In part 1, posted last week, I reported the findings of a recent survey which states that the use of SOA (service oriented architecture) is on the rise and also the role of MDM (Master Data Management) in conjunction with SOA.  In part 2, I discuss the argument that, together with MDM and cloud computing, SOA is becoming the method of choice for data management within the industry.

MDM and SOA efforts are starting to come together, led by enterprise architects who realise these initiatives need to coincide.  People designing services are working with data architects and enterprise architects so it‘s all one initiative; you can‘t worry about the data later on.  Previously, SOA people would deliver services but the data coming out wouldn‘t work or would be dirty or there was no validation.  People realised that you have to get the master data people involved from the beginning.  Clever organisations get master data teams and SOA teams together and say ‘Thou shalt work together’ from day one.

The momentum behind cloud computing is helping to move these disciplines together.  MDM is the data pairing to the architectural philosophy of SOA.  A lot of times if companies are pursuing these two elements together they will see symbiotic benefits.  The irony is that if companies are moving to the cloud they‘re kind of getting both; they‘re getting MDM whether they know it or not, and they‘re getting SOA whether they know it or not.  The cloud is really starting to put a merger on getting those two things in harmony.

Security remains a concern for MDM in the cloud.  The master data world has been a little slower to adopt cloud, but it‘s starting to happen.  Master data is among the most private and sensitive data that companies have; it‘s the crown jewels.  It is still early days for MDM, according to the Forrester report, with data governance, data integration, and data quality maturity impacting MDM maturity.  The study found that only 17% of survey respondents felt that they had a high or very high level of MDM maturity, compared to 42% admitting low or very low levels of MDM maturity.

For organizations considering an MDM initiative, expert advice sounds similar to the early days of SOA:  Stop and think through data governance.  Don‘t just blindly charge into MDM thinking it‘s a technology solution alone.  You need a data governance organisation that will control the technology people to ensure a business perspective.  Go data governance, go early; that‘s the war cry these days. You need to do some planning up front, data modelling, data profiling, and start executing policies of data governance.