What would you do if an online survey dropped into your mailbox – press the delete key or click on? Earlier today I received an e-mail inviting me to take part in some market research. Within minutes I had previewed an autumn film release, rated its advertising and answered questions about my reading habits. Before broadband, gathering detailed consumer feedback meant hauling people off the streets into focus groups. Now multimedia surveys can be completed at home by anyone with a PC and a few minutes to kill.
Pre-testing films and advertising is one of many developments fuelling the growth of online research. From a base close to zero less than a decade ago, the medium has boomed into a vast industry accounting for a quarter of all the quantitative data collected in Europe. In the US, the internet is the leading method of data gathering. But does Web-based research deserve to be as popular as it is?
Shorter turnaround times and lower costs drive many businesses to experiment with online research. Jo Rigby, head of OMD Insight, says people find that the Web creates opportunities to innovate and reach new markets. As an example, Rigby cites a car launch that she worked on. “For the same budget as commissioning four focus groups, we were able to research 5,000 car buyers and discover a whole group of drivers outside the brand’s focus who were attracted to it.”
Doing the research online also made for a livelier, more engaging experience for the participants. Rigby says: “The respondents could play films of the car, choose different colours and trims from their screen and try out special features such as the sliding door.”
Some industries have been transformed by online research. Entertainment Media Research, an online music consultancy, runs a large panel of music consumers who score snatches of music, lasting 70 seconds, for industry clients. The clips are sent to the participants’ home PCs and the results fed back to help radio stations fine-tune their play lists and guide record labels and artists on the market potential of new material.
Music critics may castigate the approach for its remorseless commercialism and lament the loss of artistic freedom, but the men in suits love it. “It’s really about providing guidance to the relevant decision makers,” says Peter Ruppert, founder of Entertainment Media Research, in defence of the practice. “It doesn’t do away with the need for artistic creativity.”
International market research is another area where the internet is turning tradition on its head, allowing multi-country studies to be managed from a central office rather than farmed out to local researchers. This winnows out cost, and provides clients with the reassurance of knowing that differing national scores are down to market factors such as culture, and not simply a reflection of how the data was collected locally.
The benefits of running international research interactively are not all to do with cost savings and standardisation, however. By injecting a cross-cultural dimension into the mix, says TNS Interactive managing director Martin Oxley, the Web can become a catalyst for creativity. “It allows you to do things like inventing recipe ideas with consumers in California and getting food journalists from, say, Australia to evaluate them.”
Build up a picture
Matt Dyke, planning director at digital marketing agency Tribal DDB, agrees that online research can inspire original thinking. But to realise this potential, he argues, researchers need to drop their old habits. “Instead of laboriously asking questions, we should be tracking people’s paths through websites and going into forums and blogs to see what people are saying and doing, when they are not subject to research conditions.” Having amassed all these clues, Dyke recommends working with industry outsiders, such as psychologists, to build a picture of the consumer, in much the same way that detectives construct portraits of suspects from tiny fragments of evidence.
Yet the Web may not suit everyone. Overshadowing the triumphant advance of online research is the niggling worry that, in their rush to exploit the Internet’s convenience, companies have overlooked the extent to which online respondents are biased towards early adopters. But, with almost 60 per cent of the UK’s population hooked up to the Internet, how truly valid is this criticism?
Clearly the Internet is not a good place to research groups that are under-represented on the Web, such as the over 55s. The same goes for testing products, such as phones or hi-fi, which rely on technology to make them work, but which target the public in general. Says James Myring, associate director at Continental Research: “There are few technophobe types among online respondents, whereas in the population they make up a considerable minority.”
For general consumer goods, however, the risks of recruiting online may already be outweighed by the medium’s advantages. “If you are testing the strapline on, say, a can of corn, there is not much danger that you are going to skew the results by recruiting over the internet,” says Darren Noyce, managing director of Skopos UK, and a spokesman for the Market Research Society.
The honest truth
Terry Prue, senior partner at HPI Research, goes further, suggesting that online samples may actually be better balanced than those recruited by traditional methods, such as hall tests. “They are biased towards people who have time to go into a hall to be interviewed,” he says.
Finding high-quality respondents who can be trusted to give their views honestly rather than simply signing on for the incentive is a challenge for all agencies that run large consumer panels. But does the internet perform worse than other media in this respect?
Julian Price, senior vice-president of marketing and product development at Lightspeed Research, acknowledges that the absence of an interviewer makes it harder to check that people are being truthful. But most of the problems that crop up, he argues, stem from lax practices and are avoidable. As an example, Price points to companies that place surveys on the Web targeting, say, 24-year-old males who drive BMWs and invite anyone who fits the bill to complete the survey and claim the cash reward. Not surprisingly, a crowd of people come forward who meet neither the demographic nor the ownership criteria. So what should agencies be doing?
Firstly, it is important to profile people’s age, sex and occupation when they first join a panel, so that panelists can be matched to projects, rather than be allowed to “self-select”. Other precautions include recruiting consumers from as many different channels as possible to prevent the panel’s composition becoming skewed towards particular groups, and keeping the financial rewards for completing surveys as low as possible. The payback from doing things properly is that the agency has a ready-made pool of people, with known characteristics, to call upon. “With traditional field work, recruiting a hard-to-reach group, such as the under 18s, might take two or three weeks,” says Price. With an online panel, the surveys can often be sent out immediately.
Sometimes an online approach has a head-start over conventional methods, as with sensitive topics. Face-to-face with an interviewer, many people would rather lie than admit to habits or opinions that they think others might frown upon. Cloaked in the anonymity of cyberspace, they feel confident to talk freely. Oxley gives a classic example of this phenomenon from Holland, where sales data suggested a lot more people were buying pornographic magazines than the numbers implied by traditional market research. When the same survey was conducted online, the figures shot up.
But what of the future? Having assimilated the internet into quantitative projects, agencies are turning their attention to qualitative work. TNS recently completed an online project for a client which suspected that focus groups were killing off potentially ground-breaking products dreamed up by its development team.
To get around the problem, TNS set up an online group and a discussion forum. This allowed the agency to float ideas with consumers, then interview and re-interview them over a fortnight to see how their thinking changed as they started to mull over how the products might help them. Technically the approach did what was required by giving the participants time to get over “the shock of the new”. But something was lost. “Not being able to see people’s body language was a handicap,” says Oxley. “What was missing was human connectivity, the emotional drama that you get with face-to-face.”
Virtual, not actual
Other agencies report similar experiences. Nunwood Consulting recently undertook a project for a client that wanted them to create a virtual nightspot in which to test concepts for drinks. Although the research team tried to put people into the right frame of mind with “flashy graphics and clubby music,” they came away with the feeling that something was not quite there. Says Nunwood client consultant Laura Morris: “It’s very difficult to replicate the experience of a live environment online.”
Such problems may, of course, be temporary. With improvements in webcam technologies and internet telephony, virtual focus groups could soon be challenging groups held on bricks and mortar premises. Whether the technology will progress to the point where people actually feel that they are sipping cocktails in a bar, instead of gazing at a PC screen, only time will tell.