrary to this view, Johnston observed that many advertisers recording no or low discounts in
fact received very low prices (for example, estate agents) and that discounts were affected by a host of
other factors besides category of advertiser (for example, size, frequency, volume commitment etc) so
that comparisons of discount levels did not mean that customers claiming to receive low discounts
typically had no competitive alternatives open to them.
5. With respect to the observation that those who would do nothing in response to a price increase
tended to receive discounts below 25 per cent, Johnston commented that there was a relatively small
difference between the percentage of those that said they would ‘do nothing’ and received discounts of
under 25 per cent (40 per cent), and the percentage of those advertisers who indicated that they would
respond to a price increase and received discounts under 25 per cent (28 per cent).
6. Commenting on the result that 8 per cent reported not getting discounts, Johnston noted that the
majority of those reporting that they received no discount or did not know the discount would have paid
net prices and regarded the rate card as irrelevant; moreover such advertisers would typically have
received net prices which represented very substantial discounts off the rate card. Johnston said that its
own evidence provided to the CC indicated that discounts were considerably more extensive than those
reported in the survey results.
7. Where the CC analysis identified certain advertisers as faring relatively worse in terms of
discounts, Johnston noted that the bulk of the advertising of recruitment agencies, local government and
education institutions cumulatively fell into the recruitment and public notices categories, of which the
Trinity Mirror titles which it proposed to acquire carried only a very small amount of advertising. It
concluded that there would be no substantial lessening of competition in relation to these categories as a
180
result of the proposed transfer. Moreover, it noted that a lower level of discount could be a reflection of a
lower level of volume or frequency of advertising from smaller businesses and consequently the lower
level of discount was not discriminatory in nature.
8. With respect to the survey questions concerning equal effectiveness of alternative print media,
Johnston said that these questions suffered from a failure to define the categories of publication and held
that they were ambiguous; arguing that ‘equally effective’ excluded more effective as well as less
effective alternatives. Johnston said that the ambiguity was revealed by an analysis of the responses, in
which recruitment advertisers indicated that paid-for titles were not equally effective as free titles (and
that they only used paid-for titles to any material degree), and that this indicated that the question was
open to the interpretation by all advertisers that alternative media were more effective rather than less
effective.
9. In commenting on the CC analysis of the survey, Trinity Mirror expressed concern that the results
of the survey might be used to draw conclusions regarding the overall structure of the regional press—
particularly as it related to the market segmentation of newspaper products and to the relevance of other
advertising products. Trinity Mirror thought that the aggregation of the responses to the survey, which
covered a number of geographic markets each with its own mixture of products, treated each market as if
it were uniform, and that this inevitably led to problems with interpretation. In consequence, Trinity
Mirror did not believe that the effectiveness of alternative products to free newspapers was being fairly
or appropriately considered. Neither did it believe that the level of discounts achieved could be used as a
measure of the success of individual advertisers or as an indication of which advertisers might or might
not be more vulnerable to a price increase. It thought that such analysis was flawed because of the way in
which different categories of advertising were priced and sold within titles and by overall levels of
discounts offered against rate cards.
10. From Trinity Mirror’s perspective, the key issue with respect to advertisers’ reactions to a 10 per
cent price increase in advertising rates was that sufficient numbers of advertisers would react to an
increase to make such a rise uneconomic. Although this was noted in a number of places within the CC
analysis, Trinity Mirror felt that this was done without any balance compared with the emphasis placed
on the 30 per cent who would do nothing.
11. Trinity Mirror said that each market must be considered on its own merits in terms of the
products that existed currently and from the perspective of new launches. Given its concerns stated above
regarding the aggregation of markets within the survey area, it felt that any general conclusions needed to
have a caveat: that the survey might or might not reflect the situation in the individual markets within the
survey, let alone in other markets. Only to the extent that the products available in other markets were
equivalent, it argued, could the survey be regarded as potentially relevant to wider market structures.
12. The remainder of this appendix is organized in three sections. The section that follows contains a
description and analysis of the survey, its limitations and the main conclusions that can be drawn from it.
The comments of the parties on the various survey results are also included within this section. The next
section is the report that ORC produced on its work. The final section lists the questions that were asked.
The CC analysis of the survey and its main results
Fieldwork and sample
13. We commissioned ORC to conduct a telephone survey. We asked the parties to send us lists of
the top 30 advertisers by spend in newspaper titles covering the three areas of interest. Table 1 lists the
33 newspapers from which advertisers were drawn and shows that they included the eight Trinity Mirror
titles, which are the subject of our inquiry, and 25 Johnston titles. Where the advertiser information that
was given to us related to more than one period for a particular title (for example, the 12 months of 2000
and the first six months of 2001) the details were consolidated and duplicate contacts were removed.
Similarly other duplicate contacts, for example where an advertiser had used more than one title, were
removed. Based on this information we transmitted to ORC a consolidated telephone contact list. It later
emerged that the contact list erroneously omitted 4 per cent of the potential respondents and 4 per cent of
the list were unobtainable telephone numbers. However, as will be explained next, the impact on the
overall sample is modest.
181
14. After the fieldwork had been completed, Trinity Mirror notified us that by mistake its contact list
had excluded advertisers in the property and motors supplements that are provided as part of certain of
the titles published by its Northampton and Peterborough centres. The effect of this was to include some
advertisers falling outside the top 30 for certain titles, but in the top 40 or 50, and to exclude some
higher-spending property and motor advertisers. Since some of these advertisers were listed elsewhere,
the net effect was the exclusion of 18 of the Trinity Mirror advertisers, 11 of which were estate agents. In
addition, Johnston drew to our attention that, as well as advertisers in the titles in Table 1, details for the
Northants Citizen and the Northampton Mercury had been provided to us but we had omitted them from
the survey. As a result, eight estate agents, three motor dealers, nine retail businesses using display
advertising and one business user of other classified advertising were omitted that might otherwise have
contributed to the survey in the Northampton area. The effect of these combined omissions is equivalent
to 4 per cent of the consolidated list of 914 contacts given to ORC. Although this will reduce the
potential accuracy of some results, namely inferences about estate agents and motor dealers and top
advertisers in the Northampton and Peterborough areas, it will improve the accuracy of inferences about
medium-spending advertisers in these areas. When ORC used the contact telephone numbers it found
that 4 per cent were unobtainable and a further 10 per cent evoked no response during the period of the
fieldwork. Effectively this reduces the sampling frame to 783 contacts. However, a response rate of over
60 per cent of these contacts was actually achieved.
TABLE 1 Newspaper titles covered by the survey
Title Publisher Area* Type
Alfreton and Ripley Echo Johnston Derby (2) FW
Alfreton Chronicle Advertiser Johnston Derby (2) FW
Banbury Citizen Johnston Northampton (3) FW
Banbury Guardian Johnston Northampton (3) PW
Belper News Johnston Derby (1) PW
Buckingham Advertiser Johnston Northampton (3) PW
Daventry Express Johnston Northampton (1) PW
Derby Trader Trinity Mirror Derby (1) FW
Derbyshire Times Johnston Derby (2) PW
Eastwood Advertiser Johnston Derby (2) PW
Fenland Citizen Johnston Peterborough (1) FW
Harborough Herald and Post Trinity Mirror Northampton (4) FW
Harborough Mail Johnston Northampton (4) PW
The Trader Trinity Mirror Derby (2) FW
Ilkeston Advertiser Johnston Derby (2) PW
Lincs Free Press Johnston Peterborough (1+2) PW
Milton Keynes Citizen Johnston Northampton (3) FW
Northampton Chronicle and Echo Johnston Northampton (1) PE
Northampton Herald and Post Trinity Mirror Northampton (1) FW
Northamptonshire Evening Telegraph Johnston Northampton (2) PE
Peterborough Citizen Johnston Peterborough (1) FW
Peterborough Evening Telegraph Johnston Peterborough (1) PE
Peterborough Herald and Post Trinity Mirror Peterborough (1) FW
Ripley and Heanor News Johnston Derby (2) PW
Rugby Advertiser Johnston Northampton (1) PW
Rugby Review Johnston Northampton (1) FW
Spalding Guardian Johnston Peterborough (1+2) PW
Stamford Citizen Johnston Peterborough FW
Stamford Herald and Post Trinity Mirror Peterborough (1+2) FW
Stamford Mercury Johnston Peterborough (1+2) & PW
Northampton (2)
Sunday Citizen Johnston Northampton (3) FS
Towcester/Brackley Post Trinity Mirror Northampton (3) FW
Wellingborough/Kettering/Corby Herald and Post Trinity Mirror Northampton (2) FW
Source: Johnston and CC.
*The areas in this table correspond to those defined in Tables 5.2 to 5.9 of Chapter 5.
Note: FS =Free Sunday; FW = Free Weekly; PE = Paid-for Evening; PW = Paid-for Weekly.
15. A total of 476 telephone interviews of approximately 15 minutes’ duration were conducted
between 2 and 18 January 2002. Over four-fifths of respondents were responsible for advertising
decisions, and the remainder were knowledgeable or influential in these decisions. Target numbers to be
contacted were set by geographical area in order to have similar representation in the three main areas of
interest.
182
16. Table 2 shows the range of businesses that were surveyed. Despite the exclusion of certain estate
agents and motor dealers as mentioned previously, these were still among the three main categories of
local business included in the survey, along with retailers. National and regional advertising agencies
were also represented. In Table 2 motor repairers have been grouped with other trades.
TABLE 2 Types of business advertiser surveyed
Count %
Estate agent 94 20
Retail store or supermarket 88 18
Motor dealer 62 13
Trades, including motor repairs 37 8
Leisure, consumer and financial services 33 7
Local government 19 4
Recruitment agency 13 3
Education 12 3
Other business categories* 41 9
National agency 48 10
Regional/local agency 29 6
Total 476 100
Source: CC calculations based on ORC survey.
*Other categories of business advertiser that individually represent less than 3 per cent.
17. Respondents were asked to confirm the county or town localities that they aimed to cover with
their advertising. Table 3 shows the areas targeted by businesses that advertised only in a local area or
region, by those that also advertised nationally, and by advertising agencies. The table shows that
respondents represented a fairly even spread of advertisers over the different areas. Table 3 also shows
that it is not possible unambiguously to associate advertisers with particular regions.
18. Johnston commented that many organizations would be regional or national in their areas of
operation and would therefore have indicated that they would seek to advertise in more than one local
area. However, this did not mean that the advertising which they sought to place was regional in nature
as the advertisements would be specific to a particular local area. Also, many of its titles whose
advertisers were surveyed were not primarily focused on any of the areas referred to in Table 3, and this
helped to explain the relatively high number of respondents referring to ‘other areas’.
