MARKET RESEARCH: CAN MEDIA RESEARCH KEEP UP? Simon Marquis examines whether the media’s research engines are able to give what this dynamic industry needs

There is rumbling disquiet about media research. But then most people in advertising can hardly remember a time when there wasn’t rumbling disquiet about media research. Only three months ago, Campaign ran a conference called ’Media Research - Time to Reassess’, because, well yes, it was time to reassess ... again.

There is rumbling disquiet about media research. But then most

people in advertising can hardly remember a time when there wasn’t

rumbling disquiet about media research. Only three months ago, Campaign

ran a conference called ’Media Research - Time to Reassess’, because,

well yes, it was time to reassess ... again.

Media research is one of those arcane areas of advertising that cries

out for repeated reassessment. Most media research - the expensive joint

industry stuff anyway - is born of committees and multi-interest boards,

nursed by technical experts and bickered over by agencies, advertisers

and media owners. Widely accepted to be technically sophisticated, the

major surveys are nevertheless masterpieces of compromise. Little wonder

that at any one time someone or other is crying out for a


Media research needs reassessing for other reasons. Most client

marketers, perhaps understandably, regard the detail of the BARB

specification, or the latest NRS rumpus, as things they pay their

agencies to deal with.

Even interesting conferences on media research attract precious few


It seems to need constant reassessment just to stop most people nodding


But by far the most compelling reason to re-evaluate our shared industry

research is that changes in the media world put their systems under new

pressures, and raise new questions with their users - both planners and

buyers. For example, the arrival of digital TV services this autumn will

put new strains on BARB as the existing methodology it uses cannot cope

with digital transmission. Digital TV may, at the start, be tiny

compared with analogue, but if we take the not unreasonable view that in

ten to 15 years’ time all television sets will be digital, then we must

ensure that our principal research tool is up to measuring it.

The arguments in favour of maintaining the JICs (as the joint industry

media research surveys are commonly known) centre around the need for a

currency with which to trade advertising airtime and space. Like

monetary currencies, they must have a degree of intrinsic worth or they

risk becoming so debased that the marketplace would have to invent new


Some of the JICs have, arguably, too much intrinsic worth already -

there is hardly a media buyer who will not occasionally question the

need for such a quantity of data. If the main objective of television

advertising is to make effective contact with consumers, it is hard to

know why we need minute-by-minute viewing data for no less than 56

audience sub-groups. This has everything to do with programme evaluation

and airtime negotiation but very little to do with advertising


BARB is one of the world’s most comprehensive media research surveys.

The question its controllers must continue to address is whether it

meets the needs of its users, especially in a period of such rapid


As audiences fragment more and more, commercial breaks notch up audience

levels so small that BARB cannot make anything more than a stab at

accurately measuring them. A recorded rating of 3.5 in a small ITV

region, for example, could - allowing for statistical variance -

actually be a rating of 0.1 or it could be a rating of 7.9. In an

industry obsessed with measurement, this sort of margin for error is


Sensibly, there is a growing view that what we need from BARB is

accurate, national data for all the main sub-groups to which factors can

be applied region by region. Similarly, it is clear that the quantity of

information about the biggest channels could be scaled down in favour of

boosting the samples on the growing number of small or specialist


Other media are facing big challenges too. In press, the NRS is the

long-established gold standard for space selling and buying. This year

it has taken a quantum leap forward with the QRS, which asks new and

different questions about readership. Specifically these provide more

detail about magazine reading and actual exposure to pages. This

development brings press research more into line with television but it

does not complete the picture of how press advertising works.

In radio, Rajar has made significant improvements and the challenge will

be to cope with the advent of digital.

The most interesting innovation here is the portable meter - a form of

wristwatch that records which station is being listened to. The gadget

is being developed jointly by IPSOS-RSL and Infratest-Burke and it is

possible that these meters will be tested within the next few years.

Rajar has also put all its data on to a website making analysis much


In outdoor as well, media research has improved. Oscar II, launched by

POSTAR in 1996, provides much more credible OTS figures than were

previously available and further refinements are in the offing.

It is easy to sneer at the JICs, but improvements have been


Media planners complain that the JICs provide the wrong sort of data -

or rather, not enough of the right sort.

What they need is not ever-more accurate counting of heads, but deeper

insights into how people actually use their media and how ads work

within a media context. This isn’t something the JICs can begin to


So, what sort of media research does modern media planning require?

