Last week, I shared a great post, written by Robin Bloor, president of the Bloor Group and a partner with Hurwitz and Associates, that described precisely why data integration is so gosh-darn hard. Lack of standards, applications built without considering data sharing, and a lack of a metadata warehouse all made Bloor's list, along with seven other causes.
The point of the list wasn't to solve those problems, many of which are unsolvable at this point anyway. It was just to explain something that some of us, myself included, have long wondered, which is why data integration continues to be such a pain in the patootie.
It does beg the question, though: If that's the stuff that makes data integration hard, what makes it easy, or, at least, easier?
It turns out, the answer appeared earlier this month in a SearchDataManagement.com article. The article is about Composite Software's recent Data Virtualization Day and what speakers there had to say about data integration. About halfway into the article, I found the answer to easier data integration, according to Ted Friedman, a vice president at Gartner and a member of the research firm's information architecture team.
Friedman says the five keys to making data integration both successful and easier are:
He's based his opinion on surveys and conversations with Gartner clients. The second half of the article is dedicated to explaining what he means by each of these, and it's not exactly what you'd expect. For instance, by "standardization" he doesn't mean necessarily adopting standards, but that "organizations should focus on repeatable processes and approaches for dealing with data-integration problems," the piece explains.
That governance piece is interesting. It seems many organizations are getting it right when it comes to data quality, security, privacy and so on-but still missing the whole governance piece of the data-integration puzzle. Friedman specifically warned governance is "certainly an insurance policy, in a way, to get the optimal value out of all these investments."
Of course, as has been noted here before, getting the right people to care about data governance is easier said than done. But if you can do it, it's nice to know it'd at least make integration easier.