TABLE 3 Areas targeted by different types of advertiser
per cent
Advertises own business Advertising agency All
Locally/
Locally/ regionally and Regional or
Local area regionally nationally National local
Peterborough town, Oundle and West
suburbs of Peterborough 18 32 33 52 24
Stamford and Spalding Rural (west & north
of Peterborough) 20 35 33 52 26
Northampton Town and West suburbs 21 32 44 59 28
Kettering, Wellingborough, Rushden and
Corby 17 30 35 59 24
Brackley and Towcester 16 20 27 48 20
Market Harborough 14 24 27 52 19
Derby Town and Belper 20 27 40 55 25
Alfreton, South Normanton, Ripley Ilkeston
and Eastwood 24 26 25 62 27
Other areas 21 23 31 28 23
Don’t know 0 0 4 0 1
Source: CC calculations based on ORC survey.
19. ORC asked advertisers about their mix between display and classified advertisements. Most
advertisers who used a mix reported mainly using display advertising, but it is probable that respondents’
perceptions of display advertising included advertisements that newspaper specialists would regard as
classified. We discuss this point again below.
183
Analysis of results
Use of and switching between communication media
20. The extent to which advertisers make use of local newspapers and other media was revealed by
unprompted answers to a question that asked them to list the different ways that they might have
deployed their advertising expenditure over the previous 12 months. Local radio was one of the more
popular alternatives to local newspapers but the position varied widely by category of advertiser. About
half of the advertising agencies and educational institutions reported using local radio, whereas only one
in ten estate agents and tradesmen (including motor repairers) did so. Table 4 summarizes the results.
Some response categories, such as classified directories and advertising-only publications, are not
designed to be mutually exclusive but to codify respondents’ unprompted answers.
21. Johnston stated that Table 4 was misleading and substantially understated the extent to which
local advertisers used advertising-only publications and other media for two reasons. First, it said that
advertising-only publications were difficult to categorize and, in many cases, to distinguish from free
newspapers. It said that, according to the World Advertising Research Centre, advertising-only
publications such as Loot, Auto Trader, Friday Ad and Job Opportunities would be consumer magazines,
and smaller, local publications would be categorized as local/regional newspapers. Therefore the terms
needed to be defined and, since the question had not done this, it had distorted the results. An analysis of
the use of advertising-only publications by Johnston’s top 30 advertisers had revealed that many
advertisers used advertising-only publications to a far greater extent than the results suggested. Second, it
said that the focus on top 30 customers of newspapers would skew the survey's results, because a survey
of only newspaper advertisers would naturally reveal that they all used newspapers to a significant
extent. Johnston had recommended that the survey should include advertisers who did not use the
publications of the parties as a means of advertising.
22. Table 4, and the survey results generally, should be viewed in the knowledge that the
respondents are major users of the parties’ publications.
TABLE 4 Advertisers reporting their use of different communication media
Per cent of Median per cent of
Advertising medium* advertisers budget used†
Free local newspapers 76 41
Paid-for local newspapers 85 47
National newspapers 11 13
Consumer magazines 10 6
Business magazines 11 13
Advertising-only publications 6 5
Classified directories 10 15
Local radio 28 11
Local television 7 4
Web/Internet 20 4
Direct mail leaflets 17 8
Outdoor advertisements 10 5
Other 5 ..
Source: CC calculations based on ORC survey.
*Categories are not mutually exclusive.
†Refers to those advertisers who reported using a particular medium and who were able to estimate their percentage
of budget spent (ie excludes ‘don’t knows’).
23. The question about types of media used was followed by questions that asked about the
percentage of advertising budget spent on each of the categories previously mentioned. Therefore Table
4 also shows the average percentage of advertising budget spent by media category mentioned, where the
median is the measure of average used. The medians were computed from the distributions of responses,
which were grouped in expenditure bands, by linearly interpolating the value at which half the
respondents reported lower expenditures and above which half reported higher expenditures, excluding
the ‘don’t knows’. There is no reason to expect these medians to add up to 100 per cent. The inter-
pretation of each median in Table 4 is that it indicates the average percentage of advertising budget spent
by the advertisers that reported the use of a particular medium, ignoring those which did not know. It is
184
indicative of the level of use of the various media from the perspective of these advertisers. So, for
example, 28 per cent of advertisers reported using local radio and this typically represented 11 per cent
of their advertising budget. The median spend on non-newspaper printed media varied between 5 and
15 per cent of budget, while spend on local television and the Web/Internet was lower than this.
24. Johnston expressed serious reservations about the robustness of the exact median figures in
Table 4 because, it said, the number of respondents was small in most categories. It pointed out that each
median average was determined only by the middle values in a range, and it believed that in a small
sample, a small number of entries could significantly influence the median. Furthermore, while accepting
that the medians would not necessarily add to 100 per cent, they added to 172 per cent in this case, and
so they could not provide an accurate indication of the level of use of the various media because
advertisers could not spend more than 100 per cent of their budget. Johnston felt that this raised
significant issues as to the reliability of the information in Table 4. For these reasons, it said that the
word ‘typically’, as used in the preceding paragraph, could not generally be used to describe the median.
25. Since we wanted to know about recent switching behaviour between newspapers and other
communication media, ORC asked how often advertisers switched more than 10 per cent of their spend
from newspapers into other media, and vice versa, from campaign to campaign or year to year. Table 4
shows that 10 per cent of budget is not an unusual proportion for these advertisers to spend on some non-
newspaper media. The question asked for responses on a five-point scale from ‘very often’ to ‘very
rarely’. Around 70 per cent of advertisers that ran several campaigns in a year and 80 per cent of those
that had one plan for the whole year said that they ‘rarely’ or ‘very rarely’ switched more than 10 per
cent of their advertising spend between newspapers and other media. Taking all of these advertisers, less
than 7 per cent answered ‘often’ or ‘very often’ with respect to switching either from newspapers to other
media or vice versa. Even considering the sampling error in these estimates, they show that less than
10 per cent of these advertisers reported switching often or very often more than 10 per cent of their
spend into or out of newspapers.
26. Johnston made a number of points in relation to this question to argue that the results understated
the degree of switching. First, it commented that a definition of other media and advertising-only
publications should have been given to the respondents. Secondly, it noted that advertisers that already
had diverted their spend to other media would not be in the sample. Thirdly, it remarked that active
competition to retain customers, which had the effect of limiting switching, was not evidence that
customers would not switch in response to unjustified price increases. Fourthly, Johnston stated that what
it regarded as the implicit assumption that switching of less than 10 per cent was not material to a
newspaper publisher was wholly without foundation. Switching by marginal customers of less than
10 per cent of their advertising would have a significant effect. Johnston also commented that, despite its
reservations about Table 4, if it were the case that 10 per cent of budget were a fairly usual proportion for
advertisers to spend on non-newspaper media, this did not in any way imply that an assessment of how
often advertisers switched 10 per cent of their advertising was a sensible measure of the effectiveness of
the alternative media as a constraint on newspaper publishers. Advertisers did not have to switch the
entirety of their spend on a particular medium to discipline a newspaper publisher and therefore spend on
a particular advertising medium did not cast any light on its effectiveness as a constraint on newspaper
publishers.
27. Respondents were prompted to think about the numbers of free and paid-for weeklies, and then
about the daily and Sunday newspapers that they used. Table 5 shows the popularity of weekly news-
papers for these advertisers; this is to be expected since the sampling frame consisted largely of users of
weekly titles. It also shows that it is common for these advertisers to report using more than one weekly
title, and the patterns of reported use for both types of weekly are very similar.
TABLE 5 Number of different weekly titles used to advertise in the local area
per cent
Free weeklies Paid weeklies
None 22 16
One 30 32
Two 19 19
Three or more 23 27
Don't know 6 6
Total 100 100
Source: CC calculations based on ORC survey.
185
28. It might be thought that the number of different newspapers used would be constrained by the
number of titles available in a locality. For example, one might expect local advertisers to report using no
more than the number of paid-for weekly newspapers in a given area. However, this is not the case.
Although the table shows that it is most often the case that advertisers use one title in the case of both
free and paid-for weeklies, it does not show whether an advertiser uses all the titles available in a given
area. As we have shown above, these advertisers do not necessarily limit themselves to particular local
areas and so it is not possible unambiguously to identify an advertiser with a specific local area and
hence to establish the number of choices available to it.
29. Respondents were also prompted to think about whether they used advertising-only publications,
and then asked how many of the publications specialized in a certain type of advertisement. The question
followed related questions, which asked how many free weekly newspaper titles (excluding advertising-
only publications) were normally used. Some respondents, who did not mention devoting expenditure to
advertising-only publications when unprompted, did so when they were prompted by this question, and it
is possible that a higher proportion of respondents might have said that they used advertising-only
publications if ORC had prompted them to recall every type of such publication by name. Table 6 shows
the responses of the 93 advertisers, that is 20 per cent of the sample, who said that they used advertising-
only publications. In this table, those reporting ‘none’ were those for whom the type of publication (ie
jobs, property, motors or general classified) did not focus on their type of advertising. Not surprisingly,
recruitment and local government emerged as users of specialist job publications, estate agents as the
overriding users of property publications, and motor dealers as predominant users of motors publications,
while a range of different types of organization used general classified publications.
TABLE 6 Number of different advertising-only publications used to advertise in the local areas
Jobs Property Motors Classified
Number % Number % Number % Number %
None 71 76 69 74 49 53 55 59
One 5 5 15 16 23 25 19 20
Two 7 8 1 1 7 8 5 5
Three or more 7 8 2 2 11 12 10 11
Don't know 3 3 6 6 3 3 4 4
Total 93 100 93 100 93 100 93 100
Source: CC calculations based on ORC survey.
Base: All that reported using advertising-only publications when prompted.
Note: Percentages may not sum to 100 because of rounding.
30. In respect of these results, Johnston criticized the survey for not having clearly defined the term
‘advertising-only publication’. It concluded that the consequence was that the term was likely not to have
been understood by a significant number of advertisers and in its judgement the results of the response to
the question were likely to be unreliable. Johnston noted that the 20 per cent who said that they used
advertising-only publications contrasted with the 6 per cent who referred to this unprompted. It thought
that ORC might have categorized responses to the earlier question as consumer magazines, business
magazines, or to newspaper categories which, according to the World Advertising Research Centre, were
advertising-only publications. Furthermore, it reiterated that a survey of only higher-spending newspaper
advertisers would unsurprisingly reveal that they all used newspapers to a significant extent and therefore
their usage of advertising-only publications was likely to be unrepresentative of the market as a whole.
Effectiveness of publications compared with free weeklies
31. We sought to know about the effectiveness, as perceived by advertisers, of paid-for weeklies,
paid-for dailies and advertising-only publications compared with free weeklies. ORC only asked the
opinions of those who said they had a paid-for weekly in their area about paid-for weeklies, but all
respondents were asked their opinions on the effectiveness of local paid dailies and advertising-only
publications. ORC asked for responses on a five-point scale ranging from ‘strongly disagree’ to ‘strongly
agree’. For simplicity, the scale is contracted in Table 7.
