The Manufacturer Transfer Playbook: Benchmark, Pilot, Cutover

Reviewed by Apollo’s production and quality team · Last reviewed July 2026

The real risk in switching manufacturers isn’t picking the wrong shop. It’s going dark in the middle: the old line stops before the new one is proven, and you stock out on a product that was selling. That gap is avoidable. You avoid it by refusing to cut over on faith. Prove the new shop on a sample of your actual product, review its specs and components, run a pilot, and keep your current manufacturer shipping until that pilot passes. Move in that order and a transfer becomes a controlled handoff, not a bet on your supply.

Educational overview: not legal, regulatory, or medical advice. Requirements change and vary by jurisdiction and sales channel. Last reviewed July 2026.

Short answer: The danger in switching manufacturers is a supply gap mid-transfer, not the choice of shop. You close the gap by moving review-first: prove the new shop on a benchmark sample, review specs and components, run a pilot, then cut over with overlap inventory so the old line keeps shipping until the new one is proven. Benchmark to pilot typically runs six to eight weeks.

Best for: Existing brands with a selling product whose current manufacturer has gone slow, opaque, MOQ-rigid, or quality-risky.

Key decision: Whether to run a review-first transfer and keep the incumbent warm, instead of a hard cutover on a promise.

Apollo path: Start a Transfer Review → /quote

Apollo runs transfers as a specialty, out of an FDA-registered, cGMP-compliant facility in Livermore, California, with manufacturing experience behind the lines. Most of what follows is how we run a transfer on the floor. It works the same whoever you move to.

Signs it’s time to move, and signs it’s just a bad week

Not every rough patch is a reason to switch. A missed date or a single off-spec lot is a conversation, not a divorce. Ask for the corrective action, and watch whether it holds.

Structural problems are different. A quality system with no real specifications. Batch-to-batch drift you can measure. Price creep with no cost driver behind it. A scaling ceiling you have already hit. Those don’t get better with another phone call, because they are how the shop is built.

If you are still diagnosing whether this is a hiccup or a real failure, that is a different read, and we wrote it up separately: signs a manufacturer is actually failing. This piece assumes you have decided to move. The rest of it is about doing that without a gap.

The review-first transfer: four gates that protect your supply

A transfer is not one decision. It is four, in order, and each one is a gate the new shop has to clear before the next opens. The order is the whole point: nothing about your live supply changes until the shop has earned the change.

Everything below is those four gates in detail, plus the two things brands most often forget: stability testing, and who actually owns the formula.

Gate 1 (the benchmark sample): prove they can match what you sell

Before any talk of volume or price, a capable shop makes a sample. Not a formulation from scratch, a benchmark: a small run matched to a unit you sell right now. You send a current retail unit and its documentation; the shop reproduces it and hands it back for a side-by-side.

It answers the only question that matters at this stage, and it does so early, before you commit to volume: can they make your product, on their equipment, and have it come out like yours? A benchmark sample surfaces the gaps while the fix is still a bench adjustment, not a recalled batch.

Watch what the shop asks for. One that requests your specifications, your ingredient list, and a real unit to work from is running a quality system. One that says “send it over, we’ll match it” without asking for any of that is guessing, and you will find out where the guess was wrong on your first production run.

Gate 2 (spec and component review): where quality actually drifts

Two shops can make the same formula and ship two different products. The formula is only half of it. The other half is how it gets made, and that is set by equipment, method, and materials that differ from one floor to the next. Gate 2 is where you map those differences before a customer does.

Here is what drifts between shops, why it drifts, and what you hold the new shop to:

The reason this gate matters more than any other is written into FDA’s own inspection record. FDA inspection data has repeatedly shown that a specifications failure ranks among the most-cited problems in supplement manufacturing: the shop either never established product specifications for identity, purity, strength, and composition, or never verified that a finished batch met them (a pattern that trade summaries of FDA enforcement data, such as Eurofins’, have tracked for years). The regulation itself, FDA’s 21 CFR 111.70 and 111.75, requires those specifications and requires you to test to confirm them.

So the spec review is not paperwork. A shop that asks for your specifications and tests to them on the first batch is showing you its quality system works. A shop that has none, or tests a single ingredient in a multi-ingredient product and calls it done, is the citation pattern waiting to happen. You want to find that out at Gate 2, not from a customer complaint.

One more thing gets settled here: components. If the new shop sources custom components for you, bottles, closures, droppers, custom labels, the same component-minimum math applies as on any build. Minimums often land in the low thousands, and any custom excess is yours. It is worth reading before you commit, because it changes what a “small” transfer batch actually costs to stand up: the component-minimum math.

Gate 3: the pilot run, and the six-to-eight-week window it takes to get there

A pilot is a real production run, sized small. Often in the hundreds of units when compatible materials are on hand, run on the equipment your reorders will run on. It is not a lab-bench mockup. What passes the pilot is what you will actually ship, which is the entire point of running one.

From an approved benchmark sample to a pilot on the line typically takes six to eight weeks. That window is not the shop being slow. It is the work that protects you: sourcing or qualifying components, sizing the batch to the line, scheduling production, and building the batch record the run is made against. Some of it runs in parallel; component lead times are the usual long pole. Think in weeks, not days. And if a formula or packaging change pulls in stability work, think in weeks-to-months for the full cutover, even though the pilot itself lands in that six-to-eight-week window.

