Last updated: October 6, 2026
By Dave Manchester, Scientific Recruiting, KORE1
Your packaging line finished a batch on Tuesday. When does it ship? Batch record review is the quality unit’s check that every production and control record for a batch is complete, correct, and explained, and under FDA rules that check comes before the batch can be released or distributed.
So the honest answer to “when does it ship” is “when somebody qualified has read the whole record.” Not when the last carton is sealed. Not when the truck shows up.
I recruit scientific and quality staff through KORE1’s life sciences staffing practice, and the batch record reviewer is the seat I see undervalued most often. It looks like paperwork from the outside. It isn’t. Then it goes empty for six weeks and the warehouse fills up with finished product nobody is allowed to sell.
This post is about the people. Who’s allowed to review a batch record, who’s allowed to sign the release, what the seats look like, how to screen for them, and what it costs you when one of them sits open.
What Batch Record Review Actually Checks
Batch record review is a line-by-line examination of the executed record for one batch of drug product, checking that every step followed the approved master record, every entry is signed and dated, every result meets its specification, and every discrepancy has been investigated before the quality unit approves the batch for release.
The executed record is big. Really big. Under 21 CFR 211.188, it has to document each significant step in making, packing, or holding the batch, and the regulation spells out what that means. Here’s a partial list of what a reviewer is reading.
- Dates, and the identity of the major equipment and lines used
- Which lot of each component went in, and how much of it was weighed or measured
- In-process and lab results
- The line clearance, meaning the inspection of the packaging and labeling area before and after use
- Actual yield, and the percentage of theoretical yield at each phase where the master record calls for it
- Specimens or copies of every label used
- Who performed each significant step, and who checked it
- Any investigation that happened along the way
Then there’s the yield. The master record sets a maximum and minimum percentage of theoretical yield for each phase, under 211.186, and a number outside that window isn’t a typo to initial and move past. It triggers an investigation. Every time.
The rule that ties all of this together is 21 CFR 211.192. All production and control records, “including those for packaging and labeling,” have to be “reviewed and approved by the quality control unit to determine compliance with all established, approved written procedures before a batch is released or distributed.” Any unexplained discrepancy has to be thoroughly investigated, and the investigation has to extend to other batches of the same product, plus other products tied to the same failure. It’s also one of the sections FDA investigators cite most often, which I dug into on our pharmaceutical recruiters page.
Reading the record is the easy half. Knowing what a missing entry means is the job. That takes years.
Who Is Allowed to Review It, and Who Signs the Release
Two different questions. Plants get into trouble when they treat them as one. It happens.
FDA gives the quality control unit the authority to review production records “to assure that no errors have occurred or, if errors have occurred, that they have been fully investigated,” under 211.22, and the final approval in 211.192 belongs to that unit. Nobody else. Not the plant manager. In practice, a lot of finished-dose plants run it in two passes. Production does a first read for completeness, often the shift supervisor or a production reviewer, and then pharma QA does the review that counts. If you’re fuzzy on where QA ends and QC begins, I wrote up QC vs QA in pharma separately.
For active ingredient plants, the ICH Q7 guidance for active pharmaceutical ingredients draws the line more explicitly. Records of critical process steps get reviewed and approved by the quality unit before an API batch is released. Records of noncritical steps “can be reviewed by qualified production personnel or other units following procedures approved by the quality unit(s).” And Q7 lists reviewing those critical-step records among the quality unit’s main responsibilities that “should not be delegated.” Same section, one more line I’d underline. “All deviation, investigation, and OOS reports should be reviewed as part of the batch record review before the batch is released.” OOS means out of specification. Read it twice. An open investigation is a held batch.
Then Europe. If your product is released for sale in the EU, the signature isn’t just a QA manager’s. EU GMP Annex 16 says each batch of finished product “must be certified by a QP within the EU” before release, meaning a Qualified Person, and every manufacturing site in the EU needs at least one. The QP can hand the checking work to “appropriately trained personnel or third parties,” including confirming that “all records are complete and endorsed by appropriate personnel,” but the certification stays theirs, and Annex 16 expects them to rely on the quality system with “on-going assurance that this reliance is well founded.”
Translation? The QP is only as good as the reviewers underneath them.
