Reading Batch Records
Without Missing Critical Details
The skill pharmaceutical employers assess in every interview — and the one most training programmes never actually teach. Work through four practical stages and build the habit that separates thorough QA reviewers from good ones.
Everyone talks about what you need to get a QA role in pharmaceutical manufacturing. Certifications, qualifications, experience. That side of things is well covered.
But what about the abilities you need to do the job? The skills that determine whether you are good at it — not just qualified for it?
You are not reading to understand how the product was made. You are reading to verify that every required element is present, complete, and consistent. Those are different tasks.
The issues that cost companies the most in FDA inspections are rarely the obvious ones. They are the quiet ones — the verifier who signed the day before the activity took place, the deviation raised but never closed, the yield that reconciled on paper but should not have. You only start catching those when you stop reading and start verifying.
This module builds that skill in four stages. Work through each one before moving on.
Before you can spot what is wrong, you need to know what belongs. Every batch record is built from the same core sections — each one exists for a reason.
Product name, batch number, batch size, manufacturing date, expiry. The who and what.
Raw material lots, quantities dispensed, CoA references, reconciliation. Every input must be traceable.
Equipment used, parameters set, actual readings, in-process checks. What was done and by whom.
Each critical step requires a performer signature and an independent verifier. Both must be present.
Temperature, humidity, cleanroom classification. Conditions during manufacture must be within limits.
Theoretical vs actual yield. Unexplained yield loss is a red flag that must be investigated.
Authorised Person sign-off, status labels, certificate of conformance. Nothing ships without this.
Any departure from the master batch record must be documented — even minor ones.
Each excerpt contains a single planted issue. Write your finding before revealing the answer. The act of committing to an answer is part of the training.
This excerpt contains three issues of different severity. Some are obvious. One is deliberately easy to miss. List all three before revealing the answer key.
Spotting errors in training is one thing. Doing it consistently under production pressure, batch after batch, is another. This is the systematic habit that makes the difference.
- 1Product identity confirmed — batch number, product name, and batch size match the master batch record exactly.
- 2All process steps are present and completed — no blank fields, no skipped steps, no see attached without an attachment.
- 3Dates and times are chronologically consistent — no verifications before the activity, no impossible sequences.
- 4No self-verification — performer and verifier are always different people for every critical step.
- 5All in-process results are within limits — OOS results have a deviation raised, not simply ticked as passing.
- 6Yield and reconciliation arithmetic is correct and within the acceptable range — every discrepancy is explained.
- 7All deviations raised have a complete loop — root cause, impact assessment, disposition, and CAPA decision all documented.
- 8Environmental records are complete — no N/A for mandatory monitoring parameters.
- Most citedIncomplete or missing records — entries left blank, steps not signed, no timestamps
- Most citedData integrity failures — backdated entries, altered records, implausible sequences
- FrequentInadequate investigations — root cause not substantiated, CAPA not raised or not effective
- FrequentFailure to investigate OOS results before releasing product
- CommonSelf-verification — same person performing and checking a critical step