The lab spreadsheet, straight to the ERP

The problem

Two parallel worlds and a technician copying cells between them.

the spreadsheet is the real source of truth, but the ERP demands manual re-entry, days behind the real data.

• The spreadsheet holds the live data: moisture per batch, ASTA color, heat, granulometry after sifting, and the aflatoxin and ethylene oxide results that come back from the external lab days later. It is the file everyone actually trusts.
• The ERP holds a transcribed, incomplete and out-of-date version of the same numbers. The practical gap runs from days to weeks, and it is the ERP that the commercial team quotes from when a customer asks for a specification.
• Transcription errors surface at the worst possible moment: preparing a certificate of analysis for a customer, or answering a complaint about color drift with a figure that turns out not to match the spreadsheet.
• And several lab and quality people log into the ERP purely to type these numbers in, each one consuming a named quality-module license that the plant pays for every year.

How it fits the IRIS system

Connect in inward listening mode — it watches the file, not the person.

Connect "inward listening" — watches the shared folder, parses with a grounded LLM, pushes to the ERP after human validation. License savings: if several lab staff currently log into the ERP just to enter this data, each consumes a quality-module license; with Connect, only the validator needs one.

The certificate of analysis the customer asks for stops being rebuilt: it is in the system, up to date, on the day they ask. That turns an afternoon's work into a lookup, and it removes the version problem entirely — there is no longer a spreadsheet figure and an ERP figure that disagree about the same batch, with nobody able to say which one is right.

See the full IRIS architecture →

Before and after

Today's lab versus the connected lab

Aspect	Today	With iLEAN Connect
Gap between real data and ERP	From days to weeks	Same shift
Manual transcription	Every batch, every day	Zero
Certificate of analysis	Rebuilt by hand	Assembled from live data
Copying errors	Surface with the customer	Eliminated
Spreadsheet format	Has to be respected	Tolerated as it is, even as it changes
Quality-module licenses	Four or five	One, the validator

manual transcription and days of lag → lab data in the ERP the same shift.

Impact estimate

Estimated impact — to validate against your actual invoice.

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.

• 💰 CFO angle: the ERP bills per named user. If four or five lab and quality people log in purely to enter results, that is four or five licenses: 50-80 % reduction, recurring every year.
• Transcription time recovered from the scarcest team in the plant, which is the quality team that also has to run the audits.
• From a days-to-weeks gap down to lab data available in the ERP within the same shift it was measured, on every batch.
• And ASTA color, heat and moisture history becomes a trend you can act on mid-campaign instead of a folder you consult after a customer has already complained.

50% to 80% savings on quality-module ERP licenses, recurring every year. *Estimate to validate*.

And the fair question from the production manager

"Several people have been changing that sheet for years — does it hold up?" — that is the anticipated case: parsing does not depend on fixed positions, it interprets columns, units, merged cells and margin notes without asking anyone to standardize anything. It is an anchored task, where the best models drop below 1.5% error [1], and the quality manager validates before the push.

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