DataOps, Software 2.0 And Other Tech Trends You Can't Afford To Ignore – Forbes

DataOps, MLOps and other “Ops” you should pay attention to
There is nothing worse than a good idea that never sees the light of day.  
Just this past week, Gartner released its Hype Cycle for Emerging Technologies.  The research piece does a great job highlighting the ideas that are spurring innovation at enterprises: “AI augmented software engineering”, “composable networks” and “self-integrating applications” made the list.
McKinsey and Company also published its “Top 10 tech trends” this past week, pointing to ideas like “next-level process automation and virtualization”, “applied AI” and even “Software 2.0”, a trend referring to the scaling of data-rich, AI-driven applications through “machine-written programs”.
Many of these ideas are big and bold. Most sound exciting.  But some could be overwhelming for tech leaders.  The CIOs I talk to confess that they often feel bombarded by such ideas on a daily basis.  If they don’t read about them online, they hear about them from their boss, co-workers or the consultancies their firm hired to help them innovate.
Pilot vs. Production
As BNY Mellon’s Chief Data Officer (CDO) Eric Hirschhorn, so brilliantly describes it in a session on the influence and impact of successful CDOs, many struggle with their “love affair with the Art of the Possible”. In short, they fail to ingest and operationalize innovation in a way that’s meaningful for their organization. 
This statement is not to be taken lightly.  Some of the most popular technology fields are riddled with unfulfilled promises. Take Artificial Intelligence (AI) for instance: 

The inability to connect ideas to impact could be existential for companies and their leaders:  the average Chief Data Officer last less than a 1,000 days, and according to Accenture, “failing to scale AI” could put 75% of organizations out of business.
Pilot vs. Production
More “Ops”, less “Oops”
Thankfully, over the last few years, research in the field of “XOps” has been made more available and organizations are starting to organize themselves effectively: according to Gartner, fewer than 10% of enterprises devised artificial intelligence (AI) orchestration platforms in 2020.  The firm expects this number to climb to 50% by 2025. 
“XOps” refers to at least 4 categories of operationalization platforms: DataOps, ML/ModelOps, SecOps and DevOps.  

Where do you buy yourself these “Ops”?  Nowhere really.  “XOps” is a combination of practices, technology and tools that teams adapt in order to make your strategy a reality.   
More “Ops”, less “Oops”
Hope is NOT a strategy
If you can’t buy “Ops”, you can measure them though.  Any of these “Ops” are designed to bring discipline to your plans, systematize execution and accelerate value creation.
There are at least 3 metrics you should rely on in order to make sure that you’re “doing Ops right”.

You’ll find a plethora of resources online for XOps metrics and practices.  But a good place to start with might be this Eckerson Group presentation (my favorite slide pasted below).
DataOps: How to Get Started
Dale Carnegie was famous for saying that “one hour of planning can save you 10 hours of doing”.  Don’t get mired in too much planning however.  A much more recent book (and one of my favorites), ”Execution” by Larry Bossidy and Ram Charan, is often quoted for this iconic truism: “Execution eats strategy for breakfast”.
There is nothing worse than a good idea that never sees the light of day.

Bruno Aziza is a technology entrepreneur. He has focused on scaling businesses and turning them into global leaders. Bruno has lead startups, medium and large-size

Bruno Aziza is a technology entrepreneur. He has focused on scaling businesses and turning them into global leaders. Bruno has lead startups, medium and large-size organizations across the US, France, the UK and Germany. Some of them include Microsoft, BusinessObjects and Oracle.  Bruno currently works at Google.  The perspectives and views in this column are his own. You can contact him @ [email protected]

source