The Seats Behind One Release
One released batch usually has more than one name behind it. Here’s how I’d map the seats. The last column is my opinion, not a regulation.
| Seat | What they review or sign | Where good ones usually come from | Yours or borrowed? |
|---|---|---|---|
| Production reviewer or shift supervisor | First pass for completeness, missing entries, and obvious errors, before the record leaves the floor | Senior operators who know the process and write cleanly | Yours |
| Pharma QA batch record reviewer (often titled QA specialist or QA associate) | The full review against the master record, yields, labeling, and attached deviations | Former operators, QC analysts, and document control staff with a careful eye | Yours for steady volume; contract for a backlog |
| QC data reviewer | Second-person review of the lab results that feed the release, before they reach QA | Experienced QC analysts | Yours; contract during stability peaks |
| QA investigator or deviation owner | Closing the deviations and investigations that are holding a batch | Process engineers and QA specialists who like root cause work | Contract works well when the queue spikes |
| Head of quality or QA manager | The release decision and the procedures every reviewer works to | QA managers who have already released commercial product | Always yours |
| Qualified Person (EU release only) | Certification of each finished batch for the EU market | EU-based QPs, sometimes through a partner site | Depends on where certification happens |
Small companies blur these. That’s normal. One person reviews, investigates, and releases, and that works fine right up until that person takes a vacation. The leadership version of that problem, the moment one quality person stops being enough, is its own question, and I worked through it in timing a director of quality hire.
What an Empty Reviewer Seat Costs You
Here’s the part nobody budgets for. The queue.
When the reviewer seat goes empty, production doesn’t slow down. The line keeps running, the batches keep finishing, and every one of them lands in quarantine waiting for a review that isn’t happening. You’ve already paid for the materials, the labor, and the testing. The product is sitting right there. You just can’t sell it. Or ship it. Or invoice for it. The finished goods pile up in a cage while the people who would normally clear it are covering two other jobs, and each week the pile is a little taller than the week before, because the backlog doesn’t stay still. It compounds.
Then the deviations. ICH Q7 has deviation and investigation reports reviewed before release, and Annex 16 asks the QP to confirm that investigations on the batch are complete enough to support certification. So a backlog of unclosed deviations turns into a backlog of unreleased product even when the reviewer is at their desk. If the same corrective and preventive actions (CAPAs) keep reopening, that’s a systems problem, and it’s the kind of work our quality engineering staffing desk fills alongside the GMP quality seats.
For a contract development and manufacturing organization (CDMO), it’s worse. The batch belongs to a client, and the client has a launch date, a clinical supply schedule, or a distributor waiting. A slow release becomes a client relationship problem in about two phone calls. Maybe one. If you staff a contract site, my CDMO primer for hiring managers is worth a read too.
And an inspection doesn’t wait for you to catch up. Records have to be “readily available for authorized inspection,” per 211.180. A stack of unreviewed records is the first thing an investigator will ask about.
None of this shows up on the job requisition. It should. Put it there.
Paper Records, Electronic Records, and Review by Exception
The reviewer you need depends on what the records look like. Paper? Electronic? A mix?
On paper, the job is attention. A reviewer turns hundreds of pages looking for a missed initial, a late entry, an overwritten number, a line clearance that wasn’t signed. Some people are wired for it. Plenty aren’t. That’s fine. Screen for it instead of assuming it comes free with a GMP background.
Electronic batch records change the work. A well-built system enforces the sequence, checks entries against limits as they’re captured, and can support what the International Society for Pharmaceutical Engineering (ISPE) calls review by exception, “an approach in which manufacturing and quality data are screened to present or report only critical process exceptions as required by approvers,” as quoted in Pharmaceutical Technology. The reviewer stops hunting for typos. They start judging exceptions. Harder work. They also need to trust the system enough to know what it already checked.
That’s a different hire. Someone who has only ever reviewed paper can be slow at an electronic site at first, and someone from an electronic site may never have built the paper-reading reflexes. Ask which they’ve done. Then ask for an example. The people who build and validate those systems, the manufacturing execution system (MES) and electronic batch record specialists, are a separate search that runs through our pharma IT staffing team.
