The morning production meeting runs on yesterday's numbers — reconstructed overnight from shift logs by someone who wasn't on the floor when any of it happened. By the time a two-minute jam or a slow filler becomes a line item in that meeting, the window in which anyone could have done something about it closed hours earlier. That's not a production review. It's an inquest.
In a new piece published on LinkedIn, Tennxt founder Joshua Otomo argues that most factories running below expectation aren't running badly — they're running blind, and blind operations produce the same disappointing numbers as genuinely constrained ones. The two problems look identical on a monthly report and demand opposite remedies.
Late data isn't data — it's history
Operational information has a shelf life. A developing fault caught in its first minutes costs a technician's time; the same fault discovered at end of shift has already cost hours of output. Once that window closes, the fact doesn't disappear — it just stops being something anyone can act on, and quietly becomes a management problem instead of an operational one.
What the shift log doesn't capture
Hand-logged downtime has a predictable set of blind spots: short stoppages that feel too minor to note individually, durations that drift toward round numbers, and reason codes chosen for speed rather than accuracy. The article also points to the "shadow systems" every plant runs on regardless — the operator's personal notebook, the maintenance team's WhatsApp thread, the supervisor whose ear can diagnose a machine before a sensor can. These aren't workarounds; they're a precise map of where the formal system has already failed.
Why aggregate numbers hide the fix
A shift that hits 82% of target could be a speed problem, a reliability problem, or a supply problem — three different diagnoses collapsed into one number that supports none of them. The piece walks through how this same flattening effect turns into one of the more expensive mistakes available to a manufacturer: approving new capacity to solve a shortfall that properly measured losses would have shown was never a capacity problem at all.
Where The Forge fits
Otomo connects the argument to The Forge, Tennxt's manufacturing operations platform, built to surface output, downtime, quality events and losses while the shift is still running — not the following morning. He's also explicit about its limits: it won't fix a line that's genuinely at capacity or substitute for maintenance discipline. What it removes is the excuse of not knowing in time to act.
The full article includes the complete breakdown of what real visibility requires, five questions worth asking before your next capacity decision, and the case for measuring losses before spending on capacity you may already own.
Visibility is not the end goal. It is what allows an organisation to identify the constraints hidden inside its operation.
The same visibility problem exists beyond the production floor. At many facilities, the factory gate is treated as an isolated security point even though it directly affects vehicle flow and yard operations — the factory gate is not a security problem, it is an operations problem.
This approach reflects a broader TENNXT engineering principle: understand the operation before designing the technology — five principles that shape how we engineer.
Visibility can also be limited by the systems surrounding the factory rather than the factory floor itself — why enterprise software projects fail before development begins.
This is a summary. Read the full article, "Your Factory Doesn't Have a Production Problem. It Has a Visibility Problem," on LinkedIn.
