Professional help has its benefits

The froth surrounding a new technology can impress businesses so much that they want to use it even when inappropriate. Take the case of the leading market research agency that was asked to conduct research into how women who are trying to conceive use pregnancy testing kits. The client was insistent that the sample should represent the target market and that interviews should be conducted online. However, as many would-be mothers do not have access to the Internet, fulfilling the client’s wishes would have biased the sample from the outset.

Fads in market research are nothing new. What is unusual, however, is that clients, rather than agencies, are rushing to embrace the latest techniques. With the rise of Web-based interviewing it has become possible for companies to contact customers themselves, without hiring field researchers and calling teams. Add to this intelligence gathered from EPOS (electronic point-of-sale) data, loyalty cards and customer feedback channels, and one might ask whether companies need external researchers at all. But can do-it-yourself practitioners hope to replicate the skills of professionals? And if the DIY approach catches on, where does this leave market research agencies?

Whatever the expectations of clients, many professionals consider the DIY approach to be unscientific and uninformative. Paul Milsom, a senior analyst at market research agency BMRB International, says: “DIY methods are the equivalent of a straw poll round the office. They will almost certainly lead to poor decision-making.”

Milsom is not expressing mere professional pique. A lot of DIY research is crude. For example, online “omnibus” surveys, which companies conduct by posting questions on an open website. The downside of such methods is that the user has no information on who has responded, and hence no knowledge of whether the sample represents the target market, even with respect to basic characteristics such as age and gender.

Online surveys are clearly susceptible, then, to sampling bias. But the problem is not intractable, provided that companies hold good customer data against which to measure the characteristics of respondents. To do this properly, however, companies need the requisite skills. Among other things, these include knowing how to design unbiased questions that are easy to understand, being able to spot potential sources of skew and having the competence to apply weighting to correct this.

Get them to join you

One way for companies to acquire market research expertise is to recruit professional researchers with agency backgrounds. Online financial services provider Egg did this when it brought a large part of its research programme in house. Head of customer research Mark Pearson says: “It makes sense to write and analyse online surveys ourselves, because we have the skills and we understand our business better than external researchers.”

The DIY approach may work for Egg, but it is not clear whether other companies would benefit in the same way. There has been a recent trend for in-house people to become less involved with the mechanics of research and more focused on using research findings to improve business performance.

If embracing DIY methods were to have the effect of turning back the clock, so that marketers spend more time worrying about technical issues, the overall effect would be negative. As Link consumer strategies director Alison Falconer says: “Getting the wording, routing and specification of data right involves days of detailed work. This is the behind-the-scenes groundwork that clients never need to know about – until they start doing it themselves.”

Fortunately, clients are not restricted to a simple choice between doing everything themselves or outsourcing the entire project. A third option is to use external expertise to set up the research and allow in-house people to concentrate on interpreting the findings and putting them to commercial use.

For companies that favour this intermediate approach, Web-based interviewing holds a particular appeal. Not only do responses arrive sooner but, when combined with the appropriate software, online media can be used to deliver real-time reports direct to employees’ desktops. As Bill Burey, director at consultancy CCB Profits from Data, says: “The great thing about online data is that it allows clients to monitor results themselves on a daily basis, and to spot trends as they occur.”

Knowing your name

Another attraction of online research is that it creates a mechanism for companies to personalise their relationships with customers. A key distinction here is that classic market research holds sacrosanct the principle of respondent anonymity, while in online research companies often emphasise that they will use respondents’ comments to create a more customised service.

For example, the online retailer eComet, uses e-mail to collect information from customers on what products they hold, how frequently they wish to hear from eComet, and the factors that motivate them to buy online. The power of this approach is that it enables firms to target their offers more accurately and to tailor their communications, for example, by emphasising cost-effectiveness to price-sensitive consumers, and product performance to people who value reliability more highly.

Whatever the benefits of online research, there is no denying that it has limitations. Egg is once again a case in point. The company has embraced online research with enthusiasm, but nevertheless continues to use external agencies for telephone research, focus groups and in-depth interviews. Why?

For a start, digital interaction lacks the flexibility to tackle complex issues effectively. As Pearson says: “Some types of research, such as focus groups, just don’t work on the Web.”

And even for quantitative surveys, online interviewing has its limitations. Burey says: “Go beyond nine or ten questions and you end up with incomplete and unusable data sets, because respondents lose interest.”

Outside looking in

But it is not only in a technical sense that DIY methods limit a company’s horizons.

As Alnoor Samji, corporate and consumer research director at market research agency MORI, says: “Data derived from loyalty cards, EPOS and online communication are valuable, but companies need an external perspective to spot where there are gaps and to validate their findings.” For instance, a company might conduct a Web-based survey to find out what motivates customers to purchase online. To get a complete picture, it would need to establish through external research what factors motivate people who purchase through retail outlets, or consume competing brands.

But, if DIY activity actually leads to more external research, does this imply that agencies can simply sit back and wait for the business to roll in? Not if they intend to maximise their share of profitable research, says Adelphi International Research research manager Stuart Green and Boots senior research manager Catherine Munro, who presented a joint paper on DIY research at the Market Research Society’s annual conference.

One of the major challenges confronting companies today is to combine data gathered from multiple research channels in a way that yields new insight and a holistic understanding of business issues.

The implication is that agencies must demonstrate that they have the skills to help clients realise this goal. If they do not, there is a risk that companies in other fields – management consultants, database managers and marketing services groups – will corner the market in data interpretation, confining the market research industry to the more resource-hungry and less profitable business of data collection. So what can market research companies do to fend off this spectre?

To begin with, agency staff need to acquaint themselves with the DIY studies conducted by their clients, and to show that they are willing to combine the findings from these projects with their own research. The difficulty here is that clients may be reluctant to share, or simply fail to see the purpose of sharing, data with agencies.

To get over this problem, argues Green, agencies need to evolve towards building relationships with clients, rather than viewing projects as one-off items. He adds: “In this respect we have a lot to learn from industries such as advertising, which tend to think more in terms of client accounts and less in terms of specific projects.”

Technology is allowing clients to adopt a DIY approach to market research. Taken to the extreme, this trend could result in companies reducing their use, or even dispensing with the services of agencies. To forestall this possibility, the market research profession must ensure that clients understand the dangers of becoming overly dependent on internal data. Companies that lack an external perspective also lack the means to benchmark their performance against that of their competitors.

Nevertheless, to dwell on the risks associated with DIY research is to miss a fundamental point. When applied with skill, DIY research yields insight into customer behaviour and creates opportunities for follow-up work. For example, a client might use purchasing data derived from its website or loyalty card to segment customers by their shopping habits, before commissioning an agency to find the reasons underlying such patterns.

What follows from this, as Green and Munro conclude in their paper, is that DIY research “is not the enemy” of market research professionals. Rather it is a positive development that agencies should welcome, and seek to build upon.