I’m a fairly boring and stable statistic when it comes to my personally identifiable information. I have only married once and moved four times over the past 25 years – all within a 10-mile parameter within the same city. But consider that 2.96 million people move every year. Why do they move? Some for job changes, some due to marriages, divorces, and others to be near aging parents. Whatever the reason, moving is just one of many causes for an annual overall contact data decay rate of 20 to 30%. (To see the full range of contributors to data decay, check out the Data Detox: The Fresh Face of Data infographic.)
Back to me. Even though my physical parameters were contained, the change still included my last name, at least three email addresses and internet providers, changes in cable service providers and multiple life stage changes from newlywed, new home owner, parent and pet owner to a distant view of an empty nest now in sight. A brand who didn’t keep up wouldn’t have a snowball’s chance of speaking to me in a meaningful way.
So how does a brand proactively maintain quality data? By starting with the four C’s of data quality – Correct, Current, Complete and Ethically Collected.
Step 1: Is Your Data Correct?
Normalizing your data means making sure that all of the data elements of the same type (all of your phone numbers, all of your home and email addresses, etc.) look like one another and follow the same format. Phone numbers, for example, serve as a great model.
For example, all phone numbers in the United States contain 10 digits. The first step is to check that each one has the proper number of digits. Then, we would also want to ensure the phone numbers were all presented the same way such as 123.456.2890.
Next, we want to check the data on a fundamental level to see if it really is correct. To continue with the phone number example, you must check to see if the phone number is actually in use and if if the number you have for Mary Smith really belongs to Mary Smith.
Step 2: Is Your Data Complete?
Once your data has been normalized and you are confident it is accurate, you can now examine your data to ensure it is also complete.
For example, if we notice the data element has 9 digits instead of 10, we can look further to identify if it is simply missing a number or if, perhaps, the wrong information was stored as a phone number when it’s really an address or a customer identification number.
The next check is if the information falls within an accepted range of quality. For example, a phone number can be checked against known rules to see if the area code is valid. Or a ZIP Code can be checked to make sure it has the right number of digits and that it matches the city or town also listed.
Step 3: Is Your Data Current?
Your data is now correct and complete, but is it current? Does Jennifer Jones still live at the address you have or has she moved? Is her last name still Jones or was she married last year and took her husband’s last name? Is her phone number still correct?
Recognizing a consumer’s current information helps businesses avoid embarrassing and costly mistakes – like sending one household four copies of the same catalog—one addressed to you, another addressed to your spouse, a third addressed to your mother-in-law and a fourth addressed to your teenage daughter. Yikes. The identity resolution process saves money and helps create stronger customer portraits by merging multiple customer records to create a more comprehensive picture of each customer.
Step 4: Is Your Data Collected Ethically?
When it comes to practices regarding your data, you should go above and beyond what is required by law. It’s key to building a trust-based relationship with customers to include policies and procedures that ensure data uses are fair and just to consumers. Some rules of thumb include:
- Data should be used to benefit both parties in a transaction.
- Every data supplier and data source should be vetted to help keep data quality high, and checked on a yearly basis.
By building repeatable front-end processes to ensure the ethical collection and use of data, you create a win-win situation for your business and consumers.
To combat data decay, marketers want to start cleaning existing data – today. They also want to clean data early in the process as it comes in. Correct records before they get in your customer database. Once the data becomes “official,” it becomes exponentially more expensive to get rid of.
Finally, because data quality decays so quickly, data must be processed regularly to minimize data quality issues. Determine the frequency and process required to keep your data correct, current and complete and find the solution to make it happen.
Take a look at the reasons data goes stale and how to combat it. Hopefully this can help reduce the challenges you face in keeping data correct, complete, current and ethically collected.