Who this checklist is for

I'm the office administrator who handles purchasing for a food company of about 40 people. Most weeks that means facilities and office supplies. But when we launched our ready-to-eat line in 2024, packaging landed on my desk too — roughly $150,000 a year now, spread across six vendors, and I report to both operations and finance on all of it.

I'm not a packaging engineer. But I did run the full retort pouch sourcing evaluation that year, and before that I lived through one first order that went wrong in a way that was entirely avoidable. So this is the checklist I wish I'd had. Seven steps. Run them in order before you place an OEM order, or use them as a re-audit if you already have a supplier.

Step 1: Lock the product parameters before you talk to anyone

Every retort pouch manufacturer asks the same five questions in the first email. If you can't answer them, you get a generic quote for a generic pouch, and it won't be comparable to anything else you receive.

  • What's going inside? Acid, fat, moisture content, bone fragments, sauces, solids-in-liquid — each one changes the required film.
  • What retort profile? 121°C for 30 minutes and 135°C for a short flash are not the same requirement. Neither is a pasteurised product that never sees a retort at all.
  • What shelf life are you designing for? Tell the supplier the target. Don't ask them to promise it.
  • How is it filled? Hot fill, cold fill, hand fill, form-fill-seal — the seal zone and headspace differ.
  • What's the packed weight? A 200g pouch and a 1kg pouch pull on the seal very differently during the retort cycle.

Write it all onto one page and send the same sheet to every supplier. That single habit did more for quote comparability than anything else I did that year.

Step 2: Specify the barrier structure in writing

This is where a lot of aluminium retort pouch OEM quotes go sideways. You ask for aluminium foil and you get back... a sentence that says aluminium foil. That's not a specification.

Ask for the full structure, layer by layer, with thickness: outer polyester or nylon, the foil layer, the adhesive system, the sealant layer. For retort work the sealant usually needs to be a retort-grade cast polypropylene (RCPP), not the standard CPP used for ambient pouches. Foil thickness matters too — 9µm and 7µm behave differently at 135°C.

Get the structure in writing on the quote itself. If a supplier won't put layer order and micron thickness in writing, that tells you something about how the rest of the order will go.

Also ask whether the foil is laminated in-house or bought in. It changes their lead time and their ability to fix a problem mid-run.

Step 3: Ask for named compliance documents, not the word “certified”

This is the step most buyers skip, and it's the one that delayed us two full weeks in Q3 2024.

“We are FDA approved” means almost nothing on its own. Ask for specific documents:

  • FDA 21 CFR 177.1390 covers laminated structures for high-temperature applications. Ask whether the specific structure they're quoting is covered, and get the supporting documentation from the film supplier, not just the pouch maker.
  • EU Regulation (EU) No 10/2011 if you sell into Europe. Retort conditions require migration testing at the actual time-and-temperature profile, not at ambient. Ask for the test conditions, not just the pass/fail line.
  • Migration test reports with dates. Check the report date. I've been handed reports in 2024 that were issued in 2019 against a different formulation.
  • The chain of documents. Resin supplier to film converter to pouch maker. If the pouch maker can't trace back to the film converter, the compliance claim stops at their door.

Verify current requirements yourself at fda.gov and at the EU Commission's food contact materials page before relying on any supplier summary. Regulations move, and summaries age badly.

Step 4: Force every quote onto the same inputs

A retort pouch quote is not one number. It's a set of inputs that have to be identical across suppliers before you compare anything at all.

Send the same sheet, with checkboxes:

  • Pouch format and dimensions (flat, stand-up, three-side seal)
  • Roll stock or pre-made pouches
  • Print method and number of colours
  • Plate or cylinder costs, and who owns them afterwards
  • MOQ, and MOQ per print run
  • Sample lead time and production lead time
  • Incoterms and who pays freight
  • Payment terms

I learned never to assume a MOQ is per design after a quote came back at a number that turned out to be per colour, per run. That one cost me a week of back-and-forth to untangle (and a fair amount of goodwill with our finance team).

Step 5: Test samples with your actual product, not water

Water tests tell you almost nothing about a retort pouch that will hold a tomato-based sauce with oil in it. Every serious retort pouch manufacturer will send samples — the question is whether you test them properly.

Run at minimum:

  • A full retort cycle at your target profile
  • Seal strength testing (ASTM F88 is the standard method for flexible barrier seal strength)
  • A drop test on the packed product
  • A dye penetration or vacuum leak check on a batch, not a handful
  • Three weeks of storage observation before you sign off on the full print run

The causation runs opposite to what people assume here. Buyers think expensive suppliers must test better. Actually, suppliers who test properly have to charge more for it — the testing cost comes first and the price follows. That is not the same thing as price proving quality, and treating it that way will push you toward the wrong vendor as often as the right one.

Step 6: Cost the whole order, not the unit price

This is where the real money hides. Unit price is the easiest number to compare and the least useful one.

Here's how I cost it now. In 2024 we compared two suppliers for a 500g stand-up pouch. Supplier A was about $0.008 per pouch cheaper at the quoted volume. On 200,000 pouches that's $1,600 saved. Then I put the rest of the numbers on the same sheet:

  • Supplier A's lead time was three weeks longer, so we'd carry six extra weeks of safety stock
  • Their reject rate in the sample batch was 4%, versus 1.2%
  • Their plates were quoted separately and we'd pay for them again at reorder

Once inventory carrying cost and a realistic reject rate went in, the $1,600 advantage turned into roughly a $2,000 disadvantage. The lowest quote cost more. That isn't a rare outcome — in my experience it's the normal one.

Build the sheet once and reuse it. Mine has unit price, plate cost, freight, MOQ-driven inventory, expected reject rate based on our own testing, and a line for the cost of a production delay. That last line is the one nobody puts in and the one that has actually hurt us.

Step 7: Run incoming QC on the first order and log the batch

First orders are where hidden problems surface, because nobody has run the process long enough for small deviations to compound into something visible.

  • Pull an AQL sample, not the box on top of the pallet.
  • Check seal width and seal appearance on the retort-cooked product, not the flat pouch.
  • Keep retained samples from every batch with the supplier's lot number written on them.
  • Verify the delivered structure against the quoted structure — film thickness checking is cheap, and it's the one thing that quietly gets swapped.
  • Photograph everything. If a dispute starts in month four, you'll want the photos from month one.

Things that catch people out

A few flags from the other side of this process.

The “local is faster” assumption. That was true 15 years ago, before a competent remote supplier could ship samples in a week and share retort data the same day. Today a well-organised overseas retort pouch manufacturer often beats a disorganised domestic one on turnaround — and I say that as someone who badly wanted the domestic option to win.

Treating certification as a checkbox. Certificates have issue dates and scopes. A supplier can be certified for one structure and quoting you another. Ask what the certificate covers, not whether they have one.

Skipping the bilingual spec sheet. If you're sourcing a custom kemasan retort pouch for an Indonesian or wider Southeast Asian market, write the specification in both languages and have both sides confirm it in writing. Two slightly different versions of “matte finish” once caused us a reprint.

Not visiting or video-auditing. If you can't visit, ask for a live video walkthrough of the lamination line and the pouch-making line. It takes about 40 minutes and it tells you more than three rounds of email ever will.

None of this is complicated. It's just that the cheap quote looks cheap right up until the moment it isn't — and by then you're the one explaining the delay to operations.