The most popular and comprehensive Open Source ECM platform
Let us know more
Technology: Data Consistency via Master Data Management
Companies are struggling to get their data house in order. Technologies like SOA and data web services are making it easier to pull together data across applications and repositories. So the data is out there and is readily accessible, but synchronizing data elements that are common across all data sources is a huge challenge.
Master Data Management (MDM) has been designed to handle the problem. MDM identifies data elements that are common across many software systems and attempts to keep those elements synchronized. Common elements might include customer, product, location, employee, vendor and asset.
The concept of MDM is a good one, and many companies have piloted MDM projects over the last few years. Now research firm Baseline Consulting says that many companies are beginning to move beyond their MDM pilot systems. Baseline Consulting co-founder Jill Dyche said that “the fact that data quality, data governance, and data enrichment processes may accompany an MDM initiative make it all the more attractive as an enterprise solution.”
Key to a successful MDM implementation is good data, and making sure that it is clean and correct is time consuming. In fact the Baseline report concludes that most projects vastly underestimate the time required to get quality data. As much as 30 percent of an MDM project budget is spent just on data gathering and quality.
What does it take to do MDM right? Most companies that have undertaken MDM have had a median of three million master data records. Projects take six months to implement and are staffed by an average team size of eight persons, with two people from the business side and six from both internal IT and external SI support.
An interesting statistic is that a third of companies that have undertaken MDM have done it at least twice. It’s not clear why they would have done it more than once. Possibly there were multiple pilots, maybe small in size and separated departmentally, or MDM implementations improved and evolved over multiple development generations, or else initial MDM projects just didn’t work out and were reimplemented.
