Top

4 Master Data Management Mistakes To Avoid

Blog post by on April 11, 2013 No Comments

If ever there was an elephant in the room when it comes to CRM, it would be master data management (MDM).

If ever there was an elephant in the room when it comes to CRM, it would be master data management (MDM).

If companies could avoid using MDM, they’d do so in a heartbeat. But as companies grow more proficient in their CRM usage, they need to tie more information together — customers, products, contacts, pricing. That means designating master data, and ensuring it’s of high enough quality to serve the needs of sales, service, marketing, as well as compliance and accounting.

Unfortunately, corporate data doesn’t naturally play well together. One customer may be present in five different systems. Three of those, however, may be derived from the first system, with one of them located in the cloud. Likewise, customer information may be stored in different ways. The finance system wants formal naming, akin to a Chamber of Commerce registry. The CRM system, meanwhile, caters to salespeople needing to trace contacts into specific business divisions, to prevent duplicate sales efforts.

Managing master data can solve all of those issues. It helps avoid the “garbage in, garbage out” problem, and can provide a concise, trusted view of each customer. But for such projects to succeed, there must be a concrete, “clear and present danger” to the business that requires MDM. That’s because MDM projects require people to settle thorny issues — who owns customer data? who owns product data? — that are almost impossible to agree on, in a vacuum.

MDM Succeeds When Failure No Option

It’s like having a neighbor: you’re not going to pick a fight over the color of a shared fence, if the fence never gets built. For example, at one Cloud Sherpas client, a medical equipment manufacturer, a project team had been trying to rethink products as product masters, and create a clear product hierarchy, to facilitate easier reporting, including automated pipeline forecasts. After three years, however, the project still hadn’t advanced, owing to there being no business imperative.

But as sales slowed, pricing became more difficult to police, and sales teams reported increased incidents of duplicate sales outreach — to the same divisions, in the same company — senior managers mandated that the MDM project need to be completed, as quickly as possible.

Cloud Sherpas came in, held discussions between the relevant stakeholders — marketing, sales, service, and finance — and got all parties to agree about how master data would be handled, including data ownership, in three days. Since the technology to handle MDM was already in place, the company’s IT department was able to rapidly implement the new rules and name changes, and the MDM program quickly went live.

Four MDM Mistakes To Avoid

While MDM projects require a clear business driver to succeed, that’s not all. Based on Cloud Sherpas’ extensive MDM project experience, we’ve also identified four common MDM project mistakes to avoid:

  • Close coupling: The #1 mistake we see in MDM projects is close coupling data between systems. The solution is “loose coupling,” which means ensuring that if you change something in one system, it’s not going to fail, in another system. Because many systems don’t deal well with changes.
  • Standalone projects: Never launch MDM as a standalone project; related success stories are rare. Instead, MDM should be part of an existing project, and made mandatory. That way, as the project advances, so will related MDM and data quality practices.
  • Overkill: Prize simplicity. What’s the minimum MDM necessary? For example, if you’re a manufacturer, must you link products all the way down to the individual SKU level? Perhaps, to assess the revenues each customer generates, you just need to see the level of products that any given customer has purchased, without getting into details about individual product purchases.
  • Heavy clouds: Historically, MDM projects happened on the premises. But as cloud CRM has taken off, salespeople need essential information in the cloud. Show salespeople not just a list of opportunities, but all sales to a given customer in the past 30 days, and they’ll sell more effectively. But to keep the cloud lean, choose wisely about which data to integrate, and which to avoid.

Small Steps Succeed

When beginning with MDM, start small. You don’t need to tie master data into 15 different systems; try just a few at the start, and don’t couple these systems too tightly. For example, say two of your customers merge. Perhaps you’ll want to de-duplicate the record in your CRM system. But you may want to keep the previous records extant in the backend financial system, to support historical revenue reporting, or because they relate to customer service contracts that are still in force.

As you progress, take small steps. Whenever possible, make MDM part of a larger project that gets rolled out in stages, such as migrating to cloud CRM. Use it to keep tabs on data quality, too. Some organizations apply data quality checks whenever new information gets entered, others use dedicated data quality teams to clean data and reduce duplicates.

Mastering these data management essentials can deliver big upsides, not least by implementing MDM practices tied to your business’s specific requirements. Deliver the actionable information that your senior executives and employees — and especially salespeople — need to do their jobs, and you’ll not only increase user adoption of business systems, but also business results. And that’s the best and only reason why you should ever use MDM.

Learn More

Cloud Sherpas is one of the world’s leading Cloud Solutions Providers and helps businesses maximize their CRM returns by identifying desired business goals, finding the right tools and technology for the job, and delivering rapid implementations that remain focused on achieving your desired business capabilities.

Post and thumbnail photos courtesy of Flickr user The Lizard Queen.

Republished with author's permission from original post.

297347

Categories: ! BlogEnterprise Technology
21 views

No comments yet.

Leave a Reply