Tips for Succeeding with MDM or Data Quality Initiatives

Loraine Lawson

SOA may or may not be dead, but one thing is for sure: MDM is certainly booming.


MDM, or master data management, is a strategy for obtaining a unified view of something, usually your customers or products. Gartner classified MDM as one of the fastest growing software segments last year, jumping 24.1 percent from 2007 revenue sales to $1.3 billion in 2008. According to this DM Review article, the MDM market is expected to total $2.8 billion by 2012.


Last year about this time, I interviewed Rob Karel, a Forrester Research principal analyst, about vendors pushing MDM on companies. At the time, Karel thought the economic climate and MDM's hefty price tag would deter that type of sale:

" truth, MDM technologies overall are very expensive. I would be surprised if large MDM investments will be made without extremely high confidence, especially in a year where talks of a potential economic downturn are top of mind."
This doesn't contradict Karel's statement, but in an odd twist of events, Gartner found companies are actually investing in MDM because of factors related to the economy, such as the need to increase business efficiency and reduce costs. Of course, compliance requirements remains a big driver as well.


So far, the big adoption markets are, predictably enough, manufacturing, retail and financial services. Garnter expects MDM adoption to continue in these industries, but also grow in communications, government, health care and utilities.


The DM Review article also notes, however, that more companies are already experiencing problems with MDM because-can anybody guess why? Come on-you've seen this time and time again with big, enterprise-wide initiatives. That's right, boys and girls! The problem projects were either too IT-led or the technology was inadequate.


MDM is nothing to mess around with. It's time-consuming, complicated and - let's not forget-expensive. Ravi Shankar, senior director of product marketing at Siperian, noted in a recent eWEEK article that an MDM deployment can cost more than $1 million, which is one reason it's still largely in the domain of global enterprises.


Shankar argues that as other companies enter the MDM market, the cost will go down, making MDM affordable for more companies. He even lists seven ways MDM can help you reduce IT costs.


Of course, there are two things you should keep in mind as you read these articles:

  1. Shankar works for Siperian, which sells an MDM solution.
  2. You've got to cut a lot of nickels to make an MDM investment payoff, even if the costs do fall below that $1 million mark.
Karel also recently wrote an article offering five tips for succeeding with MDM or, for those of you who can't afford MDM, other data quality initiatives.


He recommended that companies:

  • Find the data pain points and inefficiencies across the enterprise. In other words, go for the low-hanging fruit that will yield the biggest payback.
  • Focus on compliance-related initiatives, which are more likely to be funded
  • Partner with business stakeholders -- a smart tip, given that being too IT-led is a leading cause of failure.
  • Rescue underperforming IT investments. For more on how data quality and huge systems are related, see this earlier post, "Data Integration Strategy Could Be Key to Successful ERP, CRM, CDI."
  • Hone your business case development skills. Always good advice.
MDM is not a cure-all. In fact, it can become just another silo if it's not done correctly, as Shankar shared in an IT Business Edge interview not long ago.


Karel included this warning about MDM in his article:

"MDM remains an immature business capability and technology that offers compelling business drivers but also introduces a great deal of risk. Many aggressive organizations learned the hard way how an expensive MDM investment-if implemented improperly without effective governance-can in effect open a Pandora's Box, propagating less-than-trustworthy data across the enterprise."
MDM shows a lot of promise, and I suspect it will become very fashionable in 2009. But it's definitely not something you'll want to rush. Aligning with the business will be crucial to MDM's success, particularly since the business units-not IT-will most likely own the data involve.


Choose your projects carefully, find a business sponsor, take small, incremental steps, and establish clear metrics for measuring MDM's success or failure.

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


Add Comment      Leave a comment on this blog post
Jan 14, 2009 10:15 AM Cliff Longman Cliff Longman  says:
Hi Loraine,There is a distinction that might help your readers in relation to your comments and observations (which I agree with completely).There is Operational MDM and Analytic MDM. You blog and comments focus (I think) on Operational MDM.Operational MDM seeks to straighten out the data in the operational systems for operational reasons. That's why it is big, risky, invasive, and expensive.Analytic MDM seeks to leave the operational world alone (phew!) and simply concentrate on compiling a view of data that can be used for analytic purposes. Much quicker, cheaper and less risky!I recall last year when I presented at the UK CDI/MDM conference and suggested the attendees would be mad to go "operational first" I felt a bit like an outcast. I don't feel that way right now - to your points, Operation MDM seems to have ground to a halt with some spectacular failures. We are seeing a big uptake in analytic MDM and data governance.Thanks for your blog... Reply
Jan 14, 2009 10:18 AM Vincent McBurney Vincent McBurney  says:
I think MDM vendors have more focus now on getting the foot in the door and have priced the software accordingly. Entry level MDM price are much lower than a couple years ago for transactional or analytical but the price goes up as you add more domains and SOA capabilities later on. You can still end up paying over a million but at least you've worked out whether MDM is helping before you get to that price level.Almost every business with a data warehouse is doing some type of analytical MDM - just with varying degrees of governance! Reply
Feb 4, 2009 11:38 AM Loraine Lawson Loraine Lawson  says: in response to Cliff Longman

Thanks for clarifying that. I think it is confusing. I'm still a bit hazy on how this ties in with the other two types of MDM - the ones focused on customers and the ones focused on product. Or, possibly, companies that did operational MDM have a grasp on it and have moved on to analytical MDM. Last year, Gartner analyst John Radcliffe told TechTarget:

"A lot of the focus to date has been on operational side of MDM. 2008 should see a lot more interest in the analytical side. People can only get their arms around so much at once. MDM is reaching the stage where some of the leading people are getting the hang of operational MDM and are ready to move on to analytical MDM, and then figuring out how to fit them together. ...",289142,sid91_gci1287620,00.html


Post a comment





(Maximum characters: 1200). You have 1200 characters left.




Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.