How to conduct effective research

Market research is only valuable if you ask the right people the right questions. So how do you get the most from surveys, focus groups and mystery shoppers?

Marketeers are on an eternal quest to understand what motivates their customers. And that desire to get inside customers’ heads is what drives market research.

Marketeers are on an eternal quest to understand what motivates their customers

So although it might seem blindingly obvious, the very first thing to ask about market research is whether this is what you need. Rather than rushing to put numbers and ticks in boxes, marketeers need to stop and think about exactly what it is they want to find out, and assess whether market research can do that.

POV Marketing and Research Head of Insight Anthony Tagal says: “There’s a knee-jerk response. Clients feel they should do research when there’s nothing better to do. Business and marketing are about creating new ideas and insights for the product, brand idea or creative route. So it’s about finding meaning. Research is one way of finding meaning but we can get lazy with what we ask and assume that research will give us everything we need.”

Let’s start at the very beginning

How you start a research project is crucial if you want to avoid wasting time and money. Research International Client Services Director Robin Birn, suggests beginning by writing down what you do know and what you don’t. This will form the basis of a brief.

HPI Research Senior Partner Juliet Strachan says: “All things stem from the brief. The most important things to include are what the business problem is, how the research is to be used and how the outcome will answer that business issue.”

It’s easy to fall into the trap of using research to validate something you’ve already decided to do, cautions Vodafone Global Marketing Director of Customer Insight Andy Moore. “Don’t decide to do it and then think, ‘I’ll check it out with the customer’. A much better route of gaining genuine insight is first arming yourself with an understanding of the core consumer drivers and then making sure that’s shaping your thinking.

Great marketeers will listen to what the consumer is actually saying.”

When to outsource

Once a good brief has been drawn up, other decisions flow from that. One of the first will be whether the research can be done in-house or whether it should be outsourced. The majority of marketeers simply do not have the manpower or resources to run research projects themselves. Even companies with large departments dedicated to insight and research will usually outsource the legwork and use internal intelligence for analysis.

Besides the cost, there are risks to doing research in-house. Among them is the possibility that responses will be influenced by respondents’ awareness of who is asking the questions - an organization conducting its own research must make that fact clear. Equally, if your research base is flawed, those flaws could be perpetuated throughout the results.

“Carrying out research in-house, you run the risk that the hypothesis in the organization drives the design and outcome of the research, so you don’t get a fresh perspective,” says Strachan.

Another problem with in-house projects is that they slip down managers’ priority lists and are in danger of being dropped as more pressing matters arise. But before you rush headlong into a research project and get into detailed methodology, ask yourself whether you can solve your problems using existing information - does new information really need to be collected?

Sometimes the question or problem requires some digging closer to home. If, for example, you don’t know who your core customers are then there’s no point in looking for new ones. So doing your homework in-house may be the first stage to address. Of course, existing information, be that articles, directories or reports, won’t give you everything you need and, if what you’re after is specific to a product or company, you will probably need primary information.

Quant or qual?

As with most things in business, you get what you pay for. So the methodology will vary according to the scope of the research as well as what you’re trying to find out. The two main options to consider are of course qualitative and quantitative research, depending on whether you’re looking for more detail and depth on a topic, or hard figures on recall or use, for example.

With quantitative research people need to be questioned for their response; methods range from telephone to internet, face-to-face to printed questionnaires. a Course Director for The Institute and Founder and Managing Director of Market Research Agency Actionline Research and Training Sales, Jean Sutton says the most efficient and cost-effective route is likely to be via telephone.

This has the advantages of face to face - be it for qualitative or quantitative work - in that you get more depth and detail.

But the internet is becoming an increasingly popular route, especially as broadband and PC penetration increase. “When the internet first came along it was a very cheap way to conduct research. But internet research is no longer the compromise it was. Now it’s an extraordinarily powerful research tool because you’re talking one-to-one,” says Strachan.

Whether your questionnaire reaches its audience on- or offline, there are a number of points to consider. Should your questions be structured or “closed” - yes or no - or should you go for a semi-structured questionnaire with open questions as well? The latter provides more qualitative responses while the former lends itself to quantitative findings.

