The State of DataOps In Early 2019

It’s been roughly a year since I started talking about DataOps. It was an accident, something I mentioned during a presentation on data engineering. But that slip attracted the interest of several vendors using the term and I thought I was seeing the start of the next differentiating practice in data and analytics. I’m still waiting for that.

We’re now twelve months on and DataOps still hasn’t progressed beyond its initial hype. There are no meaningful user stories about how DataOps either saved or made money for a company. The stories that are in the market are little more than glorified tales of data integration. There’s a lack of evidence around how organizations are communicating and collaborating differently because they adopted DataOps principles. This makes sense – DataOps principles haven’t emerged because it’s not a being practiced.

The lack of evidence for DataOps doesn’t mean there isn’t value in the concept. Anything that gets companies thinking differently about how they develop, evolve and maintain data used across the company or project is worth exploring. The challenge is getting past the vendor-led hype and towards best practices, changes to organizational design and collaboration models.

The opportunity for DataOps, at least as I envision it, is to fundamentally change how organizations use and share their data assets. We’re a long way off from that. Until there are reproducible best and next practices implemented by end user organizations, DataOps will remain a term vendors use to try to look differentiated in a crowded marketplace.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.