Normally, people save predictions for the end of the year, but I personally always thought Halloween was a much more appropriate holiday for crystal balls.  Maybe I'm not the only one, because this week, blogger and software industry analyst R “Ray” Wang seems to have peered into the crystal ball and foreseen changes in the master data management market.

Wang lists seven trends he believes will drive change in MDM over the next decade, although some will certainly start to be significant within the next year. Some strike me as inevitable, others strike me as insightful, and a few are a big fat, “meh.”

For instance, given how much governance is an issue for everything related to data, Wang's number five - “Data governance and stewardship more important than ever” - strikes me as about as useful as predicting a horse will win the Kentucky Derby. That's the meh.

More significant, and quite possibly inevitable, are his first and third predictions:

• Vendors must offer MDM tools tailored to specific industries.
• MDM must support a hybrid approach to data, meaning it must be able to handle both on-premise and data from the cloud or SaaS providers.

I also like his suggestion that in the future, MDM styles will no longer matter – just the results. I'm not sure what he means by this. If he means we'll see a movement away from domain-specific solutions and more multidomain MDM, then that trend is already in play. That said, it does seem like it may take more time to catch on, considering how the economy has slowed MDM adoption overall. (For an explanation of the difference between the types of MDM, check out this 2008 article published by Hub Solution Designs.)

I'm 95 percent certain that's what he means, but he could also be referring to the architectural styles used to deploy MDM. Certainly, the needs of the business should shape which architectural style of multidomain MDM you choose, as Initiate Systems' Marty Mosely explained in the August issue of Information Management.

Mosely explains the pros and cons of the three possible styles – registry models, hybrids and centralized models. There are trade-offs with all three, he notes, which is why you need to carefully consider what the business will do with this master data now and in the future:

“Each of these three styles is not right or wrong in and of themselves, only in the context of an organization’s present and future business needs. This is why a careful business analysis is critical for identifying which style may be preferred over another. Since there are always compromises, organizations selecting a multidomain MDM architecture should actively involve an enterprise architect in the process to help them think through the business cases that they will be dealing with both today and tomorrow.”

Another interesting, and perhaps controversial, prediction Wang makes is to suggest MDM will need to to address more data types. I say controversial, because Gartner analyst Andrew White already has objected to this concept, before Wang even wrote his list. After reading another Mosely article, “Multidomain Master Data Management for Business Success,” which also suggested expanding the concept of which data is covered by MDM, White wrote:

"For me, and most users I speak with, one of the most useful parts of MDM is its laser-like focus on just specific data – the most important to a firm.  By deliberately focusing and excluding other data, users get to make some progress with what is a very complex and hard-to-do initiative.  If we include all kinds of data that is referenced, orders, and analytical, and transactional data, we will never get anywhere.”

As it stands now, implementing MDM with just master data is difficult enough. And for the most part, MDM is driven by very practical considerations - such as compliance and regulatory requirements, mergers and acquisitions, and integration of legacy systems. Still, it's always fun to peer into the crystal ball and see what pops up for the future, and as trend predictions go, Wang seems to be more on target than not.