How to Attack Your Data Defects
To facilitate a successful data quality initiative, you will need to attack the causes of poor quality data head-on. Inevitably this will involve the use of data quality software, but you will achieve greater success by formulating your strategy in such a way that you attack all of the defects polluting your database.
Here’s a 6-step guide to completing a successful assault on poor quality data:
1. Identify and Resolve
The first step is to scan your system for account and/or contact errors. The more mistakes you have in your database, the harder your team has to work, and the more waste is generated.
Once data defects have been identified, they can be corrected using data quality software. Deduplication software will identify records that can be merged together to eliminate waste; addresses will be checked for exact matches, phonetic matches and so on (matched), then combined (merged) to form a single, accurate record.
2. Change Your Processes For Data Capture
Once your data is clean, it’s time to look at the processes your employees are using to update the database. When records are entered, data could be incorrectly formatted, or the person taking care of data entry may neglect to do the proper searches before adding a new record.
If you can change processes, and/or adapt your systems, you may prevent some duplicate data being added at the point of data entry. While this will not negate the need to keep records up to date, it will certainly help to maintain a higher standard of data quality overall.
3. Correct and Verify
When a customer tells you their address, do you make any effort to check the details or spelling? If a street moves into another postcode zone, do you expect that your customers will tell you – or will they have better things to do?
Address verification is an easy way to improve data quality and cut down on duplicate records. It also means that a large amount of mail sent to customers will be delivered, not returned, since their address will match the records held by the Royal Mail.
4. Suppress Invalid Entries
As your database matures, it should become more valuable as a business asset. Yet contact information becomes less accurate as it ages, so your database is inherently subject to problems over time.
Data quality is all about scanning a database in a variety of ways, and invalid entry suppression is a very powerful method. It checks for people who are deceased, ‘gone away’ (moved house, or similar) and – if relevant – checks for companies that have merged or gone out of business.
5. Use DMA Preferences
The Direct Marketing Association (DMA) is a free service that gives consumers the power to control the emails they receive. Upon registering, the customer’s details are added to the Mailing Preference Service (MPS) Consumer File and excluded from future marketing lists.
All businesses keeping contact records for marketing purposes should ensure their database is regularly checked against the MPS Consumer File.
6. Fill In the Blanks
The last stage in attacking low quality data is to fill gaps in the records you’ve cleansed. This might mean adding data about the customer’s age, gender, income and so on. In business to business marketing, you may wish to include details about key decision makers.
Filling in blanks is the best way to ensure a database can be segmented effectively, thus paving the way for the holy grail of data quality – the single customer view.
What’s the Prize?
Improving data quality results in substantial financial gains, partly because the business avoids postal waste, data scrap and duplicated work. But the biggest benefit to the business is improved insight and the ability to make better decisions. This results in better customer care, lifetime value and greater share of the customer wallet over time.