So you're ready to take on the next game-changing business to business (B2B) marketing analytics project… maybe setting up some marketing tests, perhaps testing a predictive response model, or even thinking of trying your hand at text mining. You start the project by creating a list of your business customers and counting them. Sounds simple, right?

However, as you survey the state of customer data at your company and begin really digging into the details, the question you will likely find yourself asking over and over is this: "Who is a customer, anyway?"

• You discover that Sales aligns customers to geographies. National accounts can be split into East and West territories, or divided across subsidiaries without a need to explicitly tie the accounts together. Sales departments uniquely understand how customer buying decisions are made: for example, whether a retailer makes its buying decisions centrally, or individual stores should be treated as separate customers.
• Marketing is most interested in aligning customers strictly to a legal ownership hierarchy that will make it easy to integrate externally sourced lead lists and automate data processing. It maintains lead lists, customer models, and installed base data using externally available demographic data from companies such as Dun & Bradstreet or Experian.
• And (to throw another wrench into the machinery) the Services organization tracks individual corporate users or sites with entitlement to maintenance services, adding many more "customers" without aggregating the sites in a way that's consistent with Marketing or Sales.

With those different approaches, you realize that there is no simple answer to the question, "How many customers do we have?"

Accurately tracking customers is critically important to the operations of Sales, Marketing, and Services. However, the business can realize much more value if customer data is linked across the company, helping you to really understand your business by enabling…

• Tracking of new customers, to understand where the business is growing
• Cross-selling opportunities
• Customer models and segmentation
• Unambiguous marketing lead lists that tie to sales responsibilities
• Predictive models that prioritize customers for upgrades and renewals
• Customer lifetime value analysis calculations

A Proven Approach

This article offers a proven approach that will enable you to define a customer in a way that's useful across the company.

To define a customer, first think about what you want to know about them. There are three primary categories of customer information:

1. The customer name, location, and contact information.
2. Transactional information around what customers bought, when they bought it, what channel they purchased products and services through, and ideally what site they bought it for.
3. Customer hierarchy data, including numeric keys that show how different sites, affiliates, and subsidiaries tie together for an end-to-end view of the customer and the customer's interactions with your company.

In our example, and very likely at your company, there will be a great deal of information in the first two categories. In fact, the problem with understanding just who is a customer will likely come when customer contact and purchasing information is replicated across organizations, and the third area, the customer hierarchy, is neglected or there is no single standard available to draw together customer information from across the company.

So, in order to count your customers, you need to focus on establishing an effective customer hierarchy.

Building an Effective Customer Hierarchy

Any good customer list starts with individual sites or users specifically associated with your products and services. These people and sites may roll up to regional, national, and global headquarters. A global headquarters may be a subsidiary of a larger organization, with additional layers of regional and global HQ. Your customer list may accommodate some complexity in tracking customers across sites, regional, theater, and global HQ. But who is a customer, really? Ultimately you should be able to count the number of customers over time, track new accounts, and align customers to segments. Your end goal and the complexity of entering and maintaining this data should guide your thinking when deciding if a national retailer is one customer; three customers based on store type or region, or 500 customers—one for each store.

We recommend the adoption of a legal ownership hierarchy, modified in a reasonable and consistent way for the realities of customers' buying behaviors. In the vast majority of cases, the customer is defined at the global headquarters level. This is modified to roll down the hierarchy in cases where the global HQ is truly not a useful way to characterize the customer. And it is modified to roll up some affiliates, franchises, and "broken links" that the legal ownership hierarchy missed.

Sales intelligence is critical to modifying the customer hierarchy away from the strict legal ownership approach. Establishing the proper customer data update processes within marketing and sales operations will play a key role in maintaining the most accurate customer hierarchy.

Here are the critical elements of a customer hierarchy, and the mechanisms you can use to maintain them.

Strict Legal Ownership Hierarchy

This information would be sourced externally from a company such as Dun & Bradstreet and Experian, rolling every customer site to a global HQ. In many cases the site will be the global HQ.

Rolling Down From the Global HQ

It is not particularly useful to aggregate a customer's purchases up to a holding company such as Black Rock or Berkshire Hathaway. We would expect that Berkshire Hathaway-owned companies Fruit of Loom and Burlington Northern Santa Fe Corp., for example, would behave completely differently, based on their unique business needs; and when NetJets makes a purchase for the first time, it should be considered a new customer.

Rolling down from the global headquarters requires a mechanism to modify the hierarchy to roll down from the global headquarters to a domestic headquarters or subsidiary that is a more appropriate characterization of a customer. This is best achieved through a single, centrally managed "roll-down table" that aligns sites to a top level that is below the global HQ.

Accounting for Franchises and Affiliates

Incorporating franchises, affiliates, and subsidiaries with less than 50% legal ownership may also be required to reflect the customer's buying patterns and decision-making structure. Tables of affiliates and franchises may be purchased from external vendors, and adapted to meet your company's needs. Alternatively, this information may be built bottom-up with the support of Sales Operations and Services.

Rolling Up Broken Links

Unfortunately, customer hierarchy information will often be imperfect. External data providers may miss the linkages between sites and you will be reliant on intelligence from the field to avoid "broken links"—sites that do not roll up to their headquarters. More important, this is an opportunity to tailor your customer hierarchy to reflect hard-won discoveries about the real decision-makers, such as the owner of a string of franchises, or about the nature of procurement across local government departments.

* * *

As a B2B marketer, you need to accurately define your customers in order to market to them accurately and consistently. Unfortunately, maintaining the detailed alignment of sites to the true customer can be intimidating to many marketers, and operationalizing the definition... a Herculean task.

However, we encourage you to meet this task head on, following the guidelines provided above, both to ensure you aren't caught off guard the next time you are asked, "Who are my customers, anyway?"—and, most important, to lay the foundation for all your targeted marketing efforts!

The opinions and ideas expressed in this article are the authors' and not necessarily reflect those of their employers.