Every importer has heard it, and most have said it. A known problem shows up in the sample — a seam that puckers, a color that's slightly off, a part that fits a little loose — and rather than run another round, someone says, "It's fine, we'll fix it in production." The program keeps moving, the hard conversation is avoided, and everyone feels productive. Then the problem shows up again at pre-shipment inspection, except now it's on ten thousand units and you own it. "Fix it in production" is one of the most expensive sentences in sourcing, because it almost never happens — and the reasons it doesn't are structural.

The Sample Is the Best the Product Will Ever Be

A sample is made slowly, by the factory's best worker, with the most attention the product will ever receive. Mass production is the opposite: made fast, by many hands, under time and cost pressure, on a line that rewards throughput. Whatever quality you see in the sample is the ceiling. Production can only hold that standard or fall below it — it does not rise above it.

So "we'll fix it in production" is backwards. Production is where quality degrades, not where it improves. A flaw you can already see in the careful, hand-made sample is not a temporary condition that the factory will clean up at scale. It is the floor, and at scale it gets worse.

Your Leverage Disappears the Moment the PO Is Placed

Before you place the order, you hold every card. The business isn't committed, the deposit hasn't moved, and the factory wants the program. That is the exact moment you have the leverage to require a fix — and to make approval contingent on seeing it.

After the PO, that leverage is gone. Your deposit is paid, tooling is cut, materials are ordered, and the factory's incentive shifts from winning your business to protecting its margin on an order it already has. Asking them to invest in fixing a known issue now costs them time and money they didn't price in — so they won't, unless they have to. The time to fix a problem is while you still have the standing to demand it, not after you've handed over your leverage.

"Fix It in Production" Usually Means "Hope It Goes Away"

Notice what's almost never attached to the phrase: an actual plan. There's no revised sample, no corrective action, no explanation of what will be different on the line. If there were a real fix, it would show up in a new sample. Deferring a problem without a documented solution isn't a plan — it's optimism with a deadline.

The logic is unforgiving here. If the problem can be fixed, the factory can fix it in a sample and prove it. If they can't produce a fixed sample, nothing about a production line running thousands of times faster is going to fix it for them. "We'll fix it in production" and "we don't actually know how to fix it" are, most of the time, the same sentence.

The Cost Lands at the Worst Possible Moment

A problem caught in the sample stage costs a sample round — a courier and a week or two. The same problem deferred into production surfaces at pre-shipment inspection or, worse, at the retailer's dock. Now the options are all bad: reject the shipment and miss the window, accept it and eat the returns and chargebacks, or pay to rework the goods at destination. Every one of those costs more than the sample round would have — often by an order of magnitude — and arrives exactly when you have the least time to absorb it.

When Deferring Is Legitimate — and How to Do It

This isn't an argument that nothing can ever be resolved during production. Sometimes it genuinely can — a proven tooling adjustment, a machine setting, a change the factory has demonstrated before. The difference between a legitimate deferral and a costly gamble is entirely in the discipline around it. Never defer on a promise. Defer only with:

  • A written corrective action — specifically what will change and why it solves the problem.
  • A verification step — a pre-production or first-article sample that proves the fix before the full run, not after.
  • A checkpoint before mass production — an agreed gate where production doesn't proceed until the fix is confirmed against your golden sample.

That turns "we'll fix it in production" into "here is the documented fix, and here is the checkpoint that proves it." One is a hope. The other is a plan.

The One Question That Ends the Debate

When you hear the phrase, ask a single question: "If this can be fixed, show me the fixed sample. If you can't show me a fixed sample, what makes production different?" It reframes the conversation instantly. Either a corrected sample appears — and you've solved the problem before committing — or it doesn't, and you've just learned the problem is real, unsolved, and about to be multiplied by your entire order quantity.

The sample stage is where you have the leverage, the time, and the attention to get a product right. Production is where problems multiply and options run out. Don't move a known defect downstream and hope it improves. Fix it while you still can — or decide, with clear eyes, not to proceed.

When did "we'll fix it in production" actually work out for you — and what was different about that time that made it more than a hope? We'd like to hear where the line really is.