Data polishing and floor polishing - it's time for some unfortunate truths - The Industry 3.9 Rut. I'm kinda fired up...
I posted the article below 2.5 years ago. I've been talking about it for over 20 years.
Over the past 3 years, I've been working with teams both in public and private to build a part of the solution instead of just complaining. The problem is that some of the largest providers and analysts don't get it and maybe never will - all while the practitioners continue to struggle. Look at the latests FRED 'output per labor hour chart' - pathetic and mostly unchanged for over a decade.
One solid research team will tell you about "Pathfinders", as if productivity is some newly discovered continent for industrial operations. It's not.
One gov't under think tank sees the one data protocol to rule them all as the path. No, not the one from the 90s, or 00s, or even the 10s, but this time, seriously, really, we mean it.
There's the Data Polishing - you know, that fat old guy like me at the gym that tells everyone after decades and millions of cheeseburger joint frequent eater miles that I'll eat healthy. Really? Decades of allegedly garbage (more on the allegedly later) data and the next 6 months will turn it all around "cuz AI"?
Just for fun, we have the emerging next-gen industrial solution providers that either cobbled together the world largest stack of tech debt, or bought the table stakes of manufacturing to expand their vision.
"What triggered Paul", you might ask?
Two things:
1) a post from one of the largest, most prestigious providers in the space that talked about 'know the process' then went to 'more sensors and more data polishing' is the way.
That will not work. It never has. No epiphanae have occurred to make it so.
2) a LI chat with a lean practitioner that I deeply respect that said 'we'll use tech after we stabilize processes'. Holy moly. 6 years after covid and we still think we can model, map and stabilize (and more floor polishing)?
Every manufacturer making product and collecting checks has a process.
Most think the "Standard"oP is how things are done.
Every practitioner knows, and nearly every provider ignores (trying to be kind) the fact that the real "how" is in the data never even collected, let alone leveraged, let alone AI'd.
Been quiet.
Back on the soapbox.
But I have answers now. And some exist in the marketplace already, though the providers don't even see it, the analysts look past it and the practitioners are too busy putting out fires.
More. Soon.
The Industry 3.9 Rut: The State of Modern Manufacturing
linkedin.com/pulse/industry-…