Why data quality matters

Why data quality matters
Keith Jones, head of data services, Royal Mail

Data and marketing are two things that sometimes don’t go together very well.  Marketers often focus on the creativity and the offer they want to put out to market and forget that actually who you’re sending it to matters immensely. Obviously in direct mail the data is critical. Getting the address right, getting the recipient’s name right really makes a big difference to the way the campaign will go in terms of its execution and its effectiveness.

So address quality’s really important, and there are some very simple steps that people can take to make sure that the data they’re using is as accurate as possible. There are a number of data bureaus available where organisations can present their data lists and have them cleansed. They’re usually cleansed against things like the PAF file, the Postal Address File, and that ensures that the address location data you have is accurate. It’s also worth doing it against things like Mortascreen, which is a bereavement file. The last thing you want to do is send out a really creative piece to somebody who regrettably died just recently. Those are two very important things. 

And then there’s a third step which is all around that actual individual itself.  Bereavement solves one side of that, but the other side is making sure you’ve got the right salutation, the right name of the individual. It’s amazing how many times marketers fall down in putting the wrong salutation and the wrong name. So when you do a ‘Dear’, it comes out as ‘Dear Jones’ rather than ‘Dear Keith’ just because you had the data the wrong way round. So sometimes it’s not about the cleanliness of the data, it sometimes relates to formatting. But get those things right, and it’s quite easy to do, and it’s amazing what a difference it can make to the marketing campaign.

Data suppression
Suppressions is one of the most important areas in direct marketing, often I’m afraid forgotten. More often than not, not considered that important, and yet from a legal perspective it’s probably the most important thing you can do. There are several levels of suppressions. There’s the obvious ones of if you have a dataset that you have purchased and/or is in house and hasn’t been cleansed for a while. There’s then the level of suppression that takes out the Mail Preference Service or if you’re going to use the data for telephoning then the Telephone Preference Service. 

Those are legal obligations, so that if you have data that you’ve purchased that hasn’t been suppressed against those services you can be complained by the consumer, by the receiver, they can complain to various bodies including the Information Commissioner and that can lead to some level of censure.

And of course suppression may also be against those people that have asked you not to communicate with your brand. Consumers do have the right to opt out of communication. So really important that suppression is used.  And of course the advantage with suppression in the current climate is that you are taking away records that don’t want to receive your mailing piece. In effect you’re saving money, and suppression should be viewed as a way of increasing quality and most important, therefore, increasing response rates and ROI around the marketing campaign. I’m often amazed that marketers will often do a lot more consideration of targeting. From an analytical sense they’ll look at who they should target and then forget the simple things that some of those people may not want to receive the actual communication.  So it’s a really important step in the marketing process.

How to assess and improve data quality
Poor data quality really can have an impact. The obvious thing to start to look at is historical performance of campaigns. So you look at response rates. Response rates are a percentage usually of the number of responses compared to the volume mailed.  If your data quality is poor most of your mailing or some of your mailing, a proportion of your mailing will not be hitting the target you intended. You should therefore see a declining response rate. By definition you’re losing more than you’re gaining. So that’s certainly one important level. And as you see response rates decline from a tried and testing marketing campaign that you’ve got historical benchmark data, questions should be really asked about the targeting and the data quality that’s being used.

Some of the simple ways, how many returns do you get back from say the Royal Mail, they can’t deliver the letter, the address is unknown or more important we’re getting return to sender, so the RTS service. So a person has received it but is now telling you that individual no longer lives there. The volume of those coming back can be a key indication of whether your data quality really is eroding, and of course all those are going to impact on performance. If you’ve got a decline in response rate, if you’ve got increased returns and the goneaways, in effect what’s going to happen is your campaign costs are going to rise. You’re not only wasting material, you’re actually getting a decline in response rate so your ROI will now be declining. So somebody should be challenging the questions about is the product right, is the pricing, the offer right, but most importantly is the data quality as good as it should be? 

Data quality initiatives really important for the direct marketer, and the best place to start is with your existing customer base. And the easiest thing to do is to start to segment the customer base by age. So not only how long have you done business with them but when was the last time you did business with them?  Those that have transacted with you say in the last three to six months you can probably be fairly confident are still reasonably accurate, but those that may be over 12 months or certainly over two years you probably now are looking at the potential for significant change to have happened.

You know, Royal Mail monitor house movers and even in the low level of house moving in the marketplace at the moment we’re still getting over a million and a half people moving premises every year. That’s a lot of change. When you think of the adult population that’s probably about, what, 5% of the adult population moving?  So you can see over a year if you leave it two years, three years you can get significant deterioration in data quality.

B2B data
B2B data is different to the B2C world. There are serious areas to consider when you start building a B2B database. You get less movement on address in the business world, but you have more movement in people. And what matters with a business record is the people. So knowing who the marketing director is, the sales director, the CEO, the MD, or whatever level, the IT director that you’re dealing with, knowing that person’s individual name will add value to the record. If you just have a job title that really will impact response rates, so keeping names accurate in B2B is really important. That probably comes from continual communication as well as refreshing in the open market and buying the right sort of data.

At the same time in the B2B space there is this wonderful grey area called the SOHO market, which is the Solus Office Home Office, so this is the world of somebody working from home and it’s a difficult question as to whether they’re a consumer or a business. Knowing that on your database can make a big difference to the way you communicate with them. So really understanding how you deal with B2B data can make a big difference to the way campaigns will be executed and the response levels you get.