“It comes back to what information you need and what you want to do with it. You wouldn’t use a closed questionnaire for an audience of CEOs. Semi-structured allows for a two-way conversation,” says Sutton.

Crucial to remember is that average response rates to self-completed questionnaires range between three percent and five percent. So to get 50 responses, you need to send out 1,000 questionnaires. And you have little control over who chooses to complete - often only those who are very happy, or very unhappy, with a product or service, or those who have time on their hands, will fill them in. The only way of limiting this effect is to ensure high response rates to reduce the bias.

Offering an incentive is a common solution. But keeping the questions short and simple and making the sponsor’s identity clear will also improve the response rate.

Time to focus

The classic market research tool for gaining depth of understanding is the focus group. Many are quick to sneer at the image of a cosy room of housewives, chatting over tea and biscuits about their favourite fragrance for toilet cleaners, all keenly observed by researchers behind a two-way mirror. But this method of research is still used to great effect and can provide a forum for a target group to interact and open up in a relaxed environment.

GfK NOP Social Research Associate Director Amrita Sood, runs many focus groups for clients. She says clients are often surprised at how specific the agency can be in recruiting the right sort of people to a group. “We’ve done projects with people who’ve committed benefit fraud or crimes or with a specific range of ethnic groups. The target sample is a really important aspect of research,” she says.

Focus groups are unusual environments for most people so it’s important to compensate participants for their time, says Sood, and to be as transparent as possible about the research. While focus groups are useful for highlighting attitudes and behaviours, they should not be used for quantifying behaviour. “If you end up with 100 people in your focus groups and 80 say they love the new product, it doesn’t mean you can say 80 percent love the product,” cautions Sutton. “Focus groups are hand-picked and so you can’t extrapolate from that,” he said.

In the pursuit of ever more effective ways to understand the customer, market research is spreading its wings and techniques span ethnography, semiotics, longitudinal studies and even mystery shopping.

“Ethnographic techniques are good for immersing and gaining insight,” says Sood. “We may go to their home and spend time with them, observe their daily routines, look around their house, see how they use technology and media,” he says.

With more time and depth these research tools can be useful for context and bringing to light things that the consumer doesn’t realise are important.

Birn says FMCG clients in particular are adopting these techniques where they might go into shoppers’ homes to see how they stack tins in their cupboards or what selection of cereals they choose.

Continual and ad hoc

For many big companies, research will involve a wide range of continuous measures from brand tracking to customer satisfaction surveys, as well as ad hoc research. At Vodafone, for instance, Moore says about 40 percent of research is continual tracking, while the rest is ad hoc.

Royal Mail Group uses ongoing programs to act as a health check and to identify business challenges. It then instigates ad hoc research to delve into those challenges as necessary, says Director of Insight, Intelligence and Analysis Crispin Beale at the organisation.

The company also uses mystery shopping. One shopper survey, called Effect, is about improvement in Post Office branches. It’s carried out in partnership with employees, consulting them about what they think is important for the customer. There’s a danger that research can go wrong when companies get obsessed with measuring the wrong thing or measuring too frequently. “I’ve seen customer satisfaction measured on a weekly basis, which is too often as things move with a margin of error. I’ve been in companies where the measures were changing faster than the program could keep up with it,” says Beale. The most valuable research results are those where a performance improvement in a product can be proven to add directly to the bottom line, he says.

Research can and should produce real insight that can have a fundamental effect on the business. At Nationwide, TV advertising was developed following research revealing that people don’t think financial companies treat them as fairly as they should. Nationwide, Advertising Controller Paul Hibbs, says the theme for its advertising came from posing the question “what can we do to treat our customers fairly?” “On mortgages we don’t do ‘new customer only’ deals and we also don’t charge to use cards abroad. Most of that was driven by research into what frustrates customers,” says Hibbs.

Market research can be a detailed and complicated science but Beale boils it down to some pertinent advice for marketers: “Research should be seen as the radar of your organisation - it give you the early warning sign and then you respond to it. Research only makes a difference if you actually use it,” he says.

“It’s all about making money or saving money for the organisation, it shouldn’t happen otherwise. It should drive the business forward - how do we get ROI on this and increase income?”

The Marketeer