When the pilot run finishes, you get a batch made to your specs and a Certificate of Analysis from a vetted third-party lab confirming it. That document is the receipt that Gate 2 held. It is also the thing you hold up before you let anyone touch your live supply.

A transfer review comes back in one to two business days. If we read your product and it is not a fit for our lines, you will hear that in days, not weeks. A fast no is worth more to you than a slow maybe, because it sends you looking elsewhere while your incumbent is still shipping.

Gate 4 (cutover with safety stock): run the overlap so the clock stays yours

This is the gate brands skip, and it is the one that causes the stockout. The temptation, once the pilot passes, is to give notice and flip. Don’t. Cut over on a schedule, with inventory covering the seam.

The mechanic is simple: keep your incumbent under contract and shipping while you build a bridge of finished goods, then switch production to the new shop while that bridge covers the gap. The incumbent stays warm until the new line has shipped a clean run at volume. You are not paying for loyalty. You are paying for a fallback that is already running.

How much cover to hold before you give notice. You are covering the full transition, not a single reorder. Add it up:

Weeks of finished-goods cover to hold = the pilot window + any stability bridge + your buffer. Hold that much before notice goes out, and the transition timeline can slip without becoming a stockout.

Before you give notice, the checklist:

• Pilot batch passed spec, with a third-party CoA in hand
• Enough finished-goods cover to bridge the full window plus your buffer
• Incumbent still under contract and shipping (notice not yet given)
• New shop’s FDA facility registration and quality documentation verified
• Product-liability coverage confirmed on the new shop for the overlap period
• Contract terms reviewed with your counsel: notice period, minimum-purchase commitments, tooling and inventory ownership, and who holds your specs and artwork
• Lot and batch traceability in place so a customer complaint is answerable on either shop’s inventory

What if the new shop fails mid-transition? Then you have lost time, not supply, because you never cut over. The incumbent is still shipping. That is the entire reason the bridge exists: the new shop has to earn the switch, and until it does, your live supply does not move. A failed pilot is a bad week, not a lost quarter.

Your customer should not feel the switch. Keep the label and the experience the same unless you have a reason to change them. Hold lot traceability so any complaint is answerable to a specific batch. And time the cutover so a retail or Amazon reorder is not the batch you are still proving. Retention is a supply problem before it is a marketing one; a product that is out of stock churns customers whether or not the new formula is better.

On the contract and liability questions, draw a clean line. We support the operational side of a transfer, benchmark, review, pilot, cutover. Your counsel decides the contract-exit and liability questions. Read your incumbent agreement for notice periods and minimum commitments before you plan the switch, not after. This is the contract-manufacturing work we run review-first precisely so the operational side is predictable while your counsel handles the legal side. When it is time, you can Start a Transfer Review and we will read your product against our lines.

Stability testing: the trigger most brands forget, and how to bridge it

Stability is the quiet timeline driver, and it is the one that surprises people. Here is the mechanism. FDA’s cGMP rule for supplements (21 CFR 111) does not force an expiration date onto your label. But if you print one, you need data that supports it, and FDA treats that as an expectation, not a nicety. Your current data was built on your current formula, in your current packaging, made your current way. Change any of those three and the data may no longer describe what is in the bottle.

So the real question in a transfer is: does the change re-open the stability question?

The industry runs this to a known template, the ICH Q1A(R2) stability guideline: accelerated conditions (40°C / 75% relative humidity) for an early read, and long-term conditions (25°C / 60% relative humidity) to confirm the real shelf life. Accelerated data gives you a signal in months; the long-term study confirms the dating over a longer window.

The move is not to wait idle for that data. It is to run the retest against overlap inventory. Your existing stock keeps the channel supplied while the new shop’s product earns its dating. That is the difference between a stability retest and a stockout: one is a line item in the transfer plan, the other is what happens when nobody planned for it.

Who owns your formula? The private-label transfer trap

Here is the trap that catches brands built on a private-label or stock formula: you may not own the recipe.

In many private-label relationships, the formula and its specifications belong to the manufacturer, not the brand on the label. You own your brand. You usually own your artwork. You may not have the right to hand the exact formula to a new shop, because it was never yours to move.

What that means for a transfer:

• Your brand, your customers, your channel. Yours. They move with you, no question.
• Your label and artwork. Usually yours, if your agreement says so. Confirm it in writing.
• The formula. Often the manufacturer’s. You may need to rebuild it, either by benchmark-matching the product on the new line, or by developing your own version through custom formulation. Two to three bench rounds is typical to dial in a match you own outright.

Confirm ownership before you plan the switch, not during it. This is a counsel question, and it is the one place a transfer can stall hardest if it goes unasked. We support the transfer; your counsel decides what you own. If the answer is that you do not own the formula, the review-first path still works, you just start the new relationship at Gate 1 with a benchmark and, where needed, a short formulation effort to make the recipe yours.

Bringing it together

Done right, a transfer is boring: four gates in order, with overlap inventory the whole way through so your incumbent keeps shipping until the new line has earned the switch, and none of it touches your live supply. It’s what we built the existing-brands path around, backed by the documentation a real quality system produces. When you are ready, Start a Transfer Review: send a current unit and your specs, and you’ll hear back in one to two business days.