How to qualify a co-manufacturer before you talk about price
Co-manufacturing is how most brands get a new product onto shelf without buying a plant. It turns a capital decision into an operating one, and it gives you access to equipment and process experience that would take years and several million dollars to build in-house. The market that reflects that: U.S food and beverage co-manufacturing passed $100 billion in 2023 and is estimated at roughly $120 billion in 2025, expanding at two to four times the rate of the broader food industry.
The harder part is landing with the right partner, and in some categories that has become genuinely difficult. Protein and snack bars are one of them, along with high-value liquids that need a long shelf-life and anything that requires fluid processing.
Much of that capacity has moved into fewer hands. Private equity has spent the past few years buying independent co-packers, usually founder-owned businesses, which means the plants still exist but no longer set their own priorities. Larger contracts get scheduled first, and the number of brands competing for whatever remains has grown over the same period. Value Gene Consulting describes what follows: asset-light brands get placed at the back of the queue, or find no capacity available for expansion at all.
Scarcity of that kind raises the value of choosing well. When qualified lines are harder to come by, the gap between a plant that fits your product and one that simply takes your call gets expensive quickly. So if your product sits in one of those categories, what decides the outcome is access to the right line, and the rate you negotiate on it only matters once that is settled.
This article covers the questions that surface access and technical fit, and the order that gets you to a workable partner faster.
Why the first conversation usually starts in the wrong place
Most first calls open with price, and there is a reasonable explanation for that: price is the only number that looks comparable across suppliers, so two quotes side by side feel like information.
They are not information yet, because a quote assumes two things that aren’t verified at that stage: that the plant can actually run your product, and that it has line time available when you need it. Until both are confirmed, the comparison has nothing underneath it.
That matters because the cost of leaving them unconfirmed climbs as the process continues. Many co-manufacturers require production confirmation at least a month ahead of a run, so a dealbreaker that surfaces on the third call has already cost you a plant visit, a sample run, and a slot in a schedule you now have to give up.
Advice on what to ask instead is easy to find, and most of it is written by co-manufacturers, which is a reasonable thing for a plant to publish. We do not operate one — we screen them on behalf of brands, and that is why this list includes the answers that should end a conversation rather than the ones that make a facility look prepared.
Some questions eliminate a plant, others open a negotiation
Qualifying a co-manufacturer usually takes two or three calls before you know whether it is a real fit. What changes between those calls is the cost of the information rather than the information itself.
A dealbreaker on the first call costs an hour. The same dealbreaker on the third call costs a plant visit, a sample run, and four to six weeks of calendar you do not get back — the fact was true the entire time, and only the price of discovering it moved.
That gives you a way to sort the questions, because two of the four groups below eliminate a plant on their own while the other two almost always leave a negotiated path open.
Consider what that difference looks like in practice: a facility that cannot run your format will not acquire the ability inside your timeline, since filtration capability, carbonation under protein load and small-batch pilot capacity all come from equipment and accumulated experience, and no commercial conversation adds them to a plant.
A minimum order quantity above your first order behaves differently, because a plant might phase it across quarters, extend it over a longer term, or solve it by scheduling you alongside another customer. An equipment gap closes the conversation, whereas a volume term carries a price, and prices tend to move.
Ruling a plant out in the first hour returns that time to the candidates that could actually work, which is the whole point of asking in this order. Most brands are choosing among a handful of options, and the faster the wrong ones drop off, the more attention the real contenders get.
That is why availability and technical fit belong in the opening call, while they are still cheap to ask about, and why certification and commercial terms can wait for the second, once you already know the plant can do the work. The four groups follow that order.
Group 1: real availability
This group measures whether stated capacity exists as line time you can actually book, and it comes first because nothing in a negotiation resolves its absence.
"What production windows do you have open in the next two quarters, and which of those could you commit to in writing this month?"
The second half of that question is what separates real capacity from aspirational capacity, since most plants will happily describe what they could theoretically accommodate while far fewer will put a specific window in writing against a specific date.
There is a second question here that co-packer guides tend to skip. Ask whether any of the SKUs running on that line are the co-manufacturer's own house brands in your category, because if they are, your formulation and your shelf performance data will end up inside a competitor's building. Some brands accept that in exchange for the capacity and it can be the right call, but it works better as a decision you made than as something you found out afterwards.
Group 2: technical fit
This group measures whether the plant can make your product rather than a product in your category, and it also eliminates on its own, which is why it sits second.
"Have you run a product with this formulation profile before, and what went wrong the first time?"
Every first commercial run has a problem, so being able to describe that problem specifically, such as the viscosity drifted, the fill temperature had to change, the shelf-life study came back short, is what tells you the experience is real. An answer with no problems in it usually means the run has not happened yet.
Group 3: quality and compliance
This group measures whether compliance is maintained or was achieved once, and it sits third because it has a negotiated path: audit timing and certification scope can be worked through, at a cost and on a timeline.
"Which certifications are current — SQF, BRCGS, FSSC 22000, IFS Food — when is the next audit, and can I see the last report?"
The request for the report is the real test. SQF functions as an entry requirement for national retail in practice, since major chains will not onboard product made in an uncertified facility, so ask for the score on the most recent audit rather than only for confirmation that the certificate exists.
Group 4: commercial terms
This group measures your real cost of entry, and it goes last because it is negotiable end to end.
"What is the real MOQ per packaging configuration, not per product?"
A brand with three pack sizes can face three separate minimums, because in beverage the minimums vary more by format than by facility. Aluminum cans commonly start around 2,000 to 5,000 cases per SKU, glass and PET closer to 1,000 to 3,000 cases, and a large industrial can line can require something near a full production run. Those are market ranges rather than published figures, so treat them as order of magnitude and confirm them against the specific plant.
What a red flag actually sounds like
Each of the twelve questions has a specific answer that should end the conversation, and each one is a sentence you can write down in advance and listen for on the call. Two examples:
- Ask how the plant decides when two customers want the same production window, and the answer that should stop you is "that never happens." Scheduling conflicts occur in any facility running at healthy utilization, so denying them points to either distance from the floor or idle capacity that other customers have already declined.
- Ask about the most recent deviation or recall and what changed afterward, and the answer that should stop you is "we have never had one." Every functioning quality system records deviations, which means their absence from the record tells you the system does not capture them.
The pattern holds across the other ten. In each case the red flag is a specific phrase rather than a general category of concern, and that is what makes it usable while you are still on the call.
Which answers end the conversation, and which open a term
Of the twelve answers, five should end the conversation and five open a term you can work on.
Knowing which is which before the call changes what you do with a bad answer, because with the distinction defined in advance, you either walk or you counter. Without it, the tendency is to counter on everything, including the constraints that have no solution, which is how brands end up spending a month negotiating with a plant that was never going to work out.
Say a plant will not put any production window in writing. No counter-offer fixes that, because the thing you would be negotiating over does not exist, so you leave and give the time to a plant that will commit.
A changeover cost you consider too high works differently. It has a number attached, and numbers move — the rate, the volume it applies to, and who absorbs the time are all things a plant can adjust.
What to do this week
Start by listing the three products most likely to launch in your next eighteen months, then check whether your current or prospective partner can run all three, at what batch size, and with certification scope that covers each one. If you cannot answer that today, it is an information gap now and a launch delay later.
A plant that can run those products almost always exists somewhere. The work is identifying which one, and confirming it while you still have room in the calendar rather than after a retail date is set.
We wrote down all twelve questions in the order above, with what each answer tells you, the red flag for each one, and the table separating the five elimination criteria from the five negotiable terms.