Tools you can trust?

Turbulent is not usually an adjective that springs to mind when discussing market research. But the normally placid world of sampling is bracing itself for the shock waves the Internet has been threatening to send through the industry.

Few in the industry dispute the Internet’s potential to lower costs, increase speed and generally enhance the quality of the research. However, the rate at which the number of online surveys is growing is sparking debate about how to maintain standards in this unpredictable, brave new world. Millward Brown Intelliquest, for example, recently published figures estimating that up to 95 per cent of market research in the EC by 2005 could be conducted through the Internet. Statistics like these explain why the industry is taking sides on the issue, aligning with the cyber-phobes, cyber-philes or cyber-agnostics.

Richard Windle, European research director with Internet audience measurement company MMXI Europe, says: “We should always be looking to test new techniques and research methods. But at the same time it’s important to remember that the Internet, like any market research tool, has its limitations, particularly when it comes to representative samples.”

As someone who has been an adviser to the Market Research Society (MRS) on Internet research methodology, Windle has been involved in developing guidelines on how online surveys should be conducted. “Internet-based research doesn’t imply radically altering existing offline techniques,” he says.

“The problem with the Internet is that it is unrepresentative of the population as a whole. When Internet penetration is sufficiently high, it will then be possible to use it for surveys in the same way that the telephone is used today – but that is probably some years away.”

However, because it has become cheap and relatively straightforward to produce surveys through the Internet, those companies with an eye on the bottom line may be tempted to throw caution to the wind and commission them.

The MRS is concerned that there could be unethical research organisations only too willing to part naive customers from their money, and so undermine the integrity of the profession.

Peter Hutton, head of MORI’s new media division, believes in common with the agnostics that it’s right to be cautious. “The cowboy approach is to put together an e-mail questionnaire, send it out to addresses compiled from any number of newsgroups and wait for the response. However, you’ll end up with results that have the potential to be grossly misleading.”

Organisations such as the MRS and Esomar – the representative organisation for European-based market research professionals – have published guidelines and codes of practice for Internet-based research. But, as Windle points out, there is actually little difference in fundamental principles, whether the research is conducted on- or offline. “The issue is how to harness the new technology to be more effective rather than changing everything,” he says.

Time pressures

The focus on developing new applications is one shared by Stephen Martin, managing director at Aspect International Consulting. Martin believes that with companies spreading their tendrils internationally, the pressure to deliver results quickly is growing all the time.

“The Internet isn’t the point,” he says. “The point is that we’re dealing with an increasingly global market that is changing incredibly fast. Therefore, clients are faced with a problem – they are having to work at speeds which are very difficult to deliver against. Internet research is a good opportunity to get snapshots of markets quickly, and to ensure that companies aren’t held back by normal timelines. The Internet is a symptom of the times – and the times require fast-moving research.”

Even so, Martin agrees with Windle, Hutton and others that snapshots are not a substitute for strategic market research programmes. “The potential problem is that companies may start to neglect ‘proper’ research where they undertake detailed customer or prospect understanding. It’s up to the strategic research companies to improve standards and ensure this doesn’t happen.”

In spite of the fears of even the most cyber-phobic research manager, the majority of research companies respect the limitations of the medium and accept that, beyond certain applications – such as testing audience reaction to a website or the attitudes of representative target groups, the process becomes a lot more complex. Companies on both sides of the Atlantic are now tackling the issue by applying both traditional and online research techniques to compile databases of internet users who can be used in properly selected online surveys.

Rolling research

NOP has used traditional research techniques, as part of a rolling research programme investigating the pattern of Internet use in the UK, to conduct a series of online research projects using a database of selected and screened Internet users, collated as a by-product of the programme. But a number of companies, including research consultancy RSM, are using specially-designed sites as vehicles for online surveys.

Richard Gormley, a partner at RSM, believes that using the Internet is the best way to research the Internet itself and, for large global surveys, offers very significant savings in time and expense. It does, however, pose a number of new challenges – not least of which is the familiar one of representation.

Internet users are still predominantly affluent young males, so selecting a representative sample requires the development of new techniques. “Respondents may come from a number of sources, such as customer databases and mailing lists,” explains Gormley. “Before they take part in the survey, they will have been screened online by means of an allocated password and a series of self-verifying questions to ensure that they meet the profile.”

But why should anyone wish to take part in an online survey? Gormley identifies two reasons. “The survey will often coincide with an area of interest, and secondly, there will be an incentive for people to take part, such as a cash-prize draw or a subscription as a reward.”

Some of the claims forwarded on behalf of Internet-based research may need more scrutiny, but the growing number of alliances and joint ventures between the international research companies and the specialist online companies bear testimony to where the smart money is going. French Internet research company NetValue, for example, recently joined forces with Taylor Nelson Sofres to run Internet panels worldwide, while US Internet research house, Forrester, continues its global expansion through acquisition and mergers across the world.

Despite all the activity, a number of people remain sceptical that online research will turn out to be the panacea that many are predicting. Research Business International managing director Janet Kiddle is unconvinced that the Internet will be able to reproduce the important observations of qualitative research, such as group dynamics and human interaction. “Visual signals and body language are such crucial elements of qualitative research and they can’t be reproduced online. I can’t foresee a time when the same results could be achieved over the Internet,” she says.

Challenging concepts

However, as the medium evolves and new ways of interacting with consumers emerge, opportunities for testing opinions could challenge fundamental concepts of market research in ways that can’t yet be articulated.

Bjorn Haugland, founder of Future Information Research Management (FIRM), is one of a growing number of traditionally-trained researchers who believe that the advances are being held back by a reluctance to let go of long-held ideas. “With the Internet, people have got a new tool but they’re still using it in old-fashioned ways,” he says. Haugland sees the ability to personalise online surveys as something that may consign traditional forms of research to the history books.

Haugland views companies that send out non-specific and irrelevant e-mail questionnaires as the worst abusers of the medium. “Anonymity could be the death of the research industry,” says Haugland. “The Internet provides research options that can be either personalised or anonymous. You can create online dialogues with people according to how they want to respond. It’s built around having a close understanding of that person’s specific requirements.”

This idea implies personalised research based on information derived from any number of different sources. “Representation increases if you let people have it their way,” says Haugland. “The response becomes asynchronous so people can answer in their own time and not when it’s inconvenient.”

For all that, technological advances continue at breakneck speeds, but the industry is generally reacting at a much slower pace, and there remain concerns about new methodologies. MORI’s Hutton says: “People are unsure of the new technology, which is holding back progress. You have to control the sampling methodology, otherwise it becomes an open charter for the cowboys.”