CRM is only as good as the data you put into it. Here’s how to keep on top of data cleaning.

Untitled design

Data cleaning isn’t a one-off event, it’s an ongoing process. With new EU laws mandating ever-tighter conditions over what you can retain, there are too many ways to fall foul of both the law and your customers.

Fortunately, a quick check now can reveal some typical mistakes that might be hindering your marketing and lowering your response rate. Most of these issues are simple to fix if you stick to a few straightforward guidelines. Review a sample of your database for these common data cleaning errors today.

1. Is your representative sample… representative?

First up, make sure any data you sample really represents your database! That doesn’t mean looking at the first 100 names. It means choosing a statistically valid subset of your data to get an idea of the database as a whole.

Who is the best fit for the job? Should it be done in-house, outsourced, or through a CRM partner? This is a great question to bring up with your CRM expert. They work with many databases and often see the same mistakes in data entry. A few hours of their assistance could lead to big benefits down the line.

2. Do common-sense spring cleaning on the simple stuff

Every list has a few John Smiths on it. For more unusual names, simple queries of your data can reveal the incidence of duplicate names. If more than 2% of the entries in your database appear more than once for firstname-lastname, you’ve probably got duplicate entries for the same person.

How often this happens varies from sector to sector. Job-hopping industries like hospitality tend to see more of this than the public sector. Allocate resources to it depending on who your customers are. The query syntax is quite simple and once learnt can be applied annually, or quarterly, or however often you want to carry out data cleaning. This would be worth handling in-house.

3. Check when your data last touched the customer

A surprising number of CRM contacts have a lot more activity for surnames starting A-M than N-Z! It’s because many companies release their campaigns to a set number of names in a set number of tranches. Quite often, those tranches are straight alphabetical order.

Whatever the cause, you’ll probably find your list has plenty of valid contacts you haven’t contacted in over a year. Unless you want to warm those contacts back up, it means that they’re ripe for data cleaning.

The first task here is to check that those names are still valid – perhaps with an “Are we losing you?” campaign. You’ll probably see plenty of bouncebacks and undeliverables. Again, this suggests the best people to handle this data cleaning are your in-house team.

4. Take care if importing new names

Clean data importing is harder than you think. Even small differences in data preparation can mean fundamental errors in your database that aren’t easily visible to a human.

(For example, a postman would have no problem delivering a letter to London SE8, but if both words are in the same field in your CRM database, you’ve invented a new town called Londonse8. Many great databases have been RUINED by a single irreversible import.)

If you buy a list and merge it with existing CRM data, it’s wise to talk to an expert first. An outsourced list manager or your CRM partner will know how to format and foolproof your list so it merges seamlessly with your existing data. That saves a lot of problems in the future.

5. Make data cleaning policy as well as practice

The only way to keep your database clean on an ongoing basis is to have a proper data cleaning policy. You will need an assigned person (or persons) responsible for regularly checking for common problems and dealing with them.

That person could be in-house or outsourced, but you’ve got to write the policy first. And it isn’t easy. The policy needs to understand the shape of your database, what success looks like to set its KPIs and the actual actions for performing the cleaning.

(It may even include not allowing anyone to add names to the database without strict rules for doing so. Remember, a small but quality database is more valuable than a large one riddled with errors.)

To this end, engage a professional CRM provider to advise you. They can take a critical look at your data and suggest a range of policies for keeping it clean. Ultimately, it adds up to fewer bouncebacks, higher response rates, and more targeted lists – that’s CRM that works harder for you.

When making decisions about data cleaning, remember:

  • Check what’s worth checking, making sure your sampling’s right.
  • Deal with the simple stuff – it’s the biggest source of errors.
  • Check how warm your leads are and remove those who are coldest.
  • Engage professional help if importing or merging a list.
  • Make data cleaning a policy, not “when you get around to it”!

The ultimate guide showing why slow user adoption affects 49% of CRM projects

This post first appeared on the Redspire Blog.