This article originally appeared on the BeyeNETWORK.
One of my previous articles challenged the assumption that cross-selling would be necessarily improved by instituting master data management (MDM) due to latent barriers that exist when it comes to collaboration across lines of business. In essence, the business processes that incentivize product sales within one line of business don’t necessarily promote cooperation across lines of business. However, let’s presume that we have addressed that issue and created a way for the sales staff to be properly compensated when selling across product lines. In this case, how does master data management promote cross-selling?
To answer this, let’s consider what cross-selling is: an offer or a suggestion to a customer for products related to ones that they are already purchasing or considering purchasing. The conventional wisdom is that should someone be willing to buy a product, the customer may be willing to spend additional funds for one or more of a few reasons. Here are some examples:
- The use of the purchased product is enhanced by the use of other complementary products, such as suggesting the purchase of shin guards when buying a soccer ball. The soccer ball is usable alone, but more rugged use of a soccer ball in a competitive arena may require leg protection.
- The product may require add-on services not considered by the customer, such as wall-mounting service for a flat-panel television.
- The customer may have additional needs that are not directly related to the original product, such as credit services to enable a customer to afford to buy the product using longer term financing and incremental payments.
- Customers with similar profiles may be interested in similar sets of products, such as the collaborative filtering suggestions from online retailers (“customers who bought this product also bought…”).
- A customer’s apparent interest reflects patterns common to additional product sales, such as using terms indicating additional interest (“upgrade,” “more information,” etc.).
In most of these situations, one might consider that what is really necessary is common sense, but what emerges is that this works when one has access to the appropriate information. In a small-scale environment, typical customer-facing “lore” can help in driving the cross-sell. The floor salesperson at a consumer electronics store knows that the customer will need help getting that flat TV stuck on the wall, making that pitch an easy one. But in large-scale environments (multichannel web vendors, financial services vendors, office suppliers, for example), the more data that is available, the better, both from a customer profiling point of view as well as from the sales pattern viewpoint.
Consider this: customer profiling and segmentation can drive cross-selling by clustering similar customers and evaluating the items they purchase at the same time (their “market basket”). There is a need for two kinds of data for this type of analysis: customer data and transaction data. The customer profiling collects critical demographic information about the client base (age, residence location, locations of sale, annual income, and other derivable or acquired data characteristics) and clusters those records based on aspects of similarity. The purchase evaluation requires collective statistics regarding transactions – what was purchased, when, with which other products and by whom, etc.
Both of these analyses are enhanced as more knowledge has been collected, and this suggests where the MDM rubber starts to meet the road. Large organizations with many lines of business and catalogs of products may have multiple customer-facing applications through which individuals interact with the company and purchase products; effective customer profiling depends on a unified view of the customer characteristics collected via each one of those sales channels. Master data management provides a framework for unifying that view and enabling a comprehensive profiling process in preparation for the next phase, which includes collaborative filtering and market basket analysis.
In a nutshell, collaborative filtering looks for common purchase patterns among segments. Any time a website (or an individual as well) suggests that the same kind of customers who bought product “A” are also interested in product “B,” the implication is that there is some (perhaps related, perhaps completely irrational) correlation in interest in those products together. Making the purchaser aware of that correlation is a perhaps not too subtle approach to the cross-sell pitch, but one must have access to all of that data in order to properly assess those correlations. MDM, of course, is going to make this much more feasible by providing a master view, perhaps within the data warehouse environment, matrixed across the organization’s sales channels.
Similarly, market basket analysis evaluates the products that customers purchase together to identify any patterns inherent in the data that suggest potential cross-sell opportunities that might not have emerged from common sense. Presenting conclusions of market basket analysis subtly suggest the cross-sell, perhaps even by appealing to core affinities that individuals themselves may not even consciously recognize. Again, master data management provides the methods for accumulating a standardized view of the products and services purchased, which may simplify the market basket analysis processes.
These concepts become especially critical in environments where there is not necessarily a known relationship across products presented in siloed lines of business. The current consolidation in the financial services industry provides a good example. As more traditional banks are acquiring less conventional financial services organizations (such as institutions dealing in credit-backed obligation products, mortgage financing, hedge funds), they will have a growing need to understand how to combine their customer databases, understand who their (combined set of) customers are, what types of products they are interested in, and the kinds of patterns that can emerge from the combination of sales histories. This may be one of the gaps exposed by the absence of a master data environment and provides an area of further exploration for establishing the value proposition for cross-selling.
David is the President of Knowledge Integrity, Inc., a consulting and development company focusing on customized information management solutions including information quality solutions consulting, information quality training and business rules solutions. Loshin is the author of The Practitioner's Guide to Data Quality Improvement, Master Data Management, Enterprise Knowledge Management:The Data Quality Approach and Business Intelligence: The Savvy Manager's Guide. He is a frequent speaker on maximizing the value of information.