Customer Record Accuracy Depends on Quality External Data
The most powerful and sustainable competitive differentiator in business today is creating a compelling, customized customer experience. To accomplish this, you must know your customer. This sounds like a simple enough achievement. Yet to truly know your customer requires unprecedented levels of quality customer data which, as countless enterprises can attest, can be quite challenging to gain.
The Quest for Data Quality
The quest for high-quality customer data is a never-ending challenge for most businesses. The quest begins with the fundamentals of verifying a few core data elements such as name, address, phone number, e-mail address and gender. When this basic data is accurate, managing the rest of the customer profile becomes much easier. Because this data is so essential, it could be termed "high-impact data."
The challenge is that high-impact data is not stable. In fact, it is highly volatile.Think of something as straightforward as a person's name. It's not unusual to find four or five—even six, in some cases—different, valid combinations of first and last names used for a single person, all in one file. Different names and addresses for the same person can easily corrupt anywhere from 5 percent to 15 percent of the records in a single database.
According to the Census Bureau, in a single year, 2.3 million marriages, 4 million births and 1.1 million divorces take place, amounting to a lot of new names. In an average year, there are 40 million changes of address filed, not including the up to 32 percent of moves that are not reported.
Traditional, software-based data quality processes face severe limitations when working with high-impact data. For one, they cannot validate the complete business record, only its components. In other words, they cannot validate the accuracy of the customer, only the address—and, to some extent, the name.
An accurate address only means you found an existing mailbox. An accurate customer record means you've determined that an actual customer with a rich and detailed history of purchasing decisions occupies a validated, deliverable location. A data-driven data quality management solution uses external data resources to correct, update and validate your high-impact data so that you have truly accurate customer records.
The Power of External Data
The only way to validate a name—a customer—is to know whether there is some type of verified reason for that name to be paired with the address to which it is tied. Using a highly verified set of external data resources to vet a customer file can provide meaningful business measurements for data quality by supplying transactional evidence that a customer lives at an address at a specific point in time.
Screening customer data files with this highly verified external reference data also takes the address-hygiene process to new levels. For example, an address such as "#19867," which would be uncorrectable using typical address-hygiene software, can be corrected to a verifiable address such as "19867 SW 62nd Avenue." This is achieved by using additional customer attributes, such as name and postal code, matching them to a reference data source and then applying the missing address elements, hence vastly improving the underlying quality of the data.
Without a doubt, data quality management is a taxing job. Companies spend years implementing software matching algorithms in an attempt to eliminate duplicate customer records, to create accurate single customer views and to understand customer value. At the end of that process, it seems there is always a wall that is hit where matching algorithms simply cannot resolve every problem. Therein lies the beauty of using data-driven external resources, which can provide intelligent linkages that no matching algorithms could ever achieve.
Data-driven data quality management doesn't require a lengthy implementation period, expensive software or new project teams with external consultants. It can deliver the core information, along with the accompanying data quality scores necessary to build an accurate master customer view, in as little as 30 days.
In a recent assessment undertaken by a major U.S. manufacturer, data-driven, address-hygiene improvements alone increased duplicate account recognition by almost 20 percent. In total, this affected more than $5 billion dollars of customer value, changed 7 percent of the loyalty classification for customers, revealed an entire new population of repeat buyers and gave the marketing department new customer profiles to use in prospecting campaigns.
All of this knowledge had gone undetected with the previous data-quality management process.
The Impact of Data-Driven Data Quality Management
In this economy, where new business is hard to come by, companies are trying to maximize the value of existing customers, so pinpoint accuracy of customer data is more critical than ever. Meanwhile, customers, inundated with advertising clutter and increasingly accustomed to one-to-one marketing campaigns, have little tolerance for a misdirected pitch.
Data-driven data quality management not only corrects, updates and validates your data, but more importantly, it allows you to make a powerful impression on your customer.
Dylan Purse is the product manager of list processing at Experian Marketing Services. He can be reached at email@example.com.