Master Data Management and the Art of Politics

digg digg this | | reddit Reddit | StumbleUpon Stumble It!

“Politics, n. A strife of interests masquerading as a contest of principles.” — Ambrose Bierce, American journalist and writer

The organizational, political and cultural aspects of Master Data Management (MDM) and Customer Data Integration (CDI) can be more challenging than the business process and technology aspects.

Since most MDM and CDI initiatives span the entire enterprise by definition, if you’re not careful, you’ll run into a political minefield almost from the very beginning.

Getting funding for an MDM project can be difficult. Most people within your company will probably agree that “if it ain’t broke, don’t fix it” probably doesn’t apply to your critical corporate information assets like customer, product and supplier data. But finding someone fairly high up in the corporation who understands your data management challenges, and cares enough to finance a comprehensive project can be tough.

But rather than go off on a tangent on how to build a business case for master data management, which I’d like to write about some other time, today I’d like to talk about some of the “rookie mistakes” that people typically make in their MDM and CDI initiatives, and some best practices and other recommendations on handling the political side of things.

The first potential mistake is the “what is a customer” trap. While not a bad idea in itself—the corporation should certainly understand, across all of its parts and functional areas, what a customer is—this can be a mistake.

A novice CDI team will typically get fairly senior people together in a room for a “what is a customer” discussion. The conversation may go on for quite a while (I’ve heard of this going on for several months) but the deliverable at the end of the day usually falls flat.

The problem is that most people have a pretty good “gut level” understanding of “what is a customer”, but getting people from Marketing, Sales, Finance, Customer Service, etc. to agree on what a customer is can be almost impossible. And when the discussion gets hairy and people are disagreeing, there’s no mechanism in place yet to resolve those disagreements.

What I suggest instead is to start with a pretty basic view on “what is a customer”. Keep it simple! And don’t try to get sign-off (yet) from every single area of the enterprise. Instead, start with a minimalist approach. Generally, a company will have business customers, consumer customers, or both. Using business customers as an example, you probably need fewer than 10 core attributes from an identity point of view: Business Name, Street Address, City, State/Province, ZIP/Postal Code, Country, Phone, FAX, Web Site.

Obviously, there are variations of the business name (legal name vs. alternatives such as Doing-Business-As, abbreviations, etc.), and you could have both a Physical Address and Mailing Address. Adding contacts to your definition complicates it quite a bit, so leave them out for now. So there are lots of variations and special cases that can take you off into the weeds.

But these nine attributes will probably capture almost all of your business customers. And quickly agreeing to a standard baseline definition for “customer” helps you build momentum and sends the message that your project won’t be that easy to derail.

The second potential mistake to avoid is to begin an MDM or CDI project without a Data Governance program in place. In some cases, you’re actually better off, as far as increasing your chances of success, to think of and to publicly label the project as building a Data Governance program.

Making the business process changes and doing the technology implementation for an MDM platform or CDI hub without a Data Governance program is going to be pretty hard. But having a Data Governance program in place will give you the political framework—a new organization (typically referred to as a Data Governance Council) and a dedicated group of people (data stewards) reporting to it (typically referred to as a Data Governance Office, similar to the Program Management Office most IT organizations have instituted).

It will also give you a head start on developing your business processes and policies & procedures relative to data quality, data ownership, security, identity management, etc.

The third potential mistake is not having sponsorship from senior management. Active executive sponsorship is critical.

Data quality, data governance, master data management, customer data integration—whatever labels you use—this stuff is hard. It’s one of those areas that falls between the cracks in today’s corporations.

It’s everyone’s problem and no one’s problem. It affects virtually every area of the company, but because it’s a cross-functional issue across multiple business units, typically no one “owns” it.

But having active, involved executive sponsorship is a big help. When issues or disputes come up, as they inevitably will, you’ll have someone from senior management to escalate them to. Whether it’s your CFO, CIO or COO (or someone else), having the right executive sponsor can make all the difference.

It needs to be someone (typically) at the “C level” of the corporation, so that person has enough political clout themselves to definitely resolve difficult cross-functional issues. And that person needs to understand and care about Master Data Management and Data Governance (and that’s harder than you might think).

The fourth and biggest potential political mistake is to ignore the politics altogether. Many people would rather pretend corporate politics didn’t exist. It can be as nasty and distasteful as any other form of politics. But your MDM initiative (or Data Governance program, as I’m hoping you’ve “re-branded” it) will not succeed unless you embrace the organizational change management, program management, communications, training, etc. that you’ll need.

This stuff isn’t a “if we build it, they will come” proposition. You’ll have to carefully manage people’s perceptions and senior management’s willingness to fund it initially and be involved (and stay involved over time).

And you’ll need to bone up on your political skills. Done well, MDM and Data Governance can unlock a huge amount of value for the corporation. And that’s only going to be good for you and your career. But done poorly, these initiatives can fail spectacularly.

And it’s usually not the subtleties of business process or technology that go wrong. It’s the political side of things—the Customer Service director who just doesn’t agree with your choice of how to structure the customer master, or the Finance VP who agrees to implement a credit checking policy for new customers, but then folds when pressured by the Sales organization.

You’ve got to be a good strategist and visionary—and politician—to successfully navigate these waters. Good luck, and please comment on this post to let us know of your successes (and your failures)!


  1. Ten Best Practices for Master Data Management and Customer Data Integration « Hub Solution Designs - 10/10/2007

    […] This is true for other types of enterprise technology projects too, but given the amount of political angst that MDM and CDI initiatives typically cause, you need to make sure your project can’t be derailed by opponents pointing to sloppy project […]

%d bloggers like this: