L2L study repeats what many industrials already know: ‘Data chaos’ is stalling their digital transformations

In all the recent noise about industrial AI, and various companies’ efforts at software solutions to help customers pilot or bring the technology to scale, a fundamental idea should not be getting lost: If data is fragmented, missing or stored in places that require manual processes to retrieve it, AI is going to be of limited or no value to your industrial operation.

Podcast: It's ALL about your data

That said, we have to deliver a mea culpa in this space.

We are very tardy in getting around to spotlighting newsworthy research released last month by L2L, the Salt Lake City, Utah-based vendor of a connected manufacturing operations platform. Sometimes these studies are self-promoting for companies, but this one was revelatory.

The L2L study concluded almost three-quarters of manufacturers—74%—were “trapped” by reporting delays that slow their production lines due to a disconnect between their corporate digital investments—their “digital transformations”—and the reality of the data in their factories.

The L2L research, based on surveys of more than 600 U.S. manufacturing leaders, which purported to uncover this “data paradox,” also concluded that 65% of shop supervisors waste up to 4 hours per shift on manual data entry and reconciliation and that half of the surveyed manufacturers still rely on manual frontline logs, paper trails, and spreadsheets to make decisions—even though, mind you, plants are generating more data than ever via IIoT sensors and automated systems.

Ouch. That hardly sounds like digital transformation to us. More like an epidemic of continuing data siloes, as the L2L research seems to also conclude.

Yet, in the face of all this, as L2L also discovered, manufacturers are pouring millions into new software ($120,000 to $250,000 annually by SMBs and $15 million to $21 million per year by larger companies, according estimates outside the L2L research).

Correlating with the independent data, L2L’s research simply cited a staggering number, that 90% of organizations are increasing their software budgets. We question the ROI, considering all the problems with data, which L2L also reports are leading to productivity declines.

“Manufacturing has a data architecture problem, not an effort problem,” L2L CEO John Davagian said in a release from L2L. “Leaders are investing 20% or more of their budgets into advanced data collection, yet productivity has steadily declined since 2011."

He added: “We’re seeing a ‘digital fatigue’ where complex software adds more friction than clarity. To break this cycle, the industry must shift from systems that simply capture and archive what went wrong to systems that assist and empower the frontline to prioritize and solve problems as they happen.”

In the end, this is all about fragmented, missing, or inconsistent data.

See also: Why industrial AI requires a data ops foundation to scale

All this continues to make the point that manufacturers are trying, with new software purchases, to get to their critical data, but they are only having marginal success. And this does not bode well for their more advanced digital transformation efforts, like predictive maintenance and, yes, AI adoption.

See the last bullet point about the AI readiness gap. To wrap up, a few more of L2L’s interesting findings:

• The hidden cost of this complexity? 58% of respondents report that their current technology stacks create more friction than clarity, with three-quarters of workers forced to rely on multiple, disconnected systems to perform daily tasks.
• “Knowledge attrition.” 88% of leaders report that critical operational information disappears when experienced employees leave, creating a steep learning curve for new hires and threatening long-term standards.
• What's reality? Only 3 in 10 leaders say their operational data reflects real-time plant floor events.
• Digital tools utilized inconsistently. 63% report uneven use of digital tools on the plant floor.
• Cost savings and quality at their fingertips. Savings and quality improvements are the top-cited benefits of better operational data in the research.
• The AI readiness gap. While 87% of leaders believe AI can improve productivity, 79% admit that integration challenges and poor data quality limit its actual impact.