Optimizing Entity Data Quality, Part 1

Published
  • January 17 2008, 4:56pm EST

Undertaking a project to rationalize, cleanse and maintain entity data across an enterprise can be a daunting prospect. My columns this month and next will provide a high-level framework for establishing priorities and creating metrics that you can use to implement an entity data quality enhancement program.

The information that you have about businesses - customers, prospects, vendors, counterparties - across your enterprise is actually a bundle of facts, or attributes, about these businesses. Within your company’s databases, you may have one or several company names, one or several addresses, payment histories and relationship information. Customer service may know that the chief marketing officer at Company D does not want to be called, but sales may not have this important do-not-call information. Tackling the rationalization, cleansing and active maintenance of all this disparate entity data cannot be done as one big project. It would take too long and cost too much before any real benefits were achieved, and maintenance is a never-ending process. You need to make choices about which attributes to enhance first. What is the best way to establish priorities? And, once you begin to implement data quality projects, you will need to measure your success. What are the right metrics for each attribute type?

You need a process to a) document the attributes of entity data across your organization; b) assess the importance of entity data attributes against your company’s strategic goals; and c) create timeliness, accuracy, completeness and consistency improvement metrics that fit the attributes you are enhancing. This month’s column will focus on a and b.

Let’s start by creating a taxonomy of entity data. Here is one catalog of attributes:

  1. Identity data: company name(s); physical address(es); postal address(es), telephone number(s); email address(es); internal identifier(s); external identification identifiers such as a tax identification number.
  2. Individual identity data: names, telephone numbers, email addresses for directors, officers, employees of entities.
  3. Relationship data: corporate family trees, interlocking relationships (e.g., franchises, minority ownership positions), businesses within a common physical location.
  4. Market opportunity data: industry codes, size estimates (employees or revenues), regions/countries.
  5. Financial data: audited and filed financial statements by publicly traded companies or self-reported financial statements (e.g., privately held companies in countries without central reporting requirements).
  6. Risk data: prior payment history, external risk ratings, banking relationships, financial obligations such as long-term debt.
  7. Pricing data: end-of-day equity prices, complex indicative prices for various obligations.
  8. Transactional (or role) data: contacts (sales, customer support, Web site visits), contracts (current and former), products purchased.
  9. Preference data: do-not-contact preferences, preferred shipping or billing addresses, customer nicknames, favorite football teams.

If you agree with the Goldilocks Principle of Data Quality (“just right” quality levels that fit your company’s needs because more quality will cost too much and less quality will inhibit profitable revenue growth), you will tailor your entity data quality enhancement program to match your company’s strategic goals and derivative tactics. Your program will focus on improving the quality of attributes that are most important to goals.
Suppose a major strategic initiative for a company is to sell more products to existing customers. The tactical components of the plan are to use internal telesales teams for outbound calls, to train customer service teams to recognize unmet needs, passing opportunities to the appropriate sales team and to leverage behavioral usage characteristics to create product-line extensions. The attributes of entity data that would be the areas of focus for enhancement and active maintenance for this strategic initiative would include identity, individual identity data and transactional/role data. The telesales team needs high quality identity data about existing customers in order to create the best proposal, transactional data to avoid calling during the renewal cycle for existing products, and individual identity data to have current phone numbers and email addresses for all potential purchasers within the current customer base. The customer service team must have a clear picture of transactional data to understand what the customer purchases, when their contract is up for renewal, their prior customer service interactions, etc. The product development team needs a clear understanding of how customers use their existing products (transactional data) plus market opportunity characteristics of existing customers such as industry code and size by product purchased. The highest priority data enhancement projects for this company should be the identity, individual identity and transactional data elements needed by the sales, customer service and product development teams.

Generalizing from the example, you can establish data quality enhancement programs that match your enterprise’s long-term strategy and short-term tactics. By analyzing which attributes are most important for the success of your company each year, you can successfully allocate your limited resources.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access