How I Screen a Batch Record Reviewer
I talk to candidates the way I’d talk to a friend, and the first thing I want to know is whether they actually like this work. Some people find careful review deeply satisfying. Others took the QA desk job to get off the floor and are counting the months. Both can pass an interview. Only one stays good at it.
My favorite screen is practical. Hand finalists a short mock executed record with a few problems planted in it, every company and product detail made up, and give them twenty minutes. Good planted problems include a yield that falls just outside the master record’s range, a component weighing with no second person’s check, a correction that hides the original entry, and a deviation number referenced on a page with no closed investigation behind it.
Then ask what they found. Ask what they’d do about each one. That second question tells you more.
Here are a few interview questions that sort people quickly.
- Tell me about a batch you couldn’t approve. What happened next, and who pushed back?
- Your release target is Friday, and a deviation on the batch is still open. What do you do on Thursday?
- How do you handle a production supervisor who keeps sending records back with the same mistakes?
- Paper, electronic, or both?
That third one matters. A lot. A reviewer who can only find errors, and can’t get the floor to stop making them, will spend their career reviewing the same mistake. Resumes rarely say which records a reviewer actually approved, so I verify that on the phone, using the same approach I laid out for checking GMP experience on a resume.
Move quickly on the good ones. I think of a strong candidate like a house for sale. If it’s a great house, more than one buyer wants it. A reviewer who aced your planted-error exercise is probably acing somebody else’s this week.
Contract, Contract-to-Hire, or Permanent
When people call me asking for “recruitment support” on a review backlog, I ask one question first. Is this a spike or a steady state?
A spike looks like a pre-approval inspection on the calendar, a campaign of batches from a new client, or six months of records nobody reviewed while a seat was open. That’s contract work, and contract reviewers who already know your dosage form can start clearing records after SOP training, without the long ramp a new permanent hire needs. Our pharmaceutical staffing agency page covers how that onboarding works. Steady volume is different. A plant releasing batches every week needs reviewers who know the process and the people on the floor, and that’s a permanent seat, whether you fill it through direct hire staffing or try a contractor first and convert.
Retention matters here. A lot. A reviewer gets faster and sharper every month they spend with your records and your floor, so the hire you keep is worth more than the hire you make. KORE1’s 12-month retention rate is 92%, and on a review desk that number is the whole point.
Batch Record Review Questions From the Quality Side
Can production review its own batch records?
Production can do a first review, but the quality unit has to approve the record before the batch is released. Under 21 CFR 211.192, the final review and approval belong to the quality control unit. ICH Q7 lets qualified production staff review noncritical steps for APIs under procedures the quality unit approves, while critical-step records stay with quality.
Do we need a Qualified Person to release batches?
Only if the batch is being released for sale or supply in the European Union. EU GMP Annex 16 requires each finished batch to be certified by a QP within the EU, and every EU manufacturing site needs at least one. For U.S. distribution, release runs through your quality control unit under 21 CFR Part 211.
Batch record reviewer, QA specialist, QA associate. Same job?
Usually they’re the same pharma QA job under different titles, so read the duties rather than the title. Companies name the person who reviews executed batch records however they like. Watch for one mix-up in particular. “QA engineer” usually means software testing outside pharma, and those candidates won’t know a batch record from a build log.
Can a contract reviewer approve the release?
That depends on your procedures, not on the employment type. Release authority belongs to your quality unit, and your written procedures say who inside it may exercise that authority. Many companies keep the final release signature with an employee and use contract reviewers for the detailed review underneath it, which is a sensible split.
How long should a batch record review take?
No regulation sets a number, so the honest answer is that it depends on the length of the record, the number of deviations attached, and whether it’s paper or electronic. Track your own review time and queue length. When the queue grows for several weeks running, you have a staffing problem, not a diligence problem.
Before the Warehouse Fills Up
A batch isn’t finished when the line stops. Not really. It’s finished when somebody qualified has read the record and signed it, and that person deserves the same hiring attention as anyone on the floor. If you’re bringing in a client program, my piece on tech transfer hiring covers the people who build the batch record in the first place.
Got batches waiting on review, or a pharma QA seat that’s been open too long? Let’s Connect. Tell me how many records are in the queue and how your review is set up. I’ll tell you which seat to fill first, and whether you need a hire at all. I’m reachable on LinkedIn.
