How many full-time equivalent staff does it take to maintain a master data management center? I know it sounds like a joke, but there's no punchline. However, the answer is at least five, according to a recent survey recommendation from The Information Difference-but the average number (mean) of full-time equivalents supporting enterprise-wide MDM is 52.
So, somewhere between 5 and 52 should do it.
The survey looked at the relationship between MDM deployments and data warehouse projects. Although the two are usually discussed as separate and unrelated projects, the survey found that's actually misleading. In fact, there is often what Baseline Consulting partner and co-founder Jill Dyche calls a "symbiosis between data warehousing and MDM." Baseline Consulting was one of four survey sponsors, which I found interesting, given that Baseline's other partner and co-founder, Evan Levy, once told me MDM transcends the data warehouse.
The survey found that almost half (46 percent) of the 208 surveyed had both an MDM implementation and one or more data warehouse implementations. Dyche put this finding in context:
It's not surprising that companies with mature data warehouses are turning their heads to MDM to address data quality and reconciliation issues in a more sustainable way. What is surprising is that MDM is often adopted by data warehousing teams in advance operational applications. This reconfirms that MDM and data governance are being deployed incrementally by the early adopters.
While that information may matter more to vendors and analysts, the survey also includes several findings of interest to those considering or involved with a data warehousing or MDM initiatives.
For instance, most companies describe their MDM and data warehousing projects as at least "moderately successful." When it came to those who had only done MDM, two-thirds ranked the implementations as successful and-even better-only 1 percent labeled it as "unsuccessful." (It's worth mentioning, however, that only 7 percent had actually implemented MDM without also having at least one data warehouse.)
Given the hefty cost of MDM, that should be good news to those of you contemplating MDM.
Data governance and data quality practices also seem to be taking hold, according to the survey. Two-thirds of all organizations have data governance in place, which is up from previous surveys and deemed good news. Data quality is also addressed by many organizations, with 43 percent carrying out that validation in the ETL tool. Meanwhile, 10 percent admit they just load the data into the warehouse anyway, which may explain Evan Levy's recent column on why you shouldn't do this. (For more from Levy on successful MDM, see "Baby Steps to Master Data Management.")
You'll have to fill out a short registration form to receive a copy of the survey by e-mail, but I think it's definitely worth it. It includes a lot of practical information on what works and what doesn't when it comes to data warehouses and MDM. Check it out for yourself.