32. The question sought to identify which publications were perceived as substitutes by asking if
different types of publication were equally effective. Individual answers do not shed light on which type
186
is considered more effective but the responses indicate which are or are not close substitutes, based on
the balance of opinion either in agreement with or against the proposition of equal effectiveness, and the
pattern of actual use may in some cases enable conclusions to be drawn as to which of the alternative
publications is more or less effective. For example, the responses of recruitment consultants were
different from most of the other advertisers with respect to whether advertising in a local paid-for weekly
newspaper was equally effective as advertising in a local free weekly newspaper: two-thirds of the
recruitment agencies disagreed with the proposition. Since recruitment agencies use paid-for titles rather
than free weeklies, this suggests that they considered paid-for titles more effective than free weeklies.
33. However, Johnston said that this question was fundamentally ambiguous as ‘equally effective’
excluded both ‘more effective’ and ‘less effective’. As an illustration of this, Johnston referred to the
responses of recruitment consultants with respect to advertising in paid-for weeklies and concluded that
recruitment agencies must have been responding that paid-for titles were not equally effective, but more
effective, because recruitment agencies tended not to use free weekly newspapers to any great extent but
to use paid-for titles. It stated that this example illustrated that the responses to all the ‘equally effective’
questions could not be relied upon as it was impossible to be certain what respondents meant when they
gave a ‘disagree’ or ‘strongly disagree’ response. Thus, Johnston felt that no reasonable reliance at all
could be placed on the responses to these questions as a basis for determining substitutability.
34. The results in Table 7 show that half of the advertisers that had access to a paid-for weekly
perceived paid-for and free weeklies to be equally effective, compared with a third that disagreed. So the
balance of opinion is that these are equally effective advertising media.
35. A greater proportion of respondents was equivocal concerning the other comparisons. Opinions
concerning whether paid-for dailies were equally effective as free weeklies were more or less equally
balanced. Relative to the overall proportion, slightly more of the advertisers in the Northampton area
were of the opinion that these types of title were equally effective and fewer of them said that they did
not know. However, these differences from the overall pattern were not statistically significant.
36. With respect to the observation that more of the advertisers in the Northampton area than the
overall average were of the opinion that these types of title were equally effective, Johnston commented
that there was therefore no basis on which to regard paid-for daily newspapers as in a separate product
market from free weeklies (and indeed paid-for weeklies). Johnston also added that the fact that
respondents in Northampton and Peterborough appeared to be giving different responses again suggested
that the responses were unreliable given that the situation in terms of the number of paid-for evening and
free weekly titles was identical in Peterborough and Northampton.
37. The balance of opinion was that advertising-only publications were not equally effective as free
weeklies. The proportion disagreeing with the proposition was more than twice as large as the proportion
in agreement. This pattern of response is observed across all types of business advertiser and advertising
agency. 80 per cent of all respondents, when prompted, said that they did not use these publications, and
even though the question was neutral, it would be surprising to conclude that the majority thought that
these publications were more effective than free weeklies.
TABLE 7 Effectiveness of publications compared with free weeklies
per cent
Disagree or Agree or
strongly strongly Don’t
disagree Neutral agree know Total
Compared with free weekly …
Paid-for weekly is equally effective* 34 10 50 7 100
Paid-for daily is equally effective† 28 16 33 24 100
Advertising-only publication is equally effective† 46 11 22 22 100
Source: CC calculations based on ORC survey.
*Base: all who have a paid-for weekly in their area.
†Base: all respondents.
Note: Values may not sum to 100 per cent because of rounding.
38. However, Johnston reiterated its view that the failure to define advertising-only publications was
one indication of why the responses to this question should be treated with considerable caution and
187
observed moreover that the fact that a significant proportion (22 per cent) of advertisers regarded
advertising-only publications as equally effective was highly significant. It thought that this strongly
suggested that, in the event of a 10 per cent price rise, a significant number of advertisers would consider
switching their expenditure to advertising-only publications. It argued that it was not necessary for the
majority of advertisers to regard advertising-only publications as substitutes for these publications to act
as a competitive constraint.
39. Trinity Mirror commented that whereas the question concerning relative effectiveness was asked
only of those who had paid-weekly titles in their area, all respondents were asked their opinion about the
effectiveness of paid dailies and advertising-only publications. It felt that where paid and free weekly
titles were serving the same market, these were likely to be seen as equally effective for many types of
advertising. It equally believed that paid evening titles and free weeklies would be seen as equally
effective for many types of advertising where these genuinely served the same market. Because the
question concerning paid dailies was put to all respondents and not just to those who had access to an
evening newspaper that served their particular geographic target area, it felt that the aggregated response
was not compelling since many of the areas covered by the survey had no evening newspaper of
relevance. It noted in particular that responses concerning paid dailies by those who advertised in the
Northampton area, which was covered by two evening titles, showed almost the same proportion of
agreement to disagreement as for the paid weeklies. It remarked that free weekly titles were unlikely to
be regarded as equally effective as evening titles for certain types of advertising. Recruitment advertising
and BMDs were two such categories, which for historic reasons had tended to remain in the evening title.
40. Trinity Mirror also felt that aggregation caused problems of interpretation concerning
advertising-only publications. In its view, the question assumed that each geographic market was served
by appropriate advertising-only publications whereas, it noted, many of these publications were focused
on a single advertising category. For the aggregated results to treat these publications fairly would
necessitate that such publications existed in each geographic market covering each advertising category.
It did not believe that this was the situation in the markets covered by the survey. Since the motors
market had been served best by such publications, it felt that it would not be surprising if motor repairers
and motor dealers were the most likely advertisers to view these as effective substitutes and it noted that
the balance of those agreeing and disagreeing was almost equal.
41. Trinity Mirror felt that the key point was that advertising-only publications were a threat to local
newspapers for certain types of advertising. It was clear to Trinity Mirror that a large proportion of
motors advertisers would view advertising-only publications as effective alternatives, although there
would be differences by category of advertiser: recruitment was the obvious example. There were few
advertising-only recruitment publications and, not surprisingly in its view, none of the recruitment
advertisers in the sample agreed that these publications were equally effective. In terms of estate agents,
although the survey showed a significant balance against the proposition of equal effectiveness, in its
view this could reflect the fact that advertising-only publications did not exist in these areas. Using an
example of a London estate agents’ guide, it thought there was little difference between such an
advertising-only publication distributed through letterboxes and a traditional free newspaper.
Price seeking and experience of discounts
42. ORC asked whether it had been usual during the previous 12 months for advertisers to seek price
quotations, from free or paid-for weekly titles, or from advertising-only publications. The five-point
scale used for answers ranged from ‘very unusual’ to ‘very usual’ and is collapsed to a three-point scale
in Table 8. Although overall there is an even balance of those who had or had not usually sought
quotations, there are statistically significant differences. A greater proportion of those who advertised
their own business locally or regionally said that it was unusual to seek alternative quotes, whereas a
greater proportion of advertising agencies said that it was usual to seek quotes. Table 8 shows how the
pattern differs by type of business. A note of caution is due when breaking down these responses to a
detailed level, since low counts in individual cells can mean that results are subject to relatively large
sampling errors. However, more advertising agencies, recruitment agencies and motor dealers had
usually sought quotes than had not, whereas the opposite was the case for estate agents, leisure and
consumer services, and educational institutions. On this question, two-fifths of those respondents who
were informed about the inquiry said it was unusual or very unusual to seek alternative price quotes. A
significantly larger proportion, that is half, of those respondents who did not ask, and hence were not
informed, that the survey was being conducted on behalf of the CC, said that it was unusual or very
188
unusual to seek alternative quotes. It may be that these people are less informed as a result of their own
less inquisitive behaviour.
TABLE 8 How usual to seek price quotes from weeklies and advertising-only publications, by type of
advertiser
Type of business Advertising All
Agency
Retail Trades, Leisure,
store or incl consumer Local Recruit-
Estate super Motor motor & financial govern- ment Educa- Regional
agent market dealer repairs services ment agency tion Other National /local
number
Total 94 88 62 37 33 19 13 12 41 48 29 476
per cent
Usual or
very
usual 30 44 61 46 33 42 62 25 49 75 52 47
As often
as not 4 3 5 3 6 16 8 8 2 21 5
Unusual
or very
unusual 59 45 27 41 61 42 31 67 41 17 24 42
Don’t
know 7 7 6 11 0 0 0 0 10 6 3 6
Total 100 100 100 100 100 100 100 100 100 100 100 100
Source: CC calculations based on ORC survey.
Note: Values may not sum to 100 per cent because of rounding.
43. Johnson noted that although there was a greater proportion of those who advertised their own
business locally or regionally and said that it was unusual to seek alternative quotes, this was 49 per cent,
as opposed to 42 per cent of all respondents. It commented that the responses of estate agents were not
credible given that estate agents were one of the categories of advertiser which drove the hardest bargain
and had various alternatives, including advertising-only publications, available to them. Moreover, in
practice, estate agents, along with motor dealers, obtained the lowest rates. It considered that responses to
this question understated the extent to which competitive quotes were sought. It noted that the low counts
in the categories of leisure and consumer services and tradesmen made these results unreliable. It also
noted that the question referred to the previous 12 months whereas 38 per cent of respondents (and
47 per cent of estate agents) had indicated that they had one advertising plan for the whole year, with the
result that they were not likely to seek alternative quotes during the year. In addition, the results for
advertising and recruitment agencies, which tended to run several campaigns during a year, suggested
that the main determinant of whether alternative quotes had been sought over the last 12 months was
whether or not the advertiser had an annual plan or several campaigns in a year. Johnston added that
seeking alternative price quotes was not the only means of advertisers ensuring that prices were com-
petitive. For example, informal enquiries of other competitors and discussions with advertising agencies
might be equally informative.
44. ORC similarly asked these advertisers about their recent negotiating behaviour and success in
obtaining discounts from free and paid-for titles. A statistically significant higher proportion of those that
advertised their own business locally or regionally said that it was unusual or very unusual to obtain
discounts. As Table 9 shows, overall there is a close match between negotiating behaviour and success in
obtaining discounts from both free and paid-for titles. Generally these advertisers negotiated and were
successful in obtaining discounts; about three times as many reported that this was usual or very usual
compared with those that said it was unusual or very unusual. Advertising agencies and motor dealers are
among the most active negotiators. When respondents said that they did not know about or never
obtained discounts, ORC asked if they thought that the local newspapers were willing to give discounts.
Two-thirds of these respondents replied affirmatively.
