The lab spreadsheet, connected to the ERP in real time

The problem

The lab results live in one spreadsheet and the official record in another world.

Two parallel worlds — the real spreadsheet (live data) and the official ERP (data transcribed weeks later by someone copy-pasting).

• Every shift the quality lab measures water absorption, breaking strength, surface hardness, caliber and rectitude and writes them in a spreadsheet on a shared folder. It is quick, flexible and the lab trusts it. The problem is not the spreadsheet; it is that nothing reads it except the person who filled it in.
• That spreadsheet is the real data. The ERP holds a version transcribed weeks later by someone copying and pasting cells, with whatever gaps the copy leaves. A decimal comma moved or a row pasted one line too low is enough to misstate a batch.
• Transcription errors show up at the worst moment: when a customer asks for the declaration of performance of a batch, or when a batch has to be reclassified to another quality class. The answer has to be rebuilt from the spreadsheet, under time pressure.
• Several people log into the ERP's quality module only to type results, and each of them is a named license paid every year. The license is not paid for analysis; it is paid for typing.

How it fits the IRIS system

Connect in listen-inward mode — the folder is read, nobody copies cells.

Connect's "listen inward" mode — watches the shared folder, reads the exact delta when the file is saved, parses it with a grounded language model, validates on screen, pushes to the ERP via API.

The lab keeps its spreadsheet exactly as it is. When it is saved, iLEAN reads only what changed, a supervisor validates on screen, and the results reach the ERP via API. The licenses that existed only for typing stop being needed, and that is the line the CFO reads first. For the lab, the change is simpler: the result exists officially the moment it is saved.

See the full IRIS architecture →

Before and after

Tile lab results on a shared drive versus results posted to the ERP

Aspect	Today	With iLEAN Connect
Lag between lab result and ERP	Weeks	Zero
Hours per month copying cells	Several, every month	Zero
Water absorption per batch	In the spreadsheet only	In the ERP, validated
Declaration of performance data	Rebuilt by hand	Available by batch
The lab's own template	Untouchable, so it is retyped	Read as it is, even as it changes
Named users in the lab module	One per person typing	One, the supervisor who validates

Weeks of lag → zero latency. Hours/month transcribing → zero.

Impact estimate

Impact estimate — to be validated with your numbers.

The block below is an estimate to be validated against your plant's actual data. We put it forward so the committee has an order of magnitude; we refine it during the assessment.

• Estimated payback 4-9 months.
• CFO angle: a typical 50-80% reduction in quality and lab-module ERP licenses, from several named users to one, recurring every year.
• From weeks of lag to zero latency between the lab test and the official record.
• And the hours spent every month copying cells drop to zero, so the lab spends that time testing instead of typing. Customers asking for batch data get it the same day.

CFO angle — typical 50-80% reduction in quality/lab-module ERP licenses (from several user licenses to one). Estimated payback 4-9 months. *Estimate to validate*.

And the fair question from the production manager

“Our spreadsheet has merged cells and notes everywhere — will it hold up?” — that is the expected case: parsing does not rely on fixed positions, it interprets columns, units and margin notes as they are. It is an anchored task, where the best models drop below 1.5% error [1], and a supervisor validates on screen before anything is pushed to the ERP. A value outside the physical range of the test is flagged instead of posted.

^[1] OpenAI paper "Why Language Models Hallucinate", 2025 — on the reliability of AI in anchored tasks.