Data hygiene and accuracy have never been more important.; As we create more mature relationships with consumers (built on honesty, openness and transparency) then the way that we keep our databases up to date, relevant and accurate must also move with the times.
It is, in fact, a surprisingly mature market; we launched The GAS File (the UK’s first suppression file) in 1992, some 23 years ago. Yet in the majority of suppression files sold today, the offering remains essentially the same. This means it will find someone that it thinks has moved house or died and stop spending the client’s money on sending mail to them. On the face of it, simples! But of course it rarely is. What happens if a person moves jobs, or a business closes (50% of start-ups close in the first two years)? What happens in the event of a divorce (one in four households)? What happens if the husband has died but the wife continues to live at the house (about 34% of deaths)? What happens if the household has only moved out for six months to refurbish the house (80,000 a year)? What if the child, now at university, has the same name as the father or mother (more common that you might think)? Not quite so “simples” as it seems.
In the days where carpet bombing and mass targeting were the de-facto strategies, the idea of getting rid of another 40, 50, or 60 thousand names from your file really didn’t matter; the so called “Spot and Drop” model. However, in this new-found era of “less is more”, with marketers trying to forge “real” relationships with their customers and prospects, the number of targets prospects is much smaller. Therefore every single one should be nurtured as fully as possible, creating no room for error. If you get something wrong, they will be in the arms of a competitor faster than you can say Jack Robinson!
I still sit in many meetings and have clients say to me “Great, supplier x has achieved twice as many matches as suppliers a, b or c”. I explain that they are asking the wrong question; what in fact they should be asking is “how and why has supplier x matched more than a, b and c?”. Querying where the data is from is a remarkably infrequent question asked these days. A single source of data in such a peripatetic marketplace is unlikely to ever cut it.
The Spot and Drop model also concerns me as we should be using that data to enhance our positive targeting. How often do you undertake an analysis of those customers that have moved house without notification? Is this important? If you knew the profile of these individual customers, could you try and preempt it? How often do you take your Gone Away file and, firstly, see if they are still customers and, secondly, if they are not, try and relocate and re-engage? Experience shows that the longer you leave it after moving home, the less likely they are to return.
Over the years, we have discovered that by building a universe of population movements we now receive a raft of confirmations that lift the levels of certainty on these issues to extremely high levels. So in the case of home movements, we track the whole range: move outs, move ins, temporary changes, single move outs/ins, as well as a host of other transactional information to validate either a constant address presence or a new one. We apply the same validation rules across our entire data product suite as we know that using the wrong information will always be more expensive than doing nothing at all. This is not to discourage you from doing anything at all, but to impress upon you that, these days, using accurate, qualified data is the only way forward.
So the next time someone says they’ve just cleaned up your database and got rid of 500,000 records, think of that as the start of something, not the end of it.