45. With respect to the fact that a higher proportion of those that advertised their own business
locally or regionally said that it was unusual or very unusual to obtain discounts, Johnston thought this
gave an entirely misleading impression of the position because, it said, the difference in responses was
only six percentage points and, even in this case, twice as many respondents said that it was usual or very
189
usual to obtain discounts. As regards negotiation, Johnston noted that estate agents were also extremely
vigorous negotiators, although the survey results appeared to point in the opposite direction, and this was
another indication that the survey results had to be interpreted with considerable caution. It said that the
survey results underestimated the degree of discounting because information it had provided to the CC
]
indicated that, overall, over [✄ per cent of its advertisers in Peterborough obtained a discount from the
rate cards.
46. If there were to be bias in responses arising from respondents’ awareness of this inquiry, it might
be expected to arise in these answers. However, there was no significant difference between responses
from those that had and had not been informed that the survey was being conducted on behalf of the CC.
Johnston commented that advertisers might have been aware from contacts with advertising agencies or
others that ORC was conducting a survey on behalf of the CC.
TABLE 9 How usual to negotiate and obtain discounts off the rate card
per cent
Unusual or As often Usual or Don't
very unusual as not very usual know Total
Negotiating discounts with free titles over past
12 months 22 4 60 14 100
Success in obtaining discounts with free titles
over past 12 months 18 6 58 17 100
Negotiating discounts with paid-for titles over
past 12 months 26 5 58 11 100
Success in obtaining discounts with paid-for
titles over past 12 months 21 7 59 14 100
Source: CC calculations based on ORC survey.
Note: Values may not sum to 100 per cent because of rounding.
47. ORC asked all respondents for estimates of the discounts they obtained from the rate card for
both free and paid-for titles. About a third of the respondents said that they obtained discounts of 24 per
cent or less from free titles, or they never got discounts. A quarter said that they obtained discounts of
25 per cent or more. Thus, more report discounts below 25 per cent. The median annual advertising
expenditure in the local area (estimated in the manner explained previously) for businesses that never got
discounts or received discounts of up to 24 per cent is about £17,000. For those who received discounts
of up to 24 per cent (ie excluding those who did not get a discount at all) the median annual advertising
spend is about £19,000 and this compares with a median expenditure of around £36,000 for businesses
that obtain discounts of 25 per cent or more. (These figures exclude advertising agencies and ‘don’t
knows’.) Figure 1 illustrates the relationship between reported discounts and the level of advertising
expenditure. The figure shows that the distributions of advertising expenditure for those who reported not
getting a discount, lower discounts and higher discounts described those who spend at relatively low,
medium and high levels in the same order. Those who said that they don’t know have a very similar
distribution to the medium spenders, who reported lower discounts.
48. With respect to the use of 25 per cent to categorize advertisers according to whether they
obtained higher or lower discounts than this, Johnston commented that a discount of 25 per cent was
significant and that if the arbitrary line between ‘higher’ and ‘lower’ had been drawn at 20 per cent, the
balance of those getting higher or lower discounts would have been reversed. Furthermore, it emphasized
that 43 per cent of respondents gave the reply that they did not know and it reiterated that the majority of
those who did not know their discount would be advertisers who were unaware of the rate card charge
because they negotiated on the basis of net prices, which in fact represented substantial discounts from
the rate card. With respect to Figure 1, Johnston commented that it was to be expected that higher-
spending advertisers would obtain higher discounts and this was entirely normal. Moreover, with respect
to the distribution of ‘don’t know’ respondents, it commented that it could not be assumed from this that
these advertisers obtained discounts below 25 per cent. For example, it noted, 51 per cent of estate agents
gave a ‘don't know’ response, but in practice they obtained very substantial discounts from the rate card.
49. Trinity Mirror felt that since the bands into which respondents had reported their annual adver-
tising expenditure increased with the scale, the figures quoted above were not a fair analysis and it
questioned if these reported results were supported by facts. It expected to see some extremely high
spends among the top 30 customers, given that higher-spending customers tended to spend the greater
part of their advertising budget on the local press, but the annual spend of Trinity Mirror customers did
not support such high overall expenditure.
190
FIGURE 1
Local business advertising expenditure by different reported discounts
from free titles
100%
90%
Proportion of businesses with lower expenditure
80%
70%
60%
50%
40%
30%
20%
10%
0%
£1,000 £10,000 £100,000 £1,000,000 £10,000,000
Annual advertising expenditure reported
Obtain 0 - 24% discount Obtain 25% or more discount
Never get discount Don't know
Source: CC calculations based on ORC survey.
50. Table 10 shows that recruitment agencies, local government, tradesmen and other local
businesses are relatively more likely to report lower discounts from free titles.
TABLE 10 Estimated discounts obtained off the rate card with free newspapers, by type of advertiser
Type of business Advertising All
Agency
Retail Trades, Leisure,
store or incl consumer Local Recruit-
Discount Estate super Motor motor & financial govern- ment Educa- Regional
% agent market dealer repairs services ment agency tion Other National /local
number
Total 94 88 62 37 33 19 13 12 41 48 29 476
per cent
Never
get 14 5 5 5 3 21 15 8 12 4 7 8
0–24* 14 20 26 24 30 21 46 33 32 29 14 23
25 or
more 21 23 40 19 27 11 15 17 17 35 48 26
Don't
know 51 52 29 51 39 47 23 42 39 31 31 42
Total 100 100 100 100 100 100 100 100 100 100 100 100
Source: CC calculations based on ORC survey.
*Excludes those who never get a discount.
Note: Values may not sum to 100 per cent because of rounding.
191
51. However, Johnston commented that the results differed from its own knowledge of the facts. For
]
example, over [✄ per cent of advertisers in its free weekly Peterborough Citizen obtained discounts, and
this was also true of advertisers in the recruitment category. It added that the low counts for certain
categories had led to misleading results. For example, only three recruitment advertisers reported
discounts of under 20 per cent.
52. Table 11 shows the reported discounts off the rate card for paid-for newspapers, where again a
third said that they obtained less than 25 per cent and a quarter said that they obtained discounts of more
than this. Recruitment agencies, local government, services, tradesmen and other local businesses are
relatively more likely to report lower discounts. Despite the sample consisting of high-spending
advertisers, 8 per cent said that they obtained no discount from free and from paid-for newspapers,
although it may be that some of these advertisers were not aware of the rate-card rate.
TABLE 11 Estimated discounts obtained off the rate card with paid-for newspapers, by type of advertiser
Type of business Advertising All
Agency
Retail Trades, Leisure,
store or inc. consumer Local Recruit-
Discount Estate super Motor motor & financial govern- ment Educa- Regional
% agent market dealer repairs services ment agency tion Other National /local
number
Total 94 88 62 37 33 19 13 12 41 48 29 476
per cent
Never
get 16 2 3 11 9 16 15 0 10 4 0 8
0–24* 20 23 26 24 39 37 46 50 41 42 24 29
25 or
more 23 27 44 14 12 11 0 17 15 27 38 24
Don't
know 40 48 27 51 39 37 38 33 34 27 38 38
Total 100 100 100 100 100 100 100 100 100 100 100 100
Source: CC calculations based on ORC survey.
*Excludes those who never get a discount.
Note: Values may not sum to 100 per cent because of rounding.
53. Johnston commented that it was likely that estate agents were among those not aware of the rate-
card rate because they typically negotiated on the basis of net prices.
54. The discounts from rate cards reported in this survey represent perceptions rather than exact
computations, for the following reasons. The survey shows that many advertisers used more than one
]
title, and it is known that discounts can vary widely, for example from around [✄ per cent within a
single national advertising campaign. ORC asked for estimates of average discounts off the rate card. In
this context, respondents are likely to give a figure that they would regard as typical, and that they would
use in judging whether a particular discount is higher or lower than their expectations.
55. Two-fifths of respondents said that that they did not know the discounts. The parties suggested
that advertisers might not be aware of the rate-card prices and the discounts that they received, since
many negotiated on the actual price paid. A rational response to the question in this situation is ‘don’t
know’ and this would explain the large proportion of ‘don’t know’ answers.
56. Therefore the pattern of responses represents perceptions rather than exact computations of
discounts. Viewed in this way, the tables above indicate which advertisers are more likely to obtain
lower discounts, and that is how we have interpreted them.
57. However, Trinity Mirror thought that this analysis of aggregated responses was flawed because
of the way in which different categories of advertising were priced and sold within a title and by the way
that overall levels of discounts were offered against rate cards.
58. Trinity Mirror thought that significant discounts would be available for major regular advertisers
and that, within any category, the level of discount would reflect the scale of expenditure of a particular
192
advertiser. Since local employees in estate agencies, and other advertisers, moved and talked to each
other it would be important for the rates to be consistent and to reflect the importance of the advertiser.
Moreover, it would be important for the discount structure to retain its integrity from an internal
perspective or otherwise the whole pricing structure could collapse if individual sales staff were to
reduce rates to the lowest common denominator.
59. Trinity Mirror felt that aggregating the survey results was problematic. First, it felt that to use the
levels of discounts on an aggregated basis ignored the differences across each category. It noted that the
different categories of advertising within one title, and even within a category of advertising, would have
different pricing structures and discounts. It gave three examples. A motor advertiser would face two
schedules, one for advertising within a motors section and one within ROP (run of paper). In property,
the discounts available for general estate agency advertisements would be different from those available
for new homes whether or not the advertiser was the same. Recruitment advertising would normally
attract lower discounts than either property or motors advertising even though a particular recruitment
agency might be spending more money with a title.
60. Secondly, in its view, a valid aggregation would require that all the titles within the survey had
adopted a uniform approach to developing their rate cards. Given that advertisers did receive discounts,
the level of such discounts reflected the price each advertiser was willing to pay for the space compared
with the rate-card rate. To the extent that a publisher either doubled or halved the rate-card rate, the
discount might well vary to achieve the same rate per page. Therefore it could not be assumed that the
discount would remain static if the rate card were to increase or decrease. Thus, in its view, the discounts
achieved across a range of titles reflected in part the extent to which rate-card rates were more or less
realistic of actual rates paid by advertisers.
61. Accordingly, Trinity Mirror believed that it was misleading to use the levels of discounts
achieved as some form of indicator as to whether individual advertisers fared better or worse in terms of
discounts. However, Trinity Mirror fully supported the proposition that higher-spending advertisers
would achieve better rates than lower-spending advertisers in the same category but considered that
comparisons of discounts across categories and titles were meaningless.
62. Johnston also criticized the approach described in paragraph 56. It noted that the analysis made
no reference to the actual discounts received off rate cards provided by Johnston. It believed that the
incidence of discounts was substantially understated by the survey responses, both overall and in specific
advertiser categories. It considered that the data should not have been aggregated into the categories of
discount shown in Tables 10 and 11 but presented in the level of detail provided by the survey. For
example, it argued, local government advertisers all reported a discount of 10 per cent or more which, it
argued, was contrary to the impression given by the tables. At the same time, it noted that some of the
counts were low for certain categories of business and questioned the robustness of presenting those
results. Johnston commented that advertisers reporting no discount would typically receive net prices that
represented very substantial discounts off the rate card. It criticized the absence of guidance in the
questions as to how package discounts or other types of discount should be treated, arguing that this
would lead to advertisers categorizing the amount of the discount differently and some understating the
overall discount granted as a consequence. In view of these points, it regarded a conclusion that the
tables indicated which advertisers were likely to obtain lower discounts as being without foundation.
