Challenging the numbers

More people are now saying what no one dared to say before, that the emperor has no clothes. The "emperor" in question is the organisation (or organisations, since there is more than one of them) that does readership surveys every six months, in order to gauge which publication has how many readers""an issue of vital importance to the Rs 15,000 crore advertising business, and all the businesses that promote their products and services by spending that ad money. The readership surveys have produced increasingly quirky results, prompting comparisons with election surveys that prove to be off the mark, and it should surprise no one that the protests are mounting. The latest to do so is the Indian Newspaper Society, a body of publishers, which is so incensed by the findings of the latest Indian Readership Survey (produced by the Media Research Users' Council) that it has decided to do its own surveys from now on. The INS argument is that the IRS finding, that newspaper readership is declining, sits badly on numbers that show growing copy sales. Its question is: if there are fewer readers, why are they buying more copies of newspapers?

This is only the latest instance of survey findings being questioned. A couple of months ago, the rival National Readership Survey came out with numbers that had many publishers up in arms. In the past, protesting publishers have even gone to court on an issue that is vital to them because advertising accounts for between 50 per cent and 90 per cent of most publications' revenue. Embarrassed by the sometimes hopelessly contradictory findings of the IRS and NRS, both of which claim authentic methodologies, some moves were made recently to scrap the rival surveys and get onto a common platform, which would then eliminate the problem of contradictory numbers. But that would be like brushing the dirt under the carpet. The plain truth is that readership surveys are unreliable. And since the advertising agencies and media-buying houses plan their ad campaigns by using software packages that feed on these survey numbers, everyone uses the numbers that suit him best.

The alternative to readership surveys is the same as the alternative to election surveys: count the actual copies sold, and votes cast. The problem, however, is that the "audit" of copies sold (done by the Audit Bureau of Circulations, which is a body of publishers, advertisers and advertising agencies) uses straitlaced rules on trade margins and such like that prevailed in the publishing business once upon a time, and which have come under challenge in an increasingly competitive marketplace where publishers even give away copies of newspapers for free. In the process, many leading publishers have failed to pass the ABC's audit norms. Magazine publishers as a sub-group have publicly aired their unhappiness with ABC rules, which, in their view, do not reflect reality, and have banded together into a separate body that seeks to find alternative solutions. Faced with this challenge, the ABC has become more flexible about its norms, but advertisers continue to be more interested in readers than copy sales numbers, so the survey organisations remain in business, even though the reader-per-copy number may vary by many multiples between different surveys (for business newspapers, it has varied all the way from less than one reader per copy to as much as six readers per copy!). With more advertising money riding on the issue than ever before, the only solution seems to be to let a hundred surveys and audits bloom; and advertisers can choose what they find more credible""except that one survey's truth is the next survey's lie.