Customer Research

Knowing your customers is a basic requirement. But in an industry where the letters page is the main channel of interaction, and the relationship is usually arm’s length via the newsagent, it's not that easy.

A plethora of research techniques has been developed to help newspapers analyse their readership. Liz McMahon, whose clients at BMRB include Metro and Associated Newspapers, says, "You've got a whole variety of different types of activity ranging from the highly quantitative - for instance JICWEB and other web measurement methods, or the National Readership Survey – to more qualitative methods." Much of this research is aimed at understanding the place of the newspaper in the media landscape. Reader research, on the other hand, goes deeper into the relationship between the publication and its readers.

Though most reader panels, in-house interviews and street surveys have a strong quantitative element, they allow publications to gain a more nuanced understanding of their readers, as do focus groups. Focus groups, observations, or online interviews can be used to ask direct questions about the publication, sometimes down to specifics such as how an individual story was covered. Rather than looking for statistics, reader research is looking for insights.

At the Norwich Evening News, for instance, research focused on finding out what younger readers felt about coverage by asking them to comment on particular stories in the paper. This contributed to the redesign of the paper.

Reader panels, according to Liz McMahon, are "something that a lot of papers have invested in, but to varying degrees". However, where the real difference lies is in the use that the publication makes of its reader panels – how efficiently the information derived from the panels is communicated and used within the organisation.

Commercial support

Ideally, the reader panel should inform both editorial and marketing. Katharine King, director of research at Metro, says, "You can use a panel for feedback on your publication, and we do do that, but a significant part of what we use the panel for is to inform our ad sales team."

This dictates the choice of questions that the panel will be asked. Metro has a number of different sections in its survey; some are editorial in focus, with newsy questions on whatever's topical to generate editorial and to guide product development. (In the US, Gannett has used an e-panel to develop fresh story ideas, using it to assess specific subjects such as parenting and finance.) Other parts of the survey focus on Metro's marketing, and advertisers are also given the opportunity to use the survey to ask their own questions.

Katharine King says the reader panel has been useful because "it feeds into so many different areas". The results can be turned into case studies which will feed into Metro's trade marketing activity, as well as being used editorially - stories about members' attitudes to topical issues are often dropped into the paper. The survey results also give the sales team an opportunity to go back and see advertisers with the results. But, she warns, for this to work properly, the research initiative needs to be embedded into the editorial and marketing operations.

Editorial support

Rachael Morgan, marketing manager at the Hull Daily Mail, agrees that editorial needs to be linked into the reader panel. "Using the panel," she explains, "we can see what people are talking about." That lets the paper pick up on matters of concern to its readers, and prioritise stories based on what readers have told the paper they care about rather than having to make an editorial judgment in a vacuum.

To some extent, reader panels are doing the same job as the web in general and user generated content; telling the editor what themes and stories are getting attention. On the web, newspapers can see which areas of their coverage are being read, which stories are generating blog links or comments, and what the readers think. However, while web analytics shows what readers are doing, it doesn't show why. Using reader panels, the newspaper - not the readers – remain in charge, and can ask the questions it wants answered.

Panel frequency

The frequency with which reader panels are used differs between papers. Some use reader panels only when they are either relaunching or launching new products. Bristol News and Media, for instance, admits to having done relatively little reader research recently as there have been no new launches to research.

Most local newspapers, though, are now using reader panels and focus groups as a part of normal operations, while stepping up the research effort significantly for new launches. Alex Leys, assistant editor at the Hull Daily Mail, who heads the online Your Mail operation, says, "We use reader panels for all our new publication launches, both online and in print." The paper has just revamped the entertainment section, and carried out extensive research for the purpose.

But, Alex says, "we also use surveys to tweak the product. We have a running panel of people we use to review current content, and look at our ongoing material as well as our new stuff."

Panel sizes

Panels need to be properly set up if they are going to work. Sample sizes need to be correctly assessed, for instance. Liz McMahon sees this as a delicate balancing act - "It's the balance between getting a reliable answer, and the budget that is available."

She points out that the size of sample needed depends on the publication's focus. The more tightly targeted it is – whether geographically or by subject – the lower the panel size needs to be. Metro has a 4-5,000 strong reader panel, with a response rate of 70-90 percent; a local weekly paper might only need to survey a few hundred readers.

She also points out that while many papers only survey the readership as a whole, some have decided to dig deeper and analyse different groups of readers separately. "The more you want to dig into different subgroups, the larger your samples need to be," she explains. For instance, some national papers now compare reader responses from different geographical areas, readers of different ages, and so on. It's even possible, if the reader panel data is good enough, to create specific surveys focused on smartphone owners, or women between 35 and 45.

At the Hull Daily Mail, the reader panels are much smaller than Metro's – 50 people, in groups of 9 or 10. Obviously, the area covered is much smaller. But, as Rachael Morgan notes, good information is still needed on each member of the group – age, location, family and employment status, at the very least. That enables the Mail to make sure the focus groups reflect the demographics of the area. Rachael says, "We make sure we have a representative sample to ensure the views are balanced." An unrepresentative sample could result in misleading views being passed on to the editorial department.