63. Finally ORC asked what would be the response of advertisers were all the local weekly papers in
the area to put up their rates by 10 per cent. These responses were not prompted. Most respondents said
that they would react in some way, with 18 per cent saying they would negotiate, ask for justification, or
review requirements and a further 38 per cent saying they would switch or reduce advertising in some
way. 30 per cent said they would do nothing. Three-quarters of the recruitment agencies and half the
local government advertisers gave this reaction. Table 12 shows the details but caution should be
exercised in reading these results because of low counts for particular responses.
64. With regard to the result that three-quarters of recruitment agencies and half the local govern-
ment advertisers might do nothing, Trinity Mirror commented that they might have selected to use the
paid title for which they had no alternative. Johnston commented on the same result that the eight Trinity
Mirror titles did not carry any appreciable job advertising or public notices (which covered the vast
majority of the advertisements placed by recruitment agencies and local government advertisers), and
therefore that competition for such advertising would not be materially affected; that, as with all
advertisers, the results understated the responsiveness of recruitment and local government advertisers to
193
local newspaper price increases; and that competition was increasing in jobs advertising from
advertising-only publications targeted at local jobs and the growth of the Internet as a means of searching
for jobs.
TABLE 12 Reaction to 10 per cent rate increase in all the local weekly papers
Per cent of
Reaction respondents
Do nothing 30
Pass on the cost to customers/raise our prices 1
Negotiate a price 14
Would ask for justification 1
Review advertising strategy/client requirements 2
Switch advertising to other local paper 7
Switch advertising to other medium 14
Cut out local weekly paper advertising 7
Cut back on advertising/use them less 9
Reduce the size of adverts 1
Other reaction 5
Don't know 9
Total 100
Source: CC calculations based on ORC survey.
65. Johnston also noted several factors which, it argued, understated the responsiveness of
advertisers to price increases. As a result, it considered that the responses to the question unambiguously
indicated that, following the proposed transfer, price increases would not be profitable. First, it argued
that the responses to the question posed in the survey understated the degree of switching to advertising-
only publications because, it believed, a significant proportion of those surveyed interpreted the question
as meaning that the price of advertising-only ‘papers’ would also increase, because the question referred
to ‘local weekly papers’ rather than ‘newspapers’ and the examples of other media to which the
advertiser could switch did not include advertising-only publications. Secondly, because only the first
response of advertisers was recorded, only in the case of ‘do nothing’ would advertisers necessarily only
have a single response. It argued that many of those advertisers who said that they would reduce
advertising would probably also divert expenditure to other media, and similarly those who said that they
would negotiate would be negotiating on the basis that they had a variety of ways of disciplining the
newspaper publisher (for example, switching to other media, advertising less and so on), and that
negotiation was not possible in the absence of alternatives.
66. Johnston believed that a significant proportion of those indicating that they would ‘do nothing’
might well have meant that they would maintain the proportion of their advertising budgets allocated to
local weekly papers and would not spend any more on local ‘paper’ advertising (with the result that they
would buy less advertising space). It argued that clearly only if certain advertisers increased their
expenditure following a price increase would publishers’ revenues increase from such advertisers.
Johnston added that, for certain categories of advertiser, the level of ‘do nothing’ responses was not
credible. In particular it noted that 35 per cent of the estate agents surveyed stated that they would do
nothing in response to a 10 per cent price increase. Since, it noted, estate agents’ advertising tended to be
placed in a single publication, if any of the estate agents could divert their expenditure (and 34 per cent
said that they would), they all could. Finally, Johnston believed that the results would be dependent on
whether advertisers that gave the response of ‘switch to other local paper’ were indicating that they
would switch to paid-for evening titles or advertising-only publications.
67. Following these comments, Johnston provided a computation to show how its revenues from its
top 30 advertisers in the Peterborough and Northampton areas would fall as a result of their response to a
10 per cent price rise. It computed that, even on a conservative basis, the proportion of these top
advertisers (56 per cent) whose reactions would reduce Johnston’s revenue would be significantly greater
than the proportion (36 per cent) that would increase it with the result that price increases would not be
profitable, bearing in mind that any cost saving achieved as a result of carrying less advertising would be
entirely de minimis.
68. We can classify the respondents who said that they would do nothing in response to a price
increase of 10 per cent into those who did not get a discount, or reported discounts of 24 per cent or less
(lower discounts), and those who reported discounts of 25 per cent or more (higher discounts). When this
194
is done, we find that 40 per cent and 50 per cent respectively of those giving the ‘do nothing’ response
reported lower discounts off free and paid-for titles. This compares with smaller proportions, namely 20
per cent and 15 per cent respectively, who reported higher discounts off free and paid-for titles. The
results are shown in Figure 2.
FIGURE 2
Proportion of advertisers reacting to price increase of 10 per
cent by reported discounts
60%
50%
40%
Proportion of respondents
30%
20%
10%
0%
0-24% off 25%+ off free 0-24% off 25%+ off Don't know Don't know
free paid paid off free off paid
Discounts reported
Do nothing React to increase
Source: CC calculations based on ORC survey.
69. Johnston commented that paid-for titles were largely irrelevant in the context of the current
inquiry which concerned the proposed acquisition of eight free titles. Furthermore it said it was probable
that among those who said that they would do nothing and did not know what discount they received, or
said that they never got a discount, would be advertisers that paid net prices and regarded the rate card as
irrelevant. These did in fact receive substantial discounts. Johnston also made a number of detailed
points, including the fact that those who would respond to a price increase comprised advertisers who fell
both into the lower and higher discount categories. (In this connection, Johnston pointed out that 52 per
cent of those reporting that they would ‘do nothing’ indicated that they received no discount or did not
know the discount.) Furthermore, the discrepancy between those claiming to obtain lower discounts
from free titles who said that they would do nothing on the one hand, and those who would react on the
other (40 per cent as against 28 per cent), was not particularly substantial and would hardly suggest a
uniform tendency.
70. In order to include all these considerations, Figure 2 shows the overall picture. The proportions
who ‘don’t know’ are similar among those who would or would not react. Those who would react to a
195
rate increase in some way are fairly evenly distributed over lower discount and higher discount
categories. Those who would do nothing have a markedly different distribution, as described above.
71. A further analysis of those who would not react to a 10 per cent rate increase is shown in
Table 13. The table focuses on the 77 respondents who could report their discounts for both paid and free
local titles and who would do nothing if prices increased 10 per cent. Two-thirds of these advertisers
reported not getting a discount or having obtained discounts of 24 per cent or less off all local newspaper
rate cards during the previous year, so that those who said that they would not react are advertisers who
also tended to say that they obtain lower discounts or for whom actual discounts were not transparent. As
we have noted above, the typical annual advertising expenditure in the local area for businesses that did
not get discounts or that received discounts of 24 per cent or less is about half that of businesses that
obtained higher discounts.
TABLE 13 Discounts reported by those who said that they would not react to a rate increase
number
Average discount off local free titles
0–24% or do
not get 25% or more Total
Average discount off 0–24% or
local paid-for titles do not get 49 12 61
25% or
more 2 14 16
51 26 77
Source: CC calculations based on ORC survey.
Base: All who would do nothing if local press advertising rates increased 10 per cent and who could state their
average discount off rate cards for free and paid-for local titles.
72. Johnston, however, made a number of detailed points, which it said undermined any generaliz-
ation concerning this table. It stated that it was quite clear that many of those reporting lower discounts in
one category of title (free or paid-for) in fact received higher discounts in the other, although the small
size of the sample meant that any analysis based on the table should be treated with caution. For
example, of those who claimed to receive a discount of under 25 per cent in paid-for titles, 20 per cent
reported that they received discounts of over 25 per cent in free titles. Similarly, just under half of those
which claimed to receive a discount of over 25 per cent in relation to free titles, claimed to receive
discounts of under 25 per cent in relation to paid-for titles. It concluded that the pattern revealed in
Table 13 could only be extrapolated if there were reasonable grounds for concluding that the ‘don’t
know’ respondents received discounts which exactly mirrored those reporting specific discounts in both
categories of titles. It thought that this was wholly implausible, reiterating the point it made above that
many of those who said that they did not know, such as estate agents, would in fact receive substantial
discounts.
73. With regard to the nature of the advertisers who said that they did not know their discounts, it is
instructive to inspect Figure 1, which shows that in terms of the advertising spend ‘don’t know’
respondents as a whole are more akin to medium spenders who obtained lower discounts.
74. However, Johnston commented that it could not be assumed that simply because the size of the
advertiser was similar, the level of discount received would also be similar. It was a fact that many estate
agents would be relatively small businesses, but it said that they would nevertheless obtain substantial
discounts from the rate card, although a significant proportion of them had indicated that they did not
know what discount they obtained. Furthermore, it was also the case that a significant number of large
advertisers did not know what discount they obtained and an even higher number obtained ‘lower’
discounts.
75. When Trinity Mirror looked at the advertisers who would react to a 10 per cent price increase, it
found a clear pattern that the higher the spending the more likely such advertisers were to react, and it
concluded that its most important advertisers were the most likely to react to negate any price rise.
76. However, Trinity Mirror found difficulty in understanding exactly what was meant by the ‘do
nothing’ response. If ‘do nothing’ were taken to mean that the advertiser would pay the price increase,
the result that 35 per cent of estate agents said they would do nothing was contrary to all its evidence of
196
how estate agents had reacted to price increases (whether inflationary or otherwise), since estate agents
were not inclined to accept any price increases at all. Furthermore, it argued that there was no evidence
that rate was a significant issue for recruitment agencies, which were inclined to utilize the paid title (or
paper of record), and the reaction of these advertisers was explained by their perception that they had no
realistic alternative currently. It pointed out that, although the Trinity Mirror titles had existed for a long
time, they had achieved only a tiny share of such revenue. It remarked that local councils would typically
use a number of advertising categories and they might or might not choose to use a particular title rather
than seek the best rate. In any event, it noted, a number of councils were now producing their own
publications and Trinity Mirror produced titles for both Northampton and Corby.
77. Trinity Mirror commented that the numbers of advertisers who would ‘do nothing’ in the two
key geographic areas most relevant to the inquiry were lower than the overall result. The overall figure
was 30 per cent and among advertisers in the Northampton and Peterborough areas it was 27 per cent and
26 per cent respectively.
78. Finally Trinity Mirror commented that advertisers did not buy a consistent product. Just because
the cost of an advertisement was more expensive did not mean that the advertisement was less economic
when measured against response.
