Truth will out

You might wish that you earned twice your current salary, but would you go as far as ticking a box on a questionnaire to indicate that you do – just to impress someone you’ll never meet? Whether you, or people like you, would do such a thing is a key issue when evaluating lifestyle data for use in direct marketing campaigns.

Lifestyle data allows market segmentation by categories such as income, car ownership and consumer behaviour. Its utility is based on the assumption that the best indicator of future behaviour is past behaviour. This is the kind of assumption that means you are more likely to receive direct mail requests for donations to charities after you have been registered as having given to another charity. It can also have lateral applications – buying clothing via mail-order might be seen as showing your willingness to purchase using this channel, and makes you a much more likely proposition for producers of gardening catalogues than the mere fact of having a garden. The biggest competitor to lifestyle data can be said to be “geodemographic data”, although Paul Seligman, managing director of multidisciplinary agency 141, says: “No one ever seriously bought the proposition that you were the same as all the other people in your postcode area.”

Agencies have traditionally been wary of the validity of lifestyle data. It is believed that respondents are often economical with the truth and tick the boxes that reflect their aspirations rather than economic reality. So is this true or can lifestyle data be used with confidence? When is it the most appropriate choice? And how can researchers get the best from it?

Richard Bandell, managing director of lifestyle database company Claritas, says: “There has always been a certain amount of scepticism about the accuracy of lifestyle data – more so in its early days than now. However, lifestyle data is an increasingly important element of many organisations’ marketing plans. Its applications today go way beyond direct mail targeting to include strategic analysis, customer segmentation and geographical planning. I doubt it would have grown to play such an instrumental role in the market if it were seriously inaccurate.”

Peter Thompson, director of data analysis experts Experian Prospect Targeting says: “Ten years ago the validity of lifestyle data was indeed questionable, as volumes of data were much lower. Today’s suppliers offer lifestyle information on 35 to 40 per cent of the UK population. In the past there may have been an argument for checking the validity of data, and agencies were more sceptical about its use. In those days blanket mailings were often used, but now we conduct more sophisticated targeting for clients.”

Would I lie to you?

So were stories of lying respondents just rumours spread in order to discredit lifestyle data? Mike Hazeldine, direct marketing consultant at database marketing specialists EuroDirect, says: “People will lie out of self-interest or self-aggrandisement. There is no self-interest in lying on lifestyle questionnaires.”

Seligman agrees. “Most people are quite honest. They do not like telling lies. If we mistakenly encourage them to lie then yes, some will, but modern lifestyle questionnaires try to encourage people regardless of their answers,” he says.

Both Hazeldine and Seligman are adamant that anyone being creative with the truth is easily exposed. This view is shared by Jonathan Plowden Roberts, head of consultancy and analysis at database specialists the Database Group. Plowden Roberts says: “The question of validity is almost impossible to answer satisfactorily because there is no independent check on responses.” He adds, however, that: “Companies can spot inconsistencies in the data.”

So what other causes can create inconsistencies? “There are some areas where data is less reliable,” admits Seligman. “This is more because of consumer perceptions. Examples include ‘forgetting’ how much one drinks. Where we ask specific questions, such as ‘when is your car insurance due for renewal?’, consumers do not immediately know the answers, and hence may make something up. Where we have asked consumers how much they spend monthly on shopping, and then cross-referenced it with transactional data, the two are pretty close.”

The interpretation of dreams

There also seems to be a general feeling that even people’s occasional creativity with the truth when answering questionnaires could provide a source of important information. If you indicate that you earn a low salary but claim to drive an expensive car, this could show what your aspirations are. As Plowden Roberts notes: “This could mean that perhaps you could be in the market for a more expensive car in the future. Companies can look on this as an opportunity for long-term relationship building.”

How do you make sure that you are getting as close to truth and accuracy as possible? Cross-referencing of data is a key issue. Plowden Roberts believes lifestyle data “is a perfectly valid way of gaining information when used in conjunction with geodemographic and psychodemographic profiling.”

Seligman says: “The best measure of the accuracy of lifestyle databases is when the data is aggregated and can be compared back against census data. The bottom line is that the data stacks up pretty well nationally, but there are local gaps and corrections need to be applied.”

Perhaps the most important consideration when using lifestyle data is recency. Ashley Grainger, business development director of data marketing consultancy DMS, says: “Lifestyle data is very dependent on timing: a person may have expressed an interest in baby products two years ago, but they will probably no longer be interested as their children are older. Similarly, cars and car usage are heavily dependent on factors such as finance, which can change dramatically.”

Hazeldine agrees, commenting: “People’s circumstances can change dramatically in the space of two or three years. Take the example of a young professional couple – in that time they can get married, set up home and have a child, which may mean one of them gives up work. Suddenly they aren’t going on holiday three times a year any more. It takes much longer for postcode areas to change than it does for personal circumstances.”

Consequently Hazeldine advises that only data gathered in the preceding year should be used in fields where such changes can occur.

Direct Marketing Practice director of media and targeting Lyn Bennet says: “Lifestyle data works best in niche campaigns, for example in the financial and insurance sectors, where date of birth, renewal dates and specific qualifying data are required. It used to work well for the charity sector, but recently we have found performance has declined. This is due to a number of factors including overselling or overmailing.”

Plowden Roberts advises that lifestyle data can be most successfully used when “you know the characteristics of the people you’re after and want to speak to them on a personal basis”.

God is in the details

Bandell says: “One of the unique advantages of lifestyle data is the ability to target different individuals in the same household. That’s because most lifestyle companies collect spouse or partner data in addition to respondent data. For example, it’s well documented that females are the main decision-makers when it comes to some products, while males make the decisions in other areas. If you know that the demographic profile of a household is right for your offer, you can heighten response simply by targeting the right individual in that household.”

Thompson adds: “Lifestyle data is a source of information about consumers’ latest thinking, and so is particularly suitable for targeting new technology products. It is possible to sponsor questions on a lifestyle survey to get precise feedback. This makes it possible to generate tailor-made products for your customers, based on having a clearer understanding of their buying triggers. Sponsoring questions can also be used for market analysis, to test consumer attitudes prior to launching new consumer brands.”

As the saying goes, a bad workman always blames his tools. In the same vein, data alone cannot be blamed for poor results. It should be cross-referenced against other sources that will both ensure accuracy and enhance its profiling capabilities. It becomes most effective when matched against an accurate potential customer profile.

Seligman says: “The industry is structured around ridiculously low expectations. I consider a survey response rate of between one and three per cent to be woeful, and a sheer waste of resources. The inappropriate use of direct marketing can also switch people from being neutral towards a brand to being actively hostile. This is the real danger – the pursuit of short-term business gain at the cost of making long-term enemies.”