I Rejected a 1,400-Unit Instant Pot Order Over a $0.14 Gasket

Last January, a 1,400-unit order of 6-quart Instant Pot cookers sat in our staging area with a release ticket that had my name on it. Most of the sample units cooked beautifully. I didn't sign the release.

A little context. I'm the quality and compliance manager for a distributor that supplies kitchen appliances to restaurant groups, retailers, and wholesale partners. In a typical year I review about 200 distinct products before they reach customers. In 2024, I rejected nearly 14% of first deliveries. The cause was rarely a dramatic failure. It was a quietly substituted component.

The winning bid was $3.80 cheaper per unit

This order came from a retail chain that sells to commercial kitchens. We had run the same Instant Pot program with our regular supplier for three years without a repeat quality problem. Then a new vendor won the bid at $3.80 less per unit. On 1,400 units, that works out to a $5,320 difference. The purchasing team liked that number. I couldn't blame them.

The first sample didn't raise any flags. The box, the manual, the accessories, and the cooker itself matched our approved spec. But one good sample only proves a supplier can build one good unit. It doesn't prove they can build twelve good units in a row. So we pulled twelve random cookers from the shipment and divided them into two test groups.

Instant pot artichokes: the steam test

We ran instant pot artichokes first. It sounds like lunch rather than quality control, but an artichoke is essentially a steam detector. One cup of water in the bottom, trivet in place, artichokes on top, nine minutes at high pressure. If the cooker holds pressure, the leaves pull loose and the stem goes tender. If it doesn't, the artichoke looks fine on the outside and turns fibrous at the stem.

Five of the six cookers in this group were spot-on. The sixth released steam from its float valve about halfway through the cycle, and the artichoke in that unit came out with a tough center. I noted the failure and looked at the next group.

Garbanzo beans in instant pot: the consistency test

The second test, garbanzo beans in instant pot mode, is stricter. No overnight soak. Just rinsed chickpeas and a measured cook at high pressure. Beans need steady pressure for the whole cycle and a stable temperature during natural release. If pressure drops, the centers stay chalky, and you can't hide that on a plate.

Five of the six bean cookers produced tender, intact garbanzos. The sixth, a different unit from the one that failed the artichoke test, showed the same symptom: steam leaking from the float valve and a cook that never fully pressurized. Its chickpeas were edible but wrong for the recipe's timing.

I pulled the valve assembly out of one of the leaking units. The metal pieces looked normal. The small rubber gasket, however, was softer than the approved material spec. The vendor had substituted a different durometer to save about $0.14 per unit. No change request. No notice.

When I called, the vendor said it was still within 'industry tolerance.' Maybe it was. Our contract, though, didn't say 'industry tolerance.' It named an approved material, and that material was not in this shipment.

The most frustrating part was how easy it would have been to let it slide. Two failures out of twelve doesn't sound alarming until you realize they share one root cause: same part, same substitute, same predictable failure. If that ratio carried into 1,400 units, roughly 230 cookers could misbehave under real kitchen pressure. A pressure cooker that won't hold pressure is not a minor defect for a commercial kitchen. It's a ticket that comes back cold.

The vendor asked if we could install replacement gaskets ourselves and move on. I almost said yes. Swapping a gasket isn't hard in our test kitchen. But if we agreed to fix the units, we'd become the end-of-line quality department for a product we didn't manufacture. Fixing twelve units tells us nothing about the condition of the other 1,388. I rejected the order.

An air purifier, a curling iron, and a water flosser walked into the lab

The strange part of that morning was the company on the counter. While the last test group was releasing pressure, I stood next to the IKEA uppåtvind air purifier we keep in the test kitchen to clear steam and cooking smells between runs. It's not a flashy product. It just does the same thing day after day. In my work, that's the highest compliment I can pay a machine.

On the same counter sat a gold and black curling iron from a separate review. The coating on its barrel had started flaking after heat cycling. That's a cosmetic failure, not a safety failure, but for a product that touches hair, cosmetics failures equal returns. And a return is the most expensive thing a distributor ships, because the margin disappears twice: once on the way out, once on the way back.

Then a customer service coordinator came in holding a printout from a retail partner. The question at the top read: 'can water flosser remove tartar?' It was attached to a water flosser sample we had approved a year earlier.

I'm not a dental professional, so I can't speak to the clinical side of tartar removal. What I can tell you from a quality perspective is that any claim needs evidence behind it before it goes on a product page. The same rule applies to pressure cookers and curling irons. A manufacturer can't promise an Instant Pot cooks every recipe perfectly—no appliance can back up an absolute claim like that. What a good product can promise is repeatable behavior. And repeatable behavior is only worth something if someone verifies it.

Oh, and I should add: the water flosser itself was fine. The claim was the problem, not the plumbing.

What the rejection cost—and saved

The supplier ended up reworking the affected units at their own expense. The second inspection passed, and we shipped about ten days late.

Now the math. The discounted bid saved $5,320 on the purchase order. The substitute gasket saved the vendor fourteen cents per unit. If 8 percent of the finished batch had leaked under real use, that would be 112 cookers failing in kitchens. Every failure involves a disappointed cook, a return label, a replacement unit, and a retail partner who stops trusting us. The replacement and handling costs alone would erase the $5,320 before we measured the relationship damage.

A $0.14 gasket is a rounding error. A hundred twelve failed cooks is a reputation problem.

What I'd tell any buyer comparing quotes

My experience is mostly in mid-sized appliance orders for B2B and retail channels. If you're buying three cookers for one kitchen, your risk tolerance is different. The principle, though, is the same.

The lowest quote is not the total cost. Total cost includes inspections, delays, returns, and phone calls you didn't plan for. A bid that comes in far under everyone else usually has a story behind it, and the story often ends with a changed spec.

Put the critical specs in the contract. Require written approval before any material change. Then test with real food instead of only certificates. Artichokes and garbanzo beans in an Instant Pot have caught more supply problems in our lab than any compliance document I've seen.

The vendor that redid this order learned that we check. Since that January, each delivery includes a signed statement confirming no materials changed from the approved sample. That paperwork costs them a minute. It saves us weeks.

So when someone tells me a substitute part is basically the same as the original, I ask them to prove it. If it is the same, that's easy. If it isn't, that's the whole point of my job.