Johnston-commissioned analyses
79. Johnston drew our attention to the findings of two studies it had commissioned from the consult-
ants, Andersen. One was a review of the ORC survey and we summarize Andersen’s main conclusions
next. The second was a study based on anecdotal evidence garnered from interviews with individuals
who represented 9 out of 20 companies that were contacted (see Appendix 6.1). The persons interviewed
represented these companies or their advertising agencies and their answers were presented as a guide to
the attitudes of the largest UK advertising sponsors of the regional press during the period of the inquiry.
80. Andersen regarded ORC’s survey as reasonable in terms of its size and complexity. However, it
identified certain factors that, in its view, explained its limitations. In some cases it had reason to
question the reliability of results where these differed from the experience of Johnston. It considered that
this could be explained by a lack of clarity in certain questions which led to misinterpretation by
respondents.
81. It thought the survey suggested that a wide range of other media was used by advertisers but that
the use of certain non-newspaper media, notably advertising-only publications, was likely to be under-
stated partly because of potential confusion over certain terms. This view was supported by the fact that
inconsistent responses had been given to similar questions.
82. Andersen commented that a survey of the highest-spending advertisers in a sample of newspaper
titles would inevitably reveal that they predominantly used newspapers, otherwise they would not qualify
as the highest-spending advertisers.
83. It thought that the survey rightly sought to investigate substitutability by asking about switching
from one campaign or annual plan to the next but that a definitive interpretation in terms of the potential
response to price increases was not possible. However, it thought that the responses on discounts
suggested that respondents were able to negotiate good deals, and respondents’ reactions to a hypotheti-
cal price increase indicated that respondents would switch if stimulated to do so. It thought that these
points meant that advertisers considered other media as substitutes but might not have had cause
significantly to switch their expenditure to date. Those that had already switched substantial expenditure
out of newspapers would not have featured in the survey because they would no longer be top advertisers
in local newspapers.
84. Andersen thought that questions about the type and level of print media use did not produce
accurate results, that use of daily titles was more than that reported in the survey, and that the number of
advertising-only titles prevalent in the areas surveyed meant that the use of these publications was greater
than reported by the respondents. It commented that, without the identification of the various types of
advertising-only publication, and confirmation of familiarity of respondents with titles of each type, the
results might not be reliable.
197
85. Andersen thought that since between 22 and 50 per cent of respondents, depending on the type of
publication (see Table 7), agreed that advertising in alternative print media was equally effective as in a
local free newspaper, this suggested that a material level of switching was taking place. It thought that
the results understated the percentages agreeing that alternative print media were equally effective
because the answer ‘equally effective’ precluded the alternative of ‘more effective’ and ‘less effective’.
(These comments about alternative answers do not seem to relate to the questions as posed—see Table
14, Q23 to Q25. Andersen concluded that the results understated the proportion that agreed alternative
print media were equally effective. Its argument was that those who disagreed might do so for different
reasons, but this should not have affected the proportion agreeing.)
86. Andersen thought that questions about discounts did not capture all dimensions of commercial
pressure exerted by advertisers in negotiations and that questions about price discounts did not capture
package discounts or free advertising. It considered that evidence presented to it by Johnston suggested
that actual discounts were more extensive than suggested by the survey results.
87. It considered that the reaction to a hypothetical price increase showed that 48 per cent of
respondents would switch to other media or review/reduce their advertising spend in response to a 10 per
cent increase in prices by all weekly papers. It thought that losses from sales foregone would exceed any
extra profit resulting from a higher price and that the price rise would not be profitable.
Limitations of the survey
88. A number of points were made to us concerning the limitations of the survey. First, the sample
consisted of the higher-spending advertisers in local newspapers and therefore their advertising spend
was more likely to be in favour of newspapers. Johnston expressed serious concerns that a survey of the
parties’ top 30 advertisers would reveal that they used newspapers extensively. Johnston observed that
notwithstanding its concerns, the ORC survey failed to cover any advertisers using any other pub-
lications.
89. We adopted the approach taken because we wanted the sample to represent advertisers that
would be most affected by any change in rates in the newspapers concerned, and we wished to know the
extent to which these advertisers used other media and saw them as substitutes for local newspapers. The
survey was among higher-spending advertisers, but the observed associations between low advertising
expenditure, lower discounts and lack of reaction to rate increases could be expected to extend to lower-
spending advertisers although Johnston disagreed that the same relationships would apply.
90. A key reason to include a control group in this study would be to check whether awareness of the
purpose of the survey may have had an effect, and indeed ORC anticipated this in the research design by
identifying whether answers from those who were informed by them about the inquiry differed from
those who were not informed. We have noted above the question where responses show any effect,
according to whether the respondent did or did not ask the purpose of the survey, and also considered one
question where an effect might have been anticipated, but there is no evidence that there is any bias in
the key results. The response rate was 60 per cent and ORC asked a number of questions that classified
respondents by the type of business, to show that the proportions of different advertiser were not
markedly different from those expected.
91. The choice of 10 per cent of advertising spend used in connection with questions about switching
behaviour may not give a picture of those who switched smaller amounts, but as we have noted, this is
the order of magnitude of spending in any one other medium. The responses indicate a general lack of
switching at this level, even though Johnston considered smaller losses of advertisers’ expenditure a
serious commercial concern.
92. ORC asked about spend on classified and display advertising, and it would seem from the
responses that the distinction between display and classified was not the same as that understood by
newspaper professionals. That is, the distinction between display and classified in the publishing industry
relates essentially to the location of advertisements, so that large block advertisements in the classified
section of a newspaper are classified and not display advertisements. Consequently the answers to this
question have not been used in drawing any conclusions.
198
93. Johnston regarded this as one example of the absence of clear definitions in the survey and
concluded that the whole survey was unreliable. It said that it was unreasonable for the CC not to rely on
certain responses where there was evidence that the questions had not been understood, but not to reach
the same conclusion in respect of other questions, which in its opinion suffered from similar defects and
in respect of which there was evidence of a similar lack of understanding.
94. Another example noted by Johnston was that the survey did not offer guidance as to how
different publications should be categorized and that inconsistent terms were used. In particular, Johnston
noted that some free newspapers and ‘shoppers’ had little editorial content, that many advertising-only
publications were printed on newsprint, and that ‘advertising-only publications’ was not a defined term
in the publishing industry. Johnston reiterated that the World Advertising Research Centre, responsible
for the Advertising Association statistics, categorized the publications which Johnston regarded as
advertising-only publications under several headings (including consumer magazines and business
magazines). Nevertheless, no definition of this term had been used in the survey despite Johnston indicat-
ing to the CC in advance that this was necessary.
95. Johnston considered that questions concerning equal effectiveness of other publications com-
pared with free weeklies suffered from a failure to define the various categories of publication. Johnston
also considered these questions to be ambiguous, arguing that ‘equally effective’ excluded ‘more
effective’ as well as ‘less effective’ as alternatives. Citing the responses of recruitment advertisers that
paid-for titles were not equally effective as free titles (and noting that they only used paid-for titles to any
material degree), Johnston said that this indicated that the question was open to the interpretation by all
advertisers that the alternative media were more effective rather than less effective.
96. The questions to which Johnston has referred in the preceding paragraph asked whether different
publications were equally effective compare with free weeklies. When the balance of respondents
disagrees with the proposition of equal effectiveness this, on its own, does not indicate whether they
believe that the alternatives are more or less effective. However, it is possible in some cases to interpret
the responses based on additional knowledge.
97. With respect to the use of different daily titles, Johnston said that respondents appeared to have
interpreted daily newspapers as referring to national daily titles. Johnston also said that in most of the
local areas surveyed there was not a paid-for weekly, but only a daily, and so the reported high usage of
weekly paid-for titles must be wrong. However, Table 1 shows that the sampling frame included the top
advertisers in 33 titles, of which 13 were paid-for weeklies, and Table 3 shows that advertisers can target
more than one local area, and so it is not obvious that the responses were wrong. Evidently, Table 3 does
not imply that advertising in one locality is a substitute for advertising in another locality.
98. Reported discounts were also apparently lower than those actually given, and the parties sug-
gested that advertisers did not necessarily know that they received a discount, particularly when they said
they never got a discount. As Figure 1 shows, those who said that they did not know what discount they
obtained have an advertising expenditure distribution that is closer to businesses that reported discounts
of 24 per cent or less than to business reporting higher discounts. But rather than taking individual
estimates, we have referred to the relative proportion of different types of advertiser that reported
discounts above or below 25 per cent as being indicative of those who fared better or worse. However,
citing the example of the results for estate agents, Johnston believed that the majority of ‘don’t know’
responses were given by advertisers that achieved large discounts and that any comparison of reported
results with respect to discounts was therefore likely to be unreliable and meaningless.
99. The parties identified an anonymous respondent whose answers were said to be inconsistent. One
apparent inconsistency was that the respondent did not mention the use of free local titles when
unprompted, but when prompted said three free weeklies were used; moreover although the respondent
usually negotiated and obtained discounts, it did not know the average discount. These answers should be
compared with the answers concerning paid-for titles. When unprompted the respondent said that paid-
for regional and local newspapers were used and that they accounted for about 70 per cent of budget, and
when prompted said that nine or more paid titles were used. It was usual to negotiate discounts, and as
often as not discounts were obtained, at an estimated average of 20 to 24 per cent. It would appear that
the respondent thought more about and knew more about the paid-for titles than the free, but this
demonstrates inability to recall less important information when unprompted rather than inconsistency.
Although advertising-only publications had not been mentioned by name as a type of advertising
medium that was used, when prompted the respondent said that they were used.
199
100. We estimated the level of inconsistent answers as follows. If we assume that all respondents are
fully attentive to their use of free titles, so that any discrepancy in response is assumed not be from lack
of recall, the level of inconsistency in responses will be revealed by differences in answers to questions
about the use of free newspapers. 76 per cent of respondents said that they used free newspapers when
unprompted and, when prompted to recall their use of free weeklies, only 72 per cent of respondents said
that they used one or more titles. Ignoring the fact that some of these respondents would say that they did
not know, this suggests a level of inconsistency in response of around 5 per cent.
101. However, Johnston commented that the discrepancy between the percentage of respondents
who said that they used advertising-only publications when unprompted on the one hand, and when
prompted on the other, was much more significant (more than three times as many). Clearly, if the level
of inconsistency of respondents was of this magnitude (which could not be ruled out), this raised serious
issues as to the reliability of the survey results as a whole. It argued that it was misleading for the CC to
simply rely on the discrepancy in responses in relation to free newspapers.
