The MDM Revolution

Share it on Twitter  
Share it on Facebook  
Share it on Linked in  

Break out the Che Guevera berets. It seems a veritable revolution is sweeping the master data management world.


Generally, revolutions are started by the disenfranchised, but in this case, it seems the vendors are leading the way. First there was last week's announcement by Talend that its open source MDM solution would "democratize" MDM. Now, this week comes news that Kalido will bring MDM to the masses-and in the process, free MDM from the tyranny of IT.


"They manage it (MDM) and keep it close to the vest," Kalido CEO Bill Hewitt told IDG News Service this week.


The next release of Information Engine platform promises to change that with an easy-to-use interface, giving average joes and janes a way to browse through the master data using search, data hierachies or categories such as "product" or "sales territory."


Queue the hype-machine orchestra. Voice over: Oh, you lucky huddled masses, yearning to consume master data!


Sorry. Sometimes I can't help myself. Actually, this is a smart move on Kalido's part-and, if the number of Tweets on it are any indicator, it's a popular move as well. Business users do need to have access to master data, but that can be tricky because ... let's face it, no one wants them mucking up data once it's collated and cleaned up. Just ask Tom Kitten about giving his clothes to the Puddle Ducks if you don't believe me.


But what you do want is as many eyes as possible helping clean up the data-and Kalido's new release addresses that as well. When users find a problem, they can fill out a change request form, which is sent to a data steward, who then must submit the change request to a still higher authority, according to the article. This creates an audit trail for change.


The article quotes Merv Adrian, an analyst who gave Kalido's changes a big thumbs up:


"In general, MDM is a specialist play. ... But at the end of the day, if a sizable percentage of your data is not clean you've got serious problems. You can't solve this issue in IT. It's got to be globally."


Essentially, Kalido is focusing on MDM as it's used for business intelligence. Its Web site says as much, noting that Kalido's MDM solution is designed to "serve up the data to support accurate Business Intelligence." This is key to understanding how you would use Kalido's solution.


You may recall, all MDM is not alike. As I shared in February, there's operational, analytical and even - as one reader noted-collaborative MDM. Gartner Analyst Andrew White, who specializes in MDM, recently covered this topic, exploring the business use cases for operational and analytical MDM in a blog post.


A few years ago, Gartner surveyed its end-user inquiries for patterns relating to MDM. There was a very clear "use case" pattern for operational MDM, which focuses on "explicit intent to clean up the business processes that create the bad data in the first place," he writes. As a result, operational MDM involves business users resolving issues with bad master data.


A second, less prominent use case pattern relates to business intelligence. Gartner calls this "analytical MDM." Both types of MDM strive to identify erroneous master data, he writes. But beyond that, analytical MDM does not try to clean up the data at the original source or use the data in any business applications beyond the BI tools-both of which are functions of operational MDM.


But there seems to be a bit of scope creep with analytical MDM. Writes White:


"In the last few years more and more applications have been built on this 'BI data warehouse' that create new data; and this data has to be stored, managed, and governed, like any other business data. As such, this 'BI application' behaves like a business application."


He doesn't seem to approve of this:


"We need to keep things simple, and re-define things. A BI application is one that is used for reporting and analysis only and does not create any new business data; a business application may also have analytics, but it does tend to create business data. For BI applications, analytical MDM is adopted. For BA, operational MDM is adopted."


It would be interesting to see which approach is more popular now, given recent trend assessment by Hitachi Consulting, which suggests that companies are demanding more BI capabilities and MDM is being implemented more on a tactical, rather than enterprise-wide, strategic-basis. I can't help but think there's likely to be more use of analytical MDM.