# Lab correction trail completeness scorecard | SourceX

Canonical: https://refer.sourcex.si/resources/result-to-approved-correction-completeness-checklist
Last updated: 2026-11-04

A result-to-approved-correction trail breaks in four places: a quality event record that does not name the exact sample and report, a correction that overwrites the original value, a corrected report reissued without a new version marker, and copies outside the controlled system that nobody updates. The scorecard below turns those breaks into eight checks a partner can mark pass, partial or fail in one conversation with the quality lead, without seeing a single test value, investigation file or certificate.

This scorecard is part of the [result-to-approved-correction records guide](/guides/chain-result-to-approved-correction), which covers the metadata packet for the same chain. If you want the identifiers behind each check first, read the related page on [result-to-approved-correction record linkage](/guides/result-to-approved-correction-record-linkage).

## How to score a correction trail without seeing a result

Book 30 minutes with the quality lead and, if possible, the person who administers the laboratory information system and whoever controls issued documents. Read each check's question aloud and mark it from the answer you hear. If nobody knows, write "unknown" and score it as fail until someone confirms otherwise.

Each mark carries points: pass 2, partial 1, fail 0. Eight checks give a maximum of 16. The total summarizes the trail for your own notes; the sentence you write from it matters more than the number.

Three rules keep the scoring honest:

1. **Score the system, not the best case.** A lab that links its quality events to specimens "on the big investigations" scores partial, not pass.
2. **Score what exists now.** A planned platform replacement does not raise today's mark. Note the date instead, because migrations are when audit trails and old revisions most often get left behind.
3. **Do not ask for examples.** If anyone offers to show you a case file, decline politely. The scorecard works from structure and counts.

## The eight checks, grouped by what they test

The checks fall into five groups: coverage, linkage, timestamps, outcomes and gaps. Each row says how to check it at metadata level and what earns each mark.

| # | Group | Check | How to check it at metadata level | Pass | Partial | Fail |
|---|---|---|---|---|---|---|
| 1 | Coverage | Result history | Since which year are results and their audit trail held electronically? | Ten or more years, or a documented migration | Electronic, but older years only as printouts | Under three years, or unknown |
| 2 | Coverage | Quality event history | Since which year are quality events held electronically, and can wrong-unit or keying errors be counted? | Electronic with a countable error type | Electronic, error type in free text | Paper only, or not counted |
| 3 | Linkage | Event to specimen and certificate | Does each quality event name the specimen ID and the certificate number? | Both, as fields | Batch or client only | Neither |
| 4 | Linkage | Original value kept | When a result is corrected, does the audit trail keep the first entry beside the new one? | Kept, with a reason that cites the event number | Kept, reason in free text | Overwritten, or corrected before entry |
| 5 | Timestamps | Dated steps | Are value entry, event opened, approval and reissue each recorded with a system timestamp? | All four dated in systems | Approval only on a signed PDF | Only the reissue date survives |
| 6 | Outcomes | Report version | Does a corrected report carry a new revision and reference the one it replaces? | New revision that names its predecessor | New revision, no reference | Same number regenerated |
| 7 | Outcomes | Investigation scope | Does the investigation record which other specimens or lots were checked? | Recorded on every wrong-unit or keying event | Recorded for major deviations only | Not recorded |
| 8 | Gaps | Downstream copies | After a reissue, can the lab show where every earlier copy went and whether it was withdrawn? | Distribution list with each copy's status | Portal copies tracked, emailed PDFs not | Not tracked |

Checks 4 and 7 are where regulated labs usually score well, because the rules ask for exactly those records. For electronic records in closed systems, 21 CFR 11.10 requires secure, computer-generated, time-stamped audit trails and says record changes shall not obscure previously recorded information. For drug manufacturers and the labs testing for them, 21 CFR 211.192 requires any unexplained discrepancy to be thoroughly investigated, extended to other batches that may be associated, and written up with conclusions and follow-up. FDA's 2022 guidance on out-of-specification results describes when an investigation must expand outside the laboratory.

## Why the downstream-copies check often scores lowest

Both rules govern records inside the quality system. Neither tracks a PDF after it leaves. That is why check 8 is the one to ask about carefully.