102. It is instructive to see how a lack of recall by any respondent affects our analysis. If free local
newspapers were not mentioned when unprompted, ORC did not ask about spend in this category. Thus
the median spend on free local newspapers given in Table 4 was computed only from those who recalled
that they used this medium. Those who needed to be prompted about their use—and whose answers
would be less accurate—were not included. Similarly, where the estimated discount question prompted a
‘don’t know’ response this does not affect our comparison of reported discounts below and above 25 per
cent. As noted above, however, Johnston has argued that a majority of those advertisers who said they
did not know what discount they obtained were, in its view, likely to have received substantial discounts.
103. ORC, who conducted the survey on behalf of the CC, delivered a written report and an oral
presentation with visual aids. The main messages from its presentation have been included in the analysis
above. The next section is the report on the survey as written by ORC.
ORC report on the survey
1. Introduction
104. There is a proposal for transfer to Johnston Press of eight local newspaper titles in the East
Midlands currently published by Trinity Mirror. Research was required to:
— understand current advertising behaviour of advertising agencies and local businesses with
respect to media in general and local newspapers in particular;
— perceptions regarding advertisement effectiveness in local free weekly newspapers and paid-for
newspaper titles;
— prevalence of negotiation of discounts and the extent to which discounts can be negotiated; and
— reactions to a hypothetical 10 per cent increase in advertising rates.
105. Telephone interviews of approximately 15 minutes’ duration were conducted among the top 30
advertisers in 33 of the relevant newspaper titles in three areas. The sample list of contact names was
provided by the CC. Fieldwork was conducted between 2 and 18 January 2002 and a total of 476
interviews were finally achieved.
106. The following is a summary of the key findings of the research.
2. Sample profile
107. Quotas were set by region and each region was meant to have equal representation. The final
sample was as follows:
— Peterborough (25 per cent);
200
— Northampton (31 per cent);
— Derby (26 per cent); and
— Others (19 per cent).
108. Interviewers were instructed to begin their interviews with the standard ORC introduction. If
respondents queried where their name was obtained from then the CC was mentioned. 78 per cent of all
respondents were aware that the study was conducted on behalf of the CC whereas 22 per cent of
respondents were unaware of the CC’s role.
109. 84 per cent of those interviewed comprised businesses and the remaining 16 per cent were
advertising agencies.
110. 81 per cent of the businesses advertised regionally/locally while 19 per cent had a national
coverage as well.
111. 62 per cent of the advertising agencies had a national spread and the remaining 38 per cent
advertised locally/regionally. Advertising agencies in Northampton and Derby had a skew towards
national coverage whereas those in Peterborough had an equal split between national and regional/local
coverage.
112. In terms of business types, the biggest representation was that of estate agents (24 per cent),
followed by retailers (22 per cent) and motor repairers/dealers (19 per cent). The remaining business
types had a single digit representation.
113. The mode for annual turnover of businesses in the local area where they advertised was
£1 million to less than £10 million (17 per cent) and the median was £500,000 to £1 million. 46 per cent
of the businesses reported ‘don’t know’ which could be interpreted as a reluctance to reveal their
turnover.
114. 60 per cent of advertising agencies claimed ‘don’t know’ when asked for the annual turnover of
the organization in the local/regional area where they advertised. 13 per cent (mode), however, claimed
that it tended to be around £1 million to less than £10 million.
3. Advertising behaviour
Advertising spend
115. Both the median and mode for total advertising expenditure of businesses is £10,000 to
£25,000. The median spend for motor dealers/repairers tended to be higher—between £25,000 and
£50,000.
116. Half the advertising agencies (48 per cent) that had a local/regional spread had an annual
advertising expenditure of less than £25,000. In comparison, half the agencies with a national coverage
(52 per cent) had an advertising spend of less than £1 million.
117. 45 per cent of advertising agencies claimed a ‘don’t know’ when asked to estimate the total
expenditure in local areas in advertising in all forms of media in the past year. Another 13 per cent
(mode) estimated the expenditure to be around between £100,000 and £1 million.
Paid-for regional and local newspapers vs free regional and local newspapers
118. At an unprompted level respondents were asked to name different media in which they
advertised. Paid-for regional and local newspapers seemed to be the most common medium for
advertising spend with 85 per cent of the respondents claiming to be spending on it. The second most
popular medium appeared to be free regional and local newspapers, with 76 per cent of respondents
claiming to spend on it. The gap seems to widen between the second and the third most popular medium
201
(local radio) with 28 per cent of respondent spend. In comparison, 11 per cent of respondents used
national newspapers; 10 per cent used classified directories; and 6 per cent used advertising-only
publications.
119. The two business types which tend to buck the above trend are motor dealers (89 per cent paid-
for regional vs 94 per cent free regional and local newspapers) and trades such as builders, plumbers, etc
(70 per cent paid-for regional vs 89 per cent free regional and local newspapers).
120. Among those who advertised in paid-for regional and local newspapers and free regional and
local newspapers the median spend in the relevant medium was 40 to 59 per cent of total spend. For
14 per cent of those who advertised in paid-for regional and local newspapers their entire advertising was
on this medium in comparison with 8 per cent in free regional and local newspapers. Recruitment
agencies (36 per cent) were most likely to spend their entire advertising budget on paid-for regional and
local newspapers, while leisure and consumer services (29 per cent) showed a greater likelihood of
spending their entire advertising budget on free regional and local newspapers.
Switching behaviour to/from newspapers into other medium
121. 62 per cent of respondents had several campaigns throughout the year in their local area and the
remaining 38 per cent had one plan for the whole year. While advertising agencies (92 per cent) and
recruitment agencies (85 per cent) were more likely to have several campaigns in a year motor dealers
(48 per cent), retailers (47 per cent) and estate agents (47 per cent) were most likely to have an annual
advertising plan.
122. Among those who planned several campaigns a year, it was not very common (68 per cent
mentioned rarely/very rarely) to switch more than 10 per cent of their advertising spend on newspapers
into other media while moving on from one campaign to another. For those who had an annual
campaign, the spend on newspapers tended not to vary year on year, with 78 per cent reporting
‘rarely/very rarely’. This leads us to believe that the advertising spend on newspapers is unlikely to be
channelled into other media.
123. A similar theory applies for shifting advertising spend from other channels into newspapers.
69 per cent of those who have several campaigns in a year and 81 per cent of those who had an annual
campaign reported that they rarely/very rarely switched more than 10 per cent of advertising spend on
other media into newspapers.
124. Hence, overall the budgeted spend on newspapers appears to be fixed and, all things remaining
the same, unlikely to be fuelled into other media.
Display and classified advertisements
125. 73 per cent of respondents claimed that their advertising spend was not split between display
and classified. Advertising agencies with national coverage and recruitment agencies ranked among
those who had an approximately equal split between those who claimed that their advertising spend was
split and not split between the two types of advertisements.
126. Among those who claimed their advertising was split between display and classified adver-
tisements the median spend on display ranged from 80 to 90 per cent. Estate agents (69 per cent) were
most likely to have 80 to 99 per cent of their total advertising spend on displays.
Number of different types of newspaper titles used
127. The mode for the number of free weekly newspaper titles and paid-for newspaper titles used by
respondents is 1. 22 per cent said that they did not use any free weekly newspaper titles and 16 per cent
claimed that they did not use any paid-for newspaper titles.
128. The majority of the respondents (63 per cent) said that they did not use any daily newspapers.
202
129. The number rises when it comes to Sunday newspapers with 84 per cent reporting non-usage.
130. Since the bulk of the sample comprised local business, it probably explains the non-usage for
daily and Sunday newspapers.
131. When specifically asked whether they used advertising-only publications, ie prompted, 20 per
cent of those interviewed responded in the affirmative. This reiterates the theory that respondents are
more likely to give higher ratings when prompted than at an unprompted level. Advertising agencies with
a national coverage (29 per cent) and businesses, which deal with motor dealers/repairers (36 per cent),
were more likely to advertise in such a publication.
4. Perceptions on advertising effectiveness in different types of newspapers
132. Only 13 per cent of those surveyed said that they had local paid-for Sunday titles in their area.
When asked whether advertising in local paid-for Sunday newspapers was equally effective as local free
weekly newspapers their opinions appear to be split, with 36 per cent disagreeing (disagree/strongly
disagree) and another 30 per cent agreeing (agree/strongly agree).
133. 82 per cent of respondents had local paid-for weekly titles. 50 per cent agreed (agree/strongly
agree) that advertising in a local paid-for weekly newspaper was equally as effective as advertising in a
local free weekly newspaper. Another 34 per cent disagreed with the statement (disagree/strongly
disagree).
134. 46 per cent of respondents disagreed (disagree/strongly disagree) that advertising in
advertising-only publications was equally as effective as advertising in a local free weekly newspaper. In
comparison, only 22 per cent seemed to agree. Advertising agencies appeared to have the strongest
opinion, with 57 per cent of them disagreeing with this statement.
135. Opinions on advertising in local paid-for daily newspapers being equally effective as that in
local free weekly newspapers appears to be split with 33 per cent agreeing and 28 per cent disagreeing.
Trades, ie builders, plumbers, etc, appear not to follow the above rule, with a greater proportion who dis-
agreed (30 per cent) in comparison with 19 per cent who agreed.
136. Hence perceptually the two newspaper types, which are similar in terms of advertising
effectiveness, are local paid-for weekly newspapers and local free weekly newspapers (greater number of
those who agree with the statement than disagree).
5. Negotiation of discounts
137. The numbers of those who asked for alternative price quotes from free weeklies, paid-for
weeklies and advertising-only publications to compare prices is more or less equally split among those
who claimed it was usual (47 per cent) and those who claimed it was unusual (42 per cent). Motor
dealers/repairers (67 per cent), advertising agencies (66 per cent) and recruitment agencies (62 per cent)
were most likely to be following this behaviour. Businesses which advertised locally/regionally (49 per
cent), estate agents (59 per cent) and leisure and consumer services (71 per cent) ranked among those
who said that it was unusual for them to ask for alternative price quotes.
138. Negotiation of discounts appears to be a popular exercise with 60 per cent and 58 per cent
claiming that it was usual for them to negotiate rates or discounts off the rate card with free newspaper
titles and paid-for newspaper titles respectively. Over a third in each category (39 per cent for free
newspaper titles and 36 per cent for paid-for newspaper titles) claimed that this kind of negotiation was
‘very usual’.
139. In comparison, 22 per cent and 26 per cent claimed that it was unusual (unusual/very unusual)
to negotiate discounts off the rate card with free newspaper titles and paid-for newspaper titles
respectively. The groups which did not ask for alternative price quotes were also those groups which
were unlikely to negotiate discounts.
203
140. Those who tended to negotiate discounts also appeared to be successful, with 58 per cent and
59 per cent claiming that it was usual for them to be successful in obtaining a discount off the rate card
for free newspaper titles and paid-for newspaper titles respectively.
141. 18 per cent and 21 per cent say it was unusual for them to be successful in obtaining discounts
off the rate card for free newspaper titles and paid-for newspaper titles.
