The limitations of standard MDM


Traditional MDM comes with a hefty time burden. It takes months to ingest and transform incoming data. Because its capabilities are not sufficient for the size of the challenge, traditional MDM systems are inflexible and take a long time to gain value from.

The time sink

Traditional MDM systems often run using a fixed data model—meaning all data has to be painstakingly transformed into a single, fixed format (often imposed on you by the technology vendor). And that’s something that the data—multifaceted and highly contextual as it is—was never meant to do. This leads to the dreaded migration issues.

Even once the data’s transformed, companies can expect to wait several more years before seeing any true value or returns on their data. And when that insight finally arrives it tends to be underwhelming.

So it’s no surprise that many MDM efforts tend to be de-scoped, abandoned or put out to pasture—even after heroic efforts to make the technology work.

The quality shortfall

Traditional MDM doesn’t tackle the data quality problem—instead, it becomes another victim of it. The software has just never found a solution to the problem.

Standard MDM solutions were built to become a central location to manage data stores on the basis that they would become a ‘single version of the truth’. But this simple aim can become a huge political challenge in organizations, because it takes control of data away from the business owners of existing applications and asks them to conform in ways that may be worse for their objectives (of getting business value from data).

Ultimately, standard MDM solutions struggle at the first hurdle—they do a poor job of matching and de-duplicating records. And that’s because they’re simply not built for the high volumes of distributed, disparate data generated by the various internal and external applications and sources your records rely on.

Traditional MDM misses matches between records no matter how much manual remediation you throw at it. They are not equipped to handle the inherent data quality issues that come with pulling together multiple sources—nevermind helping you improve it.

And this can lead to serious issues, like markers of fraud being overlooked or missed. Or more benign problems—like missed opportunities for cross-selling.

Your data could (and should) be doing more for you

Everything mentioned above means that the ‘single view of the customer’ promised with traditional MDM solutions isn’t always accurate—or inherently useful.

Standard MDM misses connections, lacks context, leads to bad decision-making and leaves question marks around business value. And, as a result, many organizations have lost confidence in it—because the ineffective output from MDM solutions affects everything from customer experience to operational performance.

But don’t throw the idea of MDM out the window just yet. Because the principles underlying traditional MDM? They all remain hyper relevant.

Your organization does need a single, accurate point of reference for customer data. And it does need to be able to consolidate duplicate records—and discard outdated or incorrect information—in order to deliver the best view of the customer possible.

You just need a solution that offers a better, more reliable, method of delivery.

What are you thoughts on this? Please share by commenting below