We use cookies and other similar technologies (Cookies) to enhance your experience and to provide you with relevant content and ads. By using our website, you are agreeing to the use of Cookies. You can change your settings at any time. Cookie Policy.

Master Data Management: Moving Forward...

Originally published February 25, 2010

Between the end of January 2010 and the middle of February 2010, there were three corporate announcements regarding the market for master data management (MDM) tools that have significant impact on the future direction for MDM as a technology:

  • During the last week of January, open source vendor Talend announced the availability of their open source MDM product, based on the technology acquired from Amalto;

  • That same week, Informatica announced that it is acquiring MDM vendor Siperian;

  • The following week, IBM announced that it was acquiring MDM vendor Initiate Systems.
Let’s consider the acquisitions first. Siperian and Initiate Systems have both developed a relatively strong customer base, making them attractive targets for acquisition, and their purchases (respectively by Informatica and IBM) came as little surprise. So as acquisitions have gobbled up the remaining well-known independent MDM product vendors, it is interesting to contrast some of the terminology used in their corresponding announcements.

Informatica’s announcement refers to Siperian as belonging to the “Master Data Management (MDM) Infrastructure technology category.” However, that announcement came just two weeks after Siperian had issued a press release announcing that they had “changed the name of its award-winning MDM platform to Siperian Multidomain MDM Hub to more accurately represent the way their customers are leveraging the solution.” During that transition, Siperian’s product had evolved from being an MDM Hub, to being a “multidomain MDM Hub,” then suddenly became “infrastructure technology.” While Siperian might have viewed itself as a business application, as part of Informatica the product is infrastructure, which suggests a relatively different message than previous positioning had suggested.

IBM’s announcement, while expected, is a little more complex. IBM is already the acquirer of at least two MDM product companies (Trigo and DWL), as well as a nice handful of other vendors supporting data quality and identity resolution (Ascential and SRD). And while Initiate Systems portrayed its product as a “Master Data Service,” supporting data quality and data governance, IBM’s announcement clearly downplayed the general master data capabilities while focusing on Initiate Systems’ footprint  in the healthcare markets, waving the “electronic health record” flag. Considering the broad spectrum of existing “server” branded technologies, one might question the motivation behind IBM’s purchase.

The absorption of these independent MDM product companies by data management megavendors poses some potential issues, both for existing customers as well as organizations currently considering MDM. The first question involves continued support. As companies are purchased, typically there is a time period in which the acquired assets are “integrated” into the branded suite, and this process can take years – consider that Microsoft purchased MDM product company Stratature in the summer of 2007, identity resolution company Zoomix in the summer of 2008, and is finally bringing an MDM product to market in 2010. Will existing clients be able to expect the same level of support from their MDM vendor as these acquisitions close? Will the Initiate and Siperian products continue to be sold even as they are being integrated into the megavendor infrastructure?

The second question is more philosophical: Is it better to get all of your data management capabilities from a single vendor as an end-to-end solution, or be able to pick and choose best-of-breed products from different vendors? The soup-to-nuts solution is the one embraced by IBM (especially in terms of their “Smart Analytics Server” offering), and Informatica’s messaging is increasingly looking that way. However, based on my background in software development, I would think that there are some hurdles to jump in order to truly integrate a collection of purchased applications so that they provide seamless, optimized performance. While big business may have the luxury of purchasing the whole kit and caboodle of software technology, do small and medium businesses now need to commit to the full spectrum of a vendor suite in order to solve their business challenges?

Let’s now consider the Talend announcement. Talend, which provides open source data integration and data profiling tools, is not necessarily targeting the senior executives at the biggest companies, but rather focuses on a market that is more attuned to adopting open source products. This tends to be smaller organizations, with fewer “enterprise product standards,” with engineers who are willing to take on some degree of development. The announced MDM product takes a different tack to master data. Instead of a monolithic repository, the tool allows the data management professional to consider the ways that master objects interact, focusing on semantics and business process. And in my conversations with some of the folks at Talend, they confirmed my thoughts that they are adopting a more pragmatic approach to MDM, one that is used to address specific problems, perhaps from more of a tactical than a long-term strategic standpoint.

So now that MDM has been around for a while, and the master data terminology has drifted into our standard vocabulary, it might be worth stepping back and asking a different question:  Is MDM the revolutionary approach to organizational data consolidation and enterprise information management or is it devolving into yet another  (of many) data management tools?

We have worked for clients who want to adopt the completely monolithic, “pure” approach (i.e., single transaction hub consolidating all master data) and have seen the complexity involved in attempting to find, track, and tie together hundreds of data strands. On the other hand, we have worked for clients who use MDM as a support mechanism for data quality management and governance to help address specific business needs (such as lead generation and management to improve same-customer sales). We have read the press releases for award-winning MDM solutions, and then spoke to the winners only to find their repository is just that – a repository, used as a funnel to continuously circulate the same errors around the enterprise, just another data silo.

Framing the three announcements in that context, one might think that perhaps transitioning an existing enterprise to the pure model for MDM sets a high bar, one that would be difficult to attain without significant investment and support from business sponsors over a long period of time. On the other hand, with careful business analysis, clear objectives, and a clean slate for development, a multidomain MDM environment does not seem to be so out of reach. So let me pose a question back to the readers: Given your experiences over the past year, what is the future of MDM?

Let me know!

Recent articles by David Loshin



Want to post a comment? Login or become a member today!

Be the first to comment!