Liz McMahon points out that surveying non-readers can be equally interesting. BMRB is helping Associated Newspapers with this concept. "This is the most difficult job of all," she says - letting the newspaper know the opinions of people who don't take the paper, and being able to compare their responses with those of readers. "Non-readers give a benchmark against your readership and how they compare, and this gives you a feeling for groups that have strong potential for your brand and what you can do to engage them."

The Hull Daily Mail is doing this too – it used street surveys recently "particularly to get hold of people who don't interact with the paper," says Rachael Morgan.

Set up time

Newspapers shouldn't underestimate the time needed to set up a rolling panel. Metro's Urban Life project has been running for six years now. Members have to be recruited and selected, and the membership base refreshed from time to time (annually, in the case of Metro). Even if not all the panel members are to be interviewed regularly, newspapers still need a ready and receptive membership base. At the Hull Daily Mail, Rachael Morgan says, "We've created a massive bank of people we can go to, to stimulate debate in the community, for instance over crime and ASBOs."

Keeping it fresh

Katharine King points out that panels need to be "cycled" to ensure that new members are coming in to replace old ones. She says, "We kick about half the panel off each year." Within the 4-5,000 member panel, there are probably only a couple of hundred people left from the original panel now. Responses from new panel members are routinely compared to those of members who joined a year or more ago, to see if there is any difference between the two.

Many reader panels – though not all – offer a monetary compensation to members, whether in the form of a cash payment, discount vouchers, or entry to a prize draw. Metro, for instance, offers a prize draw for hundreds of £20 HMV vouchers as well as bigger cash prizes, while the Hull Daily Mail also offers financial incentives. But Rachael Morgan says that "while we give them a small amount of money as a thank you, that's not really why they come."

Katharine King says men tend to be more motivated by cash, while women tend to join the panel for fun. She believes that many panel members are motivated by seeing the results of surveys in Metro, which runs features about people on the panel, and also uses panel members' opinions in stories about topical issues such as the smoking ban.

She also points out that the survey needs to be interesting for the respondents. This may involve inserting some questions that are fun for the respondents to answer - but don't really contribute to your knowledge of the customer.

How much do focus groups change the newspaper's tactics? Quite a lot, according to Alex Leys. The Your Mail site was analysed by focus groups after two months of beta testing, and the message for management was that the design lacked impact and the site wasn't easy enough to navigate. It was redesigned and simplified, with wording and layout changed – in fact, says Alex, "We did a full revamp of all the artwork on the site. So the market research had quite a big impact on what was finally launched."

Reader input continued in the early months of Your Mail's operation. The site launched in February 2008 and by July, had already been fully [software] updated once. Interestingly, Alex believes newspapers aren't updating their websites nearly often enough – pointing to eBay, which updates its site once a week. And Alex says managers need to remain open-minded about the feedback they're getting from surveys – or they won't benefit from the investment. "If we ever got a resounding 'It's awful, we don't like it', we really would pull the product."

Golden rule

What are the worst mistakes made by research departments when they're using reader surveys and similar research techniques? According to Katharine King, the number one mistake is not having the buy-in from the people who are going to use it. She says, "There's no point doing research if it's not going to be implemented. Work out what you're going to do with it before you get the answers."

Editors are particularly important as there are two ways in which they can use the work of the panel. First of all, they can use the results to assess design issues, see how their existing editorial is playing with the readership, and even find new areas for editorial coverage. Secondly, they can help to ensure a high profile for the panel through featuring panel members and panel opinions in editorial.

Liz McMahon says the Metro editor is crucial as a figurehead for the panel and its work. "The results are fed back through Metro on a regular basis, to ensure readers know about it and that their contribution has meaning," she says.

The cost of running a panel can be high. Katharine King says that the Metro reader panel is "a significant investment for the company"; money has to be spent on recruiting readers, retaining them, planning the surveys, and processing information. Where focus groups rather than e-panels are used, facilitators have to be paid and rooms found for meetings.

Syndicated panels

Some newspapers decide to use syndicated panels in order to keep their costs down. However, panel members may not have the same sense of commitment that they do when on a panel for a single newspaper, and the responses will be openly available to other interested parties – meaning that the survey is no longer a major source of competitive advantage. Katharine King says, "You need to have something that's not a syndicated study that all the media agencies can see."

A halfway house is to share research teams with other newspapers. For instance, the research team that works for the Hull Daily Mail also works for three other papers in Lincoln, Scunthorpe and Grimsby. That enables them to share costs, while maintaining the advantage of being able to run the programme for themselves and create a sense of local belonging in the panel.

There's nothing startlingly new in the field of reader research. Reader panels have been around for a while – the Mirror began to run one in 2003, Metro in 2002, and they've been going for longer than that.

However, as in other areas of research, technology is increasingly changing the way the research is carried out. While focus groups continue, recruitment is often online, and some papers have almost entirely replaced face-to-face interaction with online panels. On the stock market, online research firms such as YouGov, Brainjuicer and ToLuna have formed what almost amounts to a new media subsector. Using electronic panels can cut the costs of research considerably – and there's some evidence that online surveys can attract a younger readership to join the panel.

Besides, focus groups have one great disadvantage according to one informant, who asked not to be identified - "We have people nicking the biscuits!"