Why do Data Governance projects fail?

By: David Arenbo

September 30, 2021

For as long as I’ve been working within the field of data analytics (20+ years), there has always been a need for understanding where the data comes from, when it was updated, if it can be trusted, if it is sensitive, who has access etc. There has been lots of attempts to solve all or some of those needs over the years, and I’ve seen a lot of large and expensive projects in the biggest Swedish organizations fail when having a go at it. But failing.

Why?

The common denominator for failure in these projects has in my experience been:

  1. Too much manual work is needed to keep the assets up to date

  2. The “buy in” and time that the non-IT employees need to invest is simply just not there

  3. The ambition is to high when organizations attempt to classify and define ALL systems and data in the organization

Now to the good news! There are a couple of initiatives to address at least (1) and parts of (2) listed above. One of them is Azure Purview, which now is GA and free to try out until the end of October. The solution isn’t perfect, but the connectors and roadmap shows a lot of promise.

More on Perview on a coming blogpost!