This article originally appeared on the BeyeNETWORK
In an ideal world, every data-driven marketer would have access to a state-of-the-art customer relationship management (CRM) system, including Web-enabled business intelligence, campaign management and customer touch-point capabilities. Every organization would enjoy the continuous, widespread internal dissemination of complete, accurate and compellingly packaged data. All data miners and marketers, as well as all employees interacting with customers and prospects, would have instant access to all of the data required for them to excel at their jobs. In such a world, smaller firms would operate with the same technological advantages as their larger competitors.
Unfortunately, many companies do not have the budget to invest in cutting-edge CRM systems. Hence, choices have to be made. At such times, it is important to remember “Wheaton’s Law”:
Never scrimp on marketing database content. Without best-practices content, sophisticated, data-driven CRM is impossible.
Characteristics of Best-Practices Content
Best practices marketing database content provides a consolidated view of all customers and inquirers across all channels, including – as applicable – direct mail, e-commerce, brick-and-mortar retail, telesales and field sales. It is as robust as the underlying methods of data collection are capable of supporting.
The complete history of transactional detail must be captured because high-quality content supports deep insight into the behavior patterns that form the foundation for data-driven decision-making. Everything within reason must be kept, even if its value is not immediately apparent.
Best-practices content includes the following four characteristics:
- First, transaction data – all customer events such as the purchase of products and services – must be time-stamped and maintained at the atomic level. Robust event detail provides the necessary input for seminal data mining exercises such as product affinity analysis. Remember that you can always aggregate, but you can never disaggregate.
Within reason, data must not be archived or eliminated. For example, it is difficult to do a product affinity analysis if customer events are rolled off the file every 36 months, for example. Ideally, even ancient data should be retained. Unlike ten or twenty years ago, disk space is cheap, and you never know when you might need the data.
The data “semantics” must be consistent and accurate. For example, descriptive information on products and services must be easily identifiable over time, despite any changes that might have taken place in naming conventions. Consider how untenable analysis would be if the data semantics were so inconsistent that, for example, “item number 1956” referenced a type of necktie several years ago but now references umbrellas.
- Second, post-transaction activity must be kept, such as cancels, rebates, refunds, returns, exchanges, allowances and write-offs. These are essential for important exercises such as identifying the customers who will be less likely to make future purchases without remedial action. After all, customers who are disappointed by unavailable, ill-fitting or damaged merchandise, or poorly conceived and improperly functioning services, will be less likely to purchase in the future.
- Third, ship-to/bill-to (often, gift-giver/receiver) relationships must be maintained. These enable targeted promotions to extend the customer universe beyond those who made the original purchase.
- Fourth, all promotional contacts across all available channels must be kept. This is necessary to rapidly and accurately create the past-point-in-time “views” required for most analytical projects, including predictive models (more on this later). One marquee, multibillion-dollar retailer with a substantial catalog/e-commerce division has learned the hard way the importance of including promotion history. Although it spends seven figures per year on its CRM system, the underlying database does not contain promotion history. As a result, most data mining projects take a week longer than they should because of the extraneous processing required to overcome the lack of promotion history when creating analysis files.
Multiple, Properly Linked Levels
Individuals must be scrupulously de-duped and, for business-to-consumer (B-to-C) environments, properly linked to households. For business-to-business (B-to-B) and business-to-institution (B-to-I) circumstances, individuals must be linked to sites, and sites to companies and organizations. This is essential for the calculation of accurate performance metrics, including promotional financials.
Optionally, database linkages can be supplemented with third-party overlay data to create a complete view of individual customers and inquirers, households, sites, companies and institutions. For B-to-C, the identity of additional adults within customer and inquirer households can be appended, including descriptive demographics such as exact date of birth, age and gender. For B-to-B and B-to-I, additional individuals can be appended to sites, and additional sites to companies and organizations, including “firmographics” such as industry type and number of employees.
Rapid Recreation of Past-Point-In-Time Views
Best practices marketing database content must support the ability to easily and rapidly recreate past-point-in-time customer and inquirer views (“time-0” or “freeze” files). These, in turn, form the basis for virtually all meaningful direct marketing-oriented analytics. For example, they allow the creation of the analysis and validation files required for predictive models. Likewise, they support the creation of the underlying data required for all cohort analysis, including long-term value and the monitoring of changes in customer inventories such as fluctuations in segment sizes over time.
When you do not have the budget to do everything you want, it is time for creative thinking. Successful CRM is attainable as long as you do not compromise on database content. Part 2 of this article will examine two case studies of how this was achieved. Also included will be a cautionary tale of database content that is very far from best practices.
Dig deeper on Business intelligence best practices