OnLIMS is a LIMS built for mining laboratories — in production since 1989 in mine sites, smelters, terminals and environmental labs — and the laboratories that adopt it mostly come from one place: a folder of spreadsheets. So the LIMS vs Excel comparison is worth making honestly, because Excel is not a bad tool. It is a general-purpose calculation grid being asked to behave like a system of record.
Where Excel breaks in a mining laboratory
- No field-level audit trail A changed gold grade in a cell leaves no record of the old value, who changed it, or when. Track Changes and file versioning do not operate at the level an assay dispute needs.
- No concurrency control Two chemists in the same workbook — or in two copies of it — silently overwrite each other. The one who saves last wins, and nobody is told there was a conflict.
- No QA/QC gate on sign-off A failed CRM in a spreadsheet is a red cell someone has to notice. Nothing stops the batch from being reported with the failure on board.
- Fragile sample identity Sort a range but not its neighbor column, paste over the wrong row, drag an autofill one cell too far — and results now belong to the wrong samples, with nothing to flag it.
What auditors ask
ISO 17025 assessors and the QPs behind NI 43-101 reports converge on the same questions: who changed this value, when, and from what? How do you prevent results from being reported when the batch QC failed? How is sample identity maintained from reception to certificate? A spreadsheet answers each of these with procedure and discipline — which the auditor then has to take on trust. A LIMS answers them structurally: field-level audit trail (old value, new value, user, timestamp), OnQC alarms that block sign-off at a 3SD control failure, results locked on publish, and a Pending, Ready, Approved, Published cycle that shows who approved what.
When Excel is enough
Honestly: sometimes it is. A very small laboratory — low sample counts, one person entering data, no accreditation ambitions, clients who accept results without asking how they were controlled — can run well on a disciplined workbook, and a LIMS would be overhead. The crossover comes with throughput, with multiple analysts working the same batches, and with the first client or auditor who asks the questions above. If you recognize your lab in that second list, the spreadsheet is already costing more than it looks.
What replacing the spreadsheet actually changes
Weights and instrument results arrive by direct capture — serial from balances, file parsers and bidirectional interfaces from instruments — instead of being retyped. Worksheets lay out samples with standards, blanks and duplicates in sequence, and lock against concurrent edits. QA/QC is evaluated as results land, with Shewhart, X-Control, CUSUM and Thompson-Howarth charts — a 2SD deviation warns, a 3SD failure blocks the signature. The daily work looks similar; what changes is that the record defends itself.
Frequently asked questions
When is Excel genuinely enough for a mining laboratory?
When the lab is very small: low sample counts, one person entering data, no accreditation ambitions, and clients who accept results without asking how they were controlled. In that setting a disciplined workbook can work well and a LIMS would be overhead. The crossover comes with throughput, multiple analysts on the same batches, and the first client or auditor who asks for an audit trail.
What do ISO 17025 auditors ask that Excel cannot answer structurally?
Who changed this value, when, and from what? How do you prevent results from being reported when the batch QC failed? How is sample identity maintained from reception to certificate? Excel answers these with procedure and discipline, which the auditor must take on trust; a LIMS answers with a field-level audit trail, QC alarms that block sign-off, and results locked on publish.
What actually changes day-to-day when a lab moves from Excel to a LIMS?
Weights and instrument results arrive by direct capture instead of retyping; worksheets lay out samples with standards, blanks and duplicates in sequence and lock against concurrent edits; QA/QC is evaluated as results land, with a 2SD warning and a 3SD failure blocking sign-off. The work looks similar — the difference is that the record defends itself.
Related: ICP LIMS integration · XRF LIMS integration · Balance LIMS integration · LECO LIMS integration · Instrument integration · Fire assay LIMS · Mining laboratory LIMS