SBHD: Having a database is one thing – knowing which one to buy and how to get the best out of it is a completely different matter. Systems that have been developed and nurtured over a period of years usually prove most useful.
During the late Eighties and early Nineties, developing a database was top priority for many companies. Now that anyone who is worth anything has one, the question is what to do with it.
Many practitioners say that in the rush to create a database, a lot of clients took a short cut and ended up with databases which are limited in scope and difficult to build on. Databases were built for the sake of having them, with little thought given to how they should and could be exploited.
But merely owning a database does not guarantee a competitive edge; it is a marketing tool like any other. Only when the data is converted into information through analysis, can any value be derived from it. A really successful implementation of the database should result in fundamental changes in a company’s business – but this is rarely the case.
The problem seems to lie in being able to merge marketing expertise with the techno brains. The marketing people know their stuff and there are plenty of computer whizz-kids around, but there are very few people who are well schooled in both. And they are very expensive.
Ogilvy & Mather Direct Dataconsult head Nigel Howlett says that most clients know how to develop systems around accounting and sales procedures, but they come unstuck when they build more sophisticated models which in-volve marketing strategies.
“It’s really a matter of education. Most database users are not aware of the potential of the database, and if they are, they are not sure how to go about ex-ploiting it,” he says.
Howlett says there are a number of pitfalls facing database development but the biggest is the desire to build a database as quickly as possible. “When people grasp the opportunities, they get terribly excited and that translates into having a database as soon as possible and as a result short-cuts are taken,” he says.
“Databases’ weakness used to be that technology was being stretched against delivery. Technology is now much better and cheaper, but the problem is that marketers are not ableto define what they want rigorously.
“Marketers are trained to think laterally. But you can’t brief systems developers and programmers in vague terms. The first step is for the company to agree at senior level what the objectives and business needs of having a database are. Unless these objectives are defined by the client, it is unlikely the systems people will come up with what you want,” says Howlett.
Because the investment in databases is usually so high, adds Howlett, there is often the danger of overselling the concept at board level and subsequently being bumped along to get the job done quickly.
SJA Direct director Steven Jenkins believes clients must stop thinking of a list of names and addresses as being the same as a database.
“When you add relevant information to it, a list becomes a database. A mar-keting database is relational in style and running a database is a lot more involved than keeping a mailing list clean. A database should imply something active, like taking on an employee. Clients tend to think of it as a black box that things come out of but nothing goes back in,” he says.
“In the latter part of the Eighties and early Nineties, having a database was the in thing. What was generally developed at that stage was in-company data resulting in a general transaction database, which is completely unhelpful for marketing purposes.
“Databases were not set up in a relational format, which means clients could not identify customers with specific characteristics. Unless you can do that, all you have is account-based information,” comments Jenkins.
He admits that databases involve high investment but adds that “the value of a database enables you to start calculating real customer value. With a database you can capture how much your customers are giving you in terms of margins, loyalty, longevity etc.”
Because of this high investment, says Jenkins, clients are tempted to write massive specifications for a database; in short, turning in a wish list to get their money’s worth.
“It’s not necessary to have a database that holds everything. In fact, the more data you hold, the more chance you have of making a mistake with it.
“Most people populate their database with internal information and data that they buy in. Unfortunately, because of the recession, the quality of the material you buy in has declined in recent years. Direct marketing activity has been lower and list brokers have not been able to keep their lists as clean as they might, so you end up with the garbage in/garbage out scenario,” says Jenkins.
LBV managing director Chris Lovell says his agency haschosen to call databases information or knowledge bases to reflect their potential.
Lovell also claims, however, that a truly inter-relational database would be so expensive to develop that there is no real business reason for doing it.
“The key is using data more intelligently rather than concentrating on actually build-ing it. The existence of a database makes marketers think they don’t have to do any more. But a database is like a pen or a piece of paper – it’s the using of it that counts,” adds Lovell.
He says that most clients’ knowledge of databases “ranges from the sublime to the ridiculous”.
“Marketers have not bothered to educate themselves in data areas. Either they have recruited someone in or used the in-houseIT function to develop it.
“Computers and databases started for financial reasons when accounts departments wanted a system to replace double-entry book keeping, followed by production departments. In those circumstances, computers solved a lot of problems.
“But the systems developed for accounting and production have tended to be used in marketing – and it just does not work. Computer people have tried to formalise and create a static environment for marketing to work in – because that’s the way computers work. But marketing, rather than being static, is very fast moving,” says Lovell.
In the end, it’s those databases which have been developed and nurtured over a period of time that are going to prove most useful.
The Computing Group database director Dick Davies believes it takes at least three years to develop a sound database.
“You can’t just build it and not feed it,” he says.
“When a system is put together to fulfil an immediate need, it will inevitably screw up. A company has to look at its long-term marketing objectives and build a database that reflects the whole operation.
“A company must know from the start that the database is about really understanding who its clients are and that process will take a few years,” says Davies.
Most people feel that the gap between the marketers and the technocrats needs to be bridged by the marketers.
O&M Direct Dataconsult’s Howlett says: “You shouldn’t let technical vendors write the brief. Because database marketing is relatively new and clients don’t really know what they want, the danger is to give the technical vendors the task of writing the specifications.
“The client’s needs then tend to reflect what the vendor is selling; and you can’t expect systems people to sud-denly understand marketing.”
In the end, however, all the technical wizardry in the world will not ensure intelligent use of data.
According to Howlett, the system has to be built in such a way that clients will be able to understand their customers better.
“There are some companies which have spent millions of pounds on building a data- base and then do nothing but the most cursory forms of analysis,” says Howlett.
“To understand what drives people to do the things they do, you have to do data modelling and build predictive, rather than just descriptive models.
“Many companies tend to stop at the descriptive stage, but still have no idea about what makes their customers behave the way they do,” he adds.