Dominic Owens, until recently the head of BT’s business division

advertising, has made an articulate call for a small levy on advertising

expenditure to create a new pot of research money. This could be used to

fund an agreed programme of both quantitative and qualitative studies

that would help us learn more about how the media work. Studies might

include: in-car radio listening, readership of the various sections of a

newspaper or the effect of position in an ad break on the actual viewing

of commercials.

ISBA is keen to encourage this initiative, but there is a real danger

that advertisers will fail to agree the priorities. Others may believe

it is a good idea but will be reluctant to pay when it comes to the

crunch. It is one of the oddities of marketing in the UK that client

companies will unhesitatingly commit to major brand and advertising

research studies but baulk at the idea of spending on brand-specific

media research. In truth, there are many excellent studies produced by

the media themselves, but in the end it comes down to a question of


You get what you pay for.

It is difficult to see where a resolution lies. Advertising is one of

the most effective competitive weapons at a company’s disposal. Great

creative ideas sell products and build brands. Rubbishy creative ideas

don’t. It is the same with media. Cunning, relevant, prominent,

inventive use of media can give what Richard Eyre, the chief executive

of ITV, once called ’an unfair advantage’ to the ads. Research that

reveals something previously unknown about consumer behaviour is a

competitive weapon too, particularly if it is brand-specific. It may

turn out to be difficult, therefore, to persuade advertisers that there

is much to be gained by jointly financing new research initiatives.

They’d sooner keep the good stuff to themselves.

Having said that, the world is moving on. The JICs must - and almost

certainly will - find ways of responding to the new challenges of

digital, of new sections and titles, of new outdoor opportunities. With

so much at stake for all parties, no-one can think of a better solution

than the JICs for sharing an understanding of what the media deliver.

But we must remind ourselves that they will only tell part of the story,

and that we will have to pay for the rest of it.


Name                  BARB

Contractor            RSMB and AGB

Annual cost           Approx. pounds 10 million

Method                In-home peoplemeters

Sample                4,500 homes

Output                Minute-by-minute viewing figures for 56 sub-groups

                      - electronic reporting

Future challenges     More than 200 digital TV channels

                      Niche broadcasting sample sizes

                      Picture matching

                      Passive meters

                      Data overload

                      Channel factoring

                      Identifying digital signatures

                      Live and consolidated data

                      Factored time shift viewing

                      Recruitment of niche targets


Necessary changes     At some point in the future, digital TV will

                      require reporting

                      How digital will be measured due to small sample

                      sizes for niche channels remains to be seen

                      Users could suffer from data overload

                      Some form of two-tier system will have to be

                      worked out, one for mass broadcasting and one for

                      smaller, tightly targeted channels

Source: Zenith Media.


Name                  Rajar

Contractor            RSL

Annual cost           pounds 10 million over four years

Method                Self-completion diaries, aided recall

Sample                160,000 annually

Output                Quarterly summary for larger population ILR’s,

                      bi-annual for smaller ILR’s

Future challenges     Increasingly complex diaries

                      Implementing electronic accountability

                      The launch of digital radio and the volume of


                      The search for electronic metering

                      The cost of digital hardware

Necessary changes     Measurement of the effect of analogue listening

                      when digital begins transmission

Source: Zenith Media.


Name                  NRS (the main survey, others such as Banner, QRS,

                      BBS also exist)

Contractor            IPSOS-RSL

Annual cost           pounds 2.2million

Method                CAPI face-to-face interviews

Sample                33,000 interviews

Output                Quarterly computerised update. Bi-annual volumes

Future challenges     Quality of reading questions

                      Reduction of minimum age of respondents

                      Scottish NRS

                      Average page exposure

                      First time read

                      Time lag reading

                      Day of week readership

                      Newspaper sections reading - AIRs

Necessary changes     The call for more depth of reading - how, why and

                      where publications are read- is inevitable in the

                      light of the first QRS report

Source: Zenith Media.


Name                  POSTAR (the main survey, other surveys such as

                      Busads also exist)

Contractor            NOP

Annual cost           Unknown

Method                Interviewers conducting traffic and pedestrian


Output                Monthly electronic database updates

Sample                7,500

Future challenges     Continuing refinement of calculation of campaign


                      Re-classification of all UK sites

                      Continued provision of robust data

                      Creating a neural network to model travel


Necessary changes     Reporting data below TV region levels

                      Calculating visibility adjusted impacts for car


Source: Zenith Media.