Master data management certainly has its benefits, at least according to those who do it well. In a recent Aberdeen Group survey of 176 organizations, the best-in-class companies reported a 15 percent increase in customer retention and a 16 percent increase in customer satisfaction from MDM - plus a 6 percent growth in product sales.
Top Ten Best Practices for Data Integration
Use these guidelines to help you achieve more modern, high-value and diverse uses of DI tools and techniques.
It's impressive, but is a standalone MDM solution really necessary to reap those rewards? Data Doghouse blogger and IT consultant Rick Sherman argues that it's not. In fact, he says, what you're primarily buying with a standalone tool is data integration capabilities that you probably already own in other tools:
The cautionary flag with customer data integration solutions is that they often bundle data-integration software with their offering. In fact, much of the value of the solution derives from the data integration components. ... Maybe you should renovate and revitalize what you have been doing rather than starting from scratch by buying a new stand-alone customer data integration solution.
Sherman is head of Athena IT Solutions and he's been preaching this message about MDM and its aliases - customer data integration (CDI) and product information management (PIM) - for some time now.
In his most recent post on this topic, Sherman suggests you ask yourself two questions before you invest in a standalone CDI solution:
Do the bundled data-integration products overlap with products you already have? Haven't you already started working on customer data integration in your data warehouse (DW) or operational data store (ODS)?
I used his original wording so you can see that the questions are largely rhetorical.
Keep in mind as you read this that CDI products have evolved and are often sold as MDM solutions. Even though vendors will often tell you - and have told me - their products will deal with both product and customer data, Gartner says these so-called multiform solutions are still too immature to be considered a real market yet.
So, if you already have the tools - or products, as Sherman would say - why don't you have a single view of your customers yet? It could be that you haven't gone enterprise-wide with your data integration. Sherman contends it takes an enterprise data integration platform to create a unified view of customers and he's not talking about something you buy. That brings me to the two other components Sherman says are essential to a successful customer data integration program: people and process. He writes:
People and process are critical success factors for any customer data integration initiative regardless of the software that you might purchase. You need an Enterprise Data Management (EDM) initiative if you are truly going to create and maintain a single view of your customers.
Realistically, what that means is the business has to take responsibility for data governance and IT has to establish a repeatable, best-practices approach to integration by setting up either a data integration center of excellence (COE) or an integration competency center (CC), he says.
We've heard this tale before: You can't buy your way out of the data silo problem, whether you're talking about customer data or other data. It's going to take time, a plan and getting everybody on board.