The two key auditors of magazine and newspaper readership, the Audit Bureau of Circulations and the National Readership Survey, are exploring plans to appoint a joint chief executive, as the top posts at both organisations free up later this year.
They are at pains to stress that discussions are tentative - "talks about talks", says NRS managing director Roger Pratt - yet the two bodies have spotted a clear opportunity to bring their workings closer together. A joint chief executive would accrue cost savings, but might it also be the driver to establish a single, universal readership metric that combines the robustness of the ABC with the demography of the NRS?
The proposals have been welcomed, broadly, by media buyers, although most are more concerned with ironing out discrepancies in existing ABC and NRS data.
Alex Randall, press director at Vizeum, explains: "A joint chief executive is potentially a good idea, but what press buyers really want is a readership measure that is robust and accurate. If a joint chief executive takes us there, then fantastic, but it doesn't guarantee change."
According to Randall, the bigger headache for buyers is that ABC and NRS data often fail to tally in that they can show circulation and readership figures moving in different directions.
Conflicting data
He believes the problem lies in the methodology of the NRS, which relies on sample data blown up to represent the population as a whole. It provides an estimate of the number of readers and the type of reader in terms of age, sex and other demographic, and lifestyle characteristics. The ABC, in contrast, measures the number of copies actually sold.
Nik Vyas, group press director at ZenithOptimedia, agrees. He says: "In terms of a joint chief executive, we'd be interested in hearing what benefits they think this will yield for both companies, as well as stakeholders. NRS has substantive issues to resolve, so we'd want reassurances that a wider remit wouldn't lead to a dilution of focus in resolving these."
No one is talking about combining the two, radically different sets of data. As Chris Boyd, chief executive of the ABC, notes, that would be like adding apples and elephants. But there is a call for some way of rationalising the numbers, so that the inter-relationship between the two currencies is more apparent.
And, since the NRS data is inherently and notoriously prone to fluctuations, it tends to attract most criticism.
"The discrepancy between NRS and ABC data has always been a problem," concedes Pratt. "It is irritating and frustrating and not to do with the ABC; it is because we are a sample-based exercise."
The only way to minimise the problem, he says, would be to substantially increase the size of the sample - currently 36,000 people - but that would be hugely expensive: "In order to halve the sample variation, we would have to quadruple the sample size. That's just not an option."
The NRS is, however, looking into ways of "smoothing" its data to flatten out peaks and troughs. But that has its own disadvantages, since it means going back further in time to produce an average of past estimates, therefore older data.
And the ABC has not escaped criticism entirely. Steve Auckland, head of Associated Newspapers' free newspapers division, insists the ABC should have more flexible auditing criteria to better reflect advertisers' needs. "Some supplements don't meet ABC criteria, so they get discounted," he says. "But if agencies and advertisers are happy that there's a proper audit trail, then why should they not be included?"
Greater clarity
ZenithOptimedia's Vyas wants the ABC to demand greater transparency of publishers. "At the moment, we get very generic, vague headline figures and we need greater clarity on the layers beneath this very thin veneer," says Vyas. "Magazine publishers are making welcome moves in this area, but newspapers are, at best, dragging their feet."
Another common cry from press buyers, and one that transcends anomalies in ABC and NRS data, is for research models that take into account online audiences, giving an idea of publishers' overall reach, but without duplication.
At the moment, Guardian News and Media is the only publisher to have created such a tool, called UK Total Audience, which gives a trusted measure of net print and online newspaper audiences across its titles.
Incorporating site-centric, as well as user-centric data, is something the ABC and NRS are investigating, some say too slowly. Agencies want combined figures as well as separate print and digital data, so they can draw their own conclusions.
Vyas says: "Publishers are moving rapidly towards a brand sell, but we need more data, so we can effectively de-dupe their headline audience figures."
ABC AND NRS DATA PROBLEMS
Time Lag: The ABC newspaper data and ABCe data are reported monthly, the ABC magazine data every six months and NRS data every six months. Magazine ABCs are released before the NRS figures
Flexibility: The ABC data does not cover supplements
Bias/vested interests: Some say the NRS survey favours fame over readership, since those surveyed more easily recall better-known titles. They also lie about their reading habits
Sample variation: The NRS data is based on interviews with a sample of the adult population, so results differ from sample to sample. This can lead to significant fluctuations. Increasing the sample size is difficult and costly
Discrepancies in the data sets: It is not unheard of for circulation figures for a particular title to go up and readership data to go down, or vice versa
Transparency: Planners want more insight into the numbers, for both ABC and NRS data. They want greater clarity beyond the headline figures.