Take a unit error: a result reported in mg/kg when the method specifies µg/kg. The lab corrects the LIMS record with the original kept in the audit trail, closes a deviation with an approved root cause and issues a revised report. Inside the lab, checks 3 to 7 can all pass. But the first report was already emailed as a certificate of analysis and pasted into a client's trend spreadsheet, and nothing in the lab's systems records those copies.

Ask the question in plain words: "When you reissue a report, what happens to the copies already sent?" Answers tend to fall into three groups:

- **"We keep a distribution list and confirm each withdrawal."** Pass. Note since when.
- **"The portal copy is replaced automatically; we email clients to tell them."** Partial. The lab can show the notice, not the outcome.
- **"We send the new one and assume they use it."** Fail, and an honest answer.

A lab whose history shows the correction and its reach is describing a different record from one whose history shows only the correction. Both can be worth an introduction; the mark tells SourceX which one it is.

## A scored example: a contract testing lab (Illustrative)

A fictional 310-person contract lab testing raw materials and finished products for food and supplement manufacturers. All answers are invented to show the level of detail.

| # | Check | Answer heard | Mark | Points |
|---|---|---|---|---|
| 1 | Result history | LIMS since 2012; migration documented | Pass | 2 |
| 2 | Quality event history | Quality system since 2017; unit error is a picklist type | Pass | 2 |
| 3 | Event to specimen and certificate | Specimen ID field; certificate number in free text | Partial | 1 |
| 4 | Original value kept | Audit trail keeps both; reason field cites the event since 2019 | Pass | 2 |
| 5 | Dated steps | Approval recorded as a signed PDF only | Partial | 1 |
| 6 | Report version | Revision suffix that names the replaced report | Pass | 2 |
| 7 | Investigation scope | Recorded for major events only | Partial | 1 |
| 8 | Downstream copies | Emailed certificates not tracked | Fail | 0 |
| All | Total | Eight checks scored | Sum | 11 of 16 |

How a partner would summarize it: "Electronic corrections with preserved originals since 2019, versioned reports, deviations countable by type from 2017; approvals on signed PDFs; no tracking of emailed copies." That sentence tells SourceX where the joined history starts and where it stops. For a broader view of how buyers judge record quality, see [data quality due diligence](/resources/data-quality-due-diligence); for why correction records draw interest at all, see [how to assess human review records and correction workflows](/guides/how-to-assess-human-review-records-and-correction-workflows).

## Who answers each check, and who can authorize the conversation

The scorecard points to the owners. The LIMS administrator answers checks 1 and 4, the QA manager checks 2, 3, 5 and 7, and the document control lead checks 6 and 8. None of them usually has authority to license records. That sits with the lab director, a general manager or the owner.

Before you introduce, confirm the basics: 50+ full-time employees at peak (contractors excluded), documented operating history, rights to license 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 70% of the data is in English.

What stays unknown until SourceX reviews it: which records are the lab's own to license (client results often need client consent, as the [testing lab licensing page](/industries/testing-laboratory-data-licensing) explains), how client and regulatory confidentiality would be handled, which years a buyer would want and what the lab would be paid. Partners never ask for or handle raw results, deviations or reports. If the lab sits in an investor's portfolio, the [portfolio review handoff checklist](/resources/a-data-opportunity-handoff-checklist-for-a-portfolio-review) helps time the conversation. To see how one correction reads with and without its surrounding records, use the sibling [missing-context walkthrough](/guides/result-to-approved-correction-missing-context).

## Next step

> **Refer a company. No cap on rewards.** Default reward on a company's first qualifying transaction: the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap; further qualifying transactions first agreed within 12 months of the first Closing pay 15%. Eligibility and payment conditions apply, including the USD 100,000 Minimum Transaction, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.

When the QA manager can score all eight checks and say plainly where the trail stops, [register as a partner](/register), confirm the lab on [who qualifies](/who-qualifies) and read the [program terms](/terms) before you introduce the director.

## Sources

- https://refer.sourcex.si/terms
- https://www.ecfr.gov/current/title-21/chapter-I/subchapter-C/part-211/subpart-J/section-211.192
- https://www.ecfr.gov/current/title-21/chapter-I/subchapter-A/part-11/subpart-B/section-11.10
- https://www.fda.gov/media/158416/download
