This worked case is part of the lab result correction histories guide, which covers what to inventory before a LIMS replacement. Here the job is narrower: show one correction from first error to closed corrective action, so you know what a complete trail sounds like when a laboratory director describes one.
The setup: a remapped interface and a clean water stream
Illustrative. A 280-person environmental and industrial testing laboratory analyzes a process water sample for a manufacturing client. The instrument reports copper at 120 µg/L. A few weeks earlier, the instrument interface was remapped during a software update, and the copper channel now writes its number into a LIMS field configured for mg/L.
The report goes out as number 4471, reading 120 mg/L against the client's limit of 1.0 mg/L. On paper the stream is more than a hundred times over its limit. In reality it is well within it. The client may already be acting on the first version.
Seven steps from the wrong value to the closed corrective action
Each row is a decision or event, the person who made it, and the record that captured it.
| Step | What happened and who decided | Record left | System |
|---|---|---|---|
| 1 | The instrument run produces 120 µg/L with passing calibration and quality control | Raw data file, timestamped | Instrument data system |
| 2 | The interface writes 120 into a mg/L field; the analyst releases report 4471 | LIMS result entry and audit trail; issued report | LIMS |
| 3 | The client's environmental manager emails to ask why a normally clean stream failed | Client inquiry | Email or client portal |
| 4 | The reviewing analyst compares raw file and report and finds the unit mismatch | Reviewer comment on the sample | LIMS |
| 5 | The quality manager opens a nonconformance, traces the cause to the remap and lists other affected samples | Nonconformance record NC-0412 | Quality system |
| 6 | The laboratory director approves amended report 4471-A1, reading 0.12 mg/L, with the reason stated | Amended report referencing 4471; audit trail of the change | LIMS and report archive |
| 7 | The quality manager re-validates the interface mapping and adds a unit check to second-person review | Corrective action record; revised procedure | Quality system and document control |
The original and amended reports side by side show why the chain matters.
| Field | Report 4471 | Report 4471-A1 |
|---|---|---|
| Copper result | 120 mg/L | 0.12 mg/L |
| Client limit | 1.0 mg/L | 1.0 mg/L |
| Interpretation | Exceeds limit | Within limit |
| Reason stated | Not applicable | Unit error: µg/L value entered as mg/L |
The decisions that give the trail its value
Three decisions turn a fixed number into evidence of how the lab reasons.
The reviewer went back to the raw file (step 4). Instead of re-running the sample or trusting the LIMS value, the analyst compared the instrument's own output with the report. The comment records the reasoning, not just the answer.
The quality manager looked beyond one sample (step 5). The nonconformance names the cause and lists every other result written through the remapped channel. That turns one correction into a bounded investigation with a scope anyone can check.
The amendment was issued as a new document (step 6). Report 4471 still exists; 4471-A1 points back to it and says why the number changed. That matches how NIST summarizes clause 7.8.8 of ISO/IEC 17025:2017. A clinical lab would add CLIA duties on top: tell whoever ordered the test, issue the corrected report promptly, and keep both versions for the required period. This page is general information, not regulatory advice.
Why this sequence is useful evidence
A status field would say "amended." The chain says what went wrong, how it was caught, who decided, what else was checked and how the process changed. That sequence of problem, authorized action and checked result is the kind of record AI developers describe needing when they study correction workflows; how to assess human review records and correction workflows describes the general pattern, and exception handling data for AI agents explains why exceptions carry more signal than routine runs.
The chain is also fragile. The raw file may sit on a workstation nobody backs up. The unit setting that caused the error is a configuration value, so a system change can keep 0.12 mg/L while losing any sign that it was once 120. The client's email may exist only in one person's inbox. The system map for correction records shows the keys that hold the steps together.
Recognizing similar cases at a client
Laboratory directors rarely call this "exception data." Listen for these signals in ordinary LIMS work:
- A recent interface remap, instrument swap or software update that touched units or methods.
- Reissued reports that carry a suffix such as "-A1" and point back to the first version.
- Investigations that list every other sample run through the same faulty step.
- Second-person review that leaves written reasoning, not only an approval tick.
- Clients who query results by email, and a team that files those queries against the sample.
Then confirm, at metadata level only, what exists: roughly how many reissued reports per year, whether investigations name affected samples, who owns each system and whether any change of system is scheduled. Never ask for reports, raw files or client names. Rights differ by record: a contract lab's results usually belong to its clients, while its review comments, nonconformance files and validation records are often the lab's own. The testing laboratory data licensing page covers that split, and the who can license decision tree shows whom to bring in.
The lab needs 50+ full-time employees at peak (contractors excluded), a documented operating history, rights to license the records and an authorized contact. Companies in the U.S., Canada and the EU qualify, as do companies in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia when at least 90% of the data is in English. What stays unknown until SourceX reviews it: which record classes the lab may license, what must be removed or de-identified, SourceX's current data needs and any offer.
Next step
Refer a company. Earn up to $100,000. Default reward: 25% of Actual Supplier Payout, capped at $100,000 cumulatively per referred company. Eligibility and payment conditions apply, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.
If a laboratory client keeps correction trails like this one and the legacy systems that hold them have an end date, register as a partner, check the lab against who qualifies and read the program terms before you raise it with the director.