142. A quarter of respondents (26 per cent) reported that they got 25 per cent or more off the rate
while negotiating a discount off the rate card when using free newspaper titles. 42 per cent preferred to
respond with a ‘don’t know’, while 8 per cent claimed that they never got a discount.
143. In comparison with free newspaper titles, for paid-for newspaper titles there appeared to be no
difference in terms of the percentage of discount they were able to negotiate (24 per cent obtaining
25 per cent or more; 8 per cent who never get a discount and 38 per cent who ‘don’t know’).
Willingness/unwillingness to give discounts
144. 67 per cent and 68 per cent respectively of respondents were of the opinion that local free
weekly newspapers and local paid-for newspapers were willing to give discounts off the rate card
charges. This probably explains why it was acceptable for them to ask and be successful in negotiating
discounts.
145. Among the reasons cited for willingness to give discounts, the most common are as follows:
— for greater business (49 per cent);
— to maintain the functioning of the business (25 per cent); and
— to outdo competition (13 per cent).
146. The common reasons cited for unwillingness to give discounts are as follows:
— operating in a monopoly situation and hence do not need to attract customers as customers will
come in any case (31 per cent); and
— demand for advertising space higher than supply (6 per cent).
6. Reactions to a hypothetical increase in advertising rate
147. When asked what they would do if all the local weekly papers in their area were to put up their
advertising rates by 10 per cent, the following was the reaction of the respondents:
— A third (30 per cent) were unconcerned and said that they would do nothing.
— 14 per cent said that they would switch advertising to another medium. Retailers were more
likely to do this.
— 14 per cent said that they would negotiate a price. Those who were currently more likely to be
negotiating prices, namely advertising agencies, motor dealers/repairers, recruitment agencies
and retailers, were more likely to follow this measure.
— 9 per cent reported that they would cut back on advertising.
— 7 per cent would switch advertising to another local paper.
— 7 per cent would cut out local weekly paper advertising.
204
The questions
148. The interview began with a preamble in which the interviewer introduced him- or herself,
saying that they were calling from ORC to ask about the way the organization used advertising media. If
asked, the interviewer said that a particular newspaper had given their name and telephone number when
the CC asked for a list of major advertisers. The interviewer went on to explain that there was a proposal
to transfer the ownership of certain free local newspapers in an area where they advertised, and that the
CC had been asked to look into this and to decide if this would be against the public interest.
149. There followed a number of screening questions to establish that the person was responsible or
knowledgeable about advertising decisions, or if necessary arrange a time to call back. In Table 14, a
‘qualifying respondent’ is a person responsible or knowledgeable about advertising decisions. In
practice, only those who were qualifying respondents were retained in the final data analysis.
TABLE 14 The questions used in the advertiser survey
Q1. Record if interviewer mentioned the CC. Base: All respondents
Q2. Are you able to speak for the organization on Base: All respondents
advertising decisions?
Q3. Before I begin, can I reassure you that your Base: All qualifying respondents
views will be anonymous and combined with
those of other advertisers in a report? May I take
about 15 minutes of your time?
Q4. Could you tell me if you advertise your own Base: All qualifying respondents
business locally/regionally or nationally, or are
you an advertising agency?
Q5. Would you say that yours is a national, regional Base: All who are an advertising
or local agency? agency
Q6. Can you confirm your organization’s type of Base: All who advertise own
business please? business
Q7. My questions will refer to the organization or Base: All qualifying respondents
business outlets where you have responsibility
over the local advertising in the town or county.
Can you identify the county or town localities
that you aim to cover when advertising this
organization? Is it (READ OUT) …
Q8. Can you estimate what is the annual turnover of Base: All who advertise own
the organization in the local areas (that is the business
town or county) that you advertise in?
Q9. Can you estimate the total expenditure of the Base: All who advertise own
organization in the local areas on advertising business
communications in all media in the last 12
months? Please exclude promotional
expenditure.
Q10. Can you estimate what is the annual turnover of Base: All who are an advertising
the organization or organizations in the local agency
area (that is the county or town) that you
advertise in?
Q11. Can you estimate the total expenditure in the Base: All who are an advertising
local areas on advertising communications in all agency
di i th l t 12 th f th
205
media in the last 12 months for these
organizations? Please exclude promotional
expenditure.
Q12. Could you now please identify all the various Base: All qualifying respondents
ways you might have spent this expenditure on
advertising communications. DO NOT READ
OUT
Q13A. Can you estimate what per cent of that total Base: All identifying free regional
spend in the past 12 months went on ‘free and local newspapers
regional and local newspapers’?
Q13B. Can you estimate what per cent of that total Base: All identifying paid-for
spend in the past 12 months went on ‘paid-for regional and local newspapers
regional and local newspapers’?
Q13C. Can you estimate what per cent of that total Base: All identifying national
spend in the past 12 months went on ‘national newspapers
newspapers’?
Q13D. Can you estimate what per cent of that total Base: All identifying consumer
spend in the past 12 months went on ‘consumer magazines
magazines’?
Q13E. Can you estimate what per cent of that total Base: All identifying business
spend in the past 12 months went on ‘business magazines
magazines’?
Q13F. Can you estimate what per cent of that total Base: All identifying advertising only
spend in the past 12 months went on publications
‘advertising only publications’?
Q13G. Can you estimate what per cent of that total Base: All identifying classified
spend in the past 12 months went on ‘classified directories
directories’?
Q13H. Can you estimate what per cent of that total Base: All identifying local radio
spend in the past 12 months went on ‘local
radio’?
Q13I. Can you estimate what per cent of that total Base: All identifying local TV
spend in the past 12 months went on ‘local TV’?
Q13J. Can you estimate what per cent of that total Base: All identifying the Internet or
spend in the past 12 months went on ‘the Web
Internet or Web’?
Q13K. Can you estimate what per cent of that total Base: All identifying direct mail
spend in the past 12 months went on ‘direct mail leaflets
leaflets’?
Q13L. Can you estimate what per cent of that total Base: All identifying outdoor
spend in the past 12 months went on ‘outdoor advertising
advertising’?
Q14. Do you plan your advertising in the local area Base: All qualifying respondents
with several campaigns throughout the year, or
do you have one plan for the whole year?
Q15A1 When moving from one campaign to the next, Base: All who plan advertising with
. how often have you switched more than 10 per several campaigns per year
206
. cent of your spend on newspapers into other several campaigns per year
communication media?
Q15A2 When moving from one year to the next, how Base: All who plan advertising with
. often have you switched more than 10 per cent one plan for the whole year
of your spend on newspapers into other
communication media?
Q15B1 When moving from one campaign to the next, Base: All who plan advertising with
. how often have you switched more than 10 per several campaigns per year
cent of your spend on other communication
media into newspapers?
Q15B2 When moving from one year to the next, how Base: All who plan advertising with
. often have you switched more than 10 per cent one plan for the whole year
of your spend on other communication media
into newspapers?
Q16. Is your advertising spend split between display Base: All qualifying respondents
and classified?
Q17. What percentage spend goes on display? Base: All respondents whose
advertising spend is split between
display and classified
Q18A. How many free weekly newspaper titles Base: All qualifying respondents
(excluding advertising-only publications) do you
normally use?
Q18B. How many paid-for weekly newspaper titles Base: All qualifying respondents
(excluding advertising-only publications) do you
normally use?
Q18C. How many daily newspapers do you normally Base: All qualifying respondents
use?
Q18D. How many Sunday newspapers do you normally Base: All qualifying respondents
use?
Q19. Do you use any advertising-only publications? Base: All qualifying respondents
Q20A. Can you tell me how many of the advertising- Base: All respondents using
only publications you normally use specialize in advertising only publications
jobs?
Q20B. Can you tell me how many of the advertising- Base: All respondents using
only publications you normally use specialize in advertising only publications
property?
Q20C. Can you tell me how many of the advertising- Base: All respondents using
only publications you normally use specialize in advertising only publications
motors?
Q20D. Can you tell me how many of the advertising- Base: All respondents using
only publications you normally use specialize in advertising only publications
classified?
Q21. Do you have local paid-for Sunday titles in the Base: All qualifying respondents
area?
Q22. Would you agree or disagree that advertising in Base: All respondents who have
a local paid-for Sunday newspaper is equally as local paid-for Sunday titles in the
207
effective as advertising in a local free weekly area
newspaper?
Q22X. Do you have local paid-for weekly titles in the Base: All qualifying respondents
area?
Q23. Would you agree or disagree that advertising in Base: All respondents who have
a local paid-for weekly newspaper is equally as local paid-for weekly titles in the
effective as advertising in a local free weekly area
newspaper?
Q24. Would you agree or disagree that advertising in Base: All qualifying respondents
advertising-only publications is equally as
effective as advertising in a local free weekly
newspaper?
Q25. Would you agree or disagree that advertising in Base: All qualifying respondents
a local paid-for daily newsaper is equally as
effective as advertising in a local free weekly
newspaper?
Q26. Over the past 12 months … How usual was it for Base: All qualifying respondents
you to ask for alternative price quotes from free
weeklies, paid-for weeklies or advertising-only
publications, to compare prices?
Q27. Over the past 12 months … How usual was it for Base: All qualifying respondents
you to negotiate rates or discounts off the rate
card with free newspaper titles?
Q28. Over the past 12 months … How usual was it for Base: All qualifying respondents
you to be successful in obtaining a discount off
the rate card charges for free newspaper titles?
Q29. Could you perhaps estimate your average Base: All qualifying respondents
discount off the rate card when using free
newspaper titles?
Q30. Do you think that the local free newspapers are Base: All respondents who don’t
willing to give discounts off the rate card know or never get discount from
charges? free newspaper titles
Q31. Over the past 12 months … How usual was it for Base: All qualifying respondents
you to negotiate rates or discounts off the rate
card with paid-for newspaper titles?
Q32. Over the past 12 months … How usual was it for Base: All qualifying respondents
you to be successful in obtaining discounts off
the rate card charges for paid-for newspaper
titles?
Q33. Could you estimate your average discount off Base: All qualifying respondents
the rate card when using paid-for newspaper
titles?
Q34. Do you think that the local paid-for newspapers Base: All respondents who don’t
are willing to give discounts off the rate card know or never get discount from
charges? paid-for newspaper titles
Q35. Why do you think they are they willing or not Base: All qualifying respondents
willing to give discounts off the rate card
charges? DO NOT READ OUT
208
Q35A. Why do you think they are willing to give Base: All saying ‘Willing’
discounts off the rate card charges?
Q35B. Why do you think they are not willing to give Base: All saying ‘NOT Willing’
discounts off the rate card charges?
Q36. Finally, if all the local weekly papers in the area Base: All qualifying respondents
where you advertise were to put up their
advertising rates by 10 per cent, what would you
do? DO NOT READ OUT/DO NOT PROMPT
209
Related docs
