Your regular salt supplier is sold out. You have a canning run scheduled, the product needs that salt, and you have maybe a day to solve it. So you do what any owner does: you find another source. You get it. It's actually a little cheaper. You feel like you dodged something.
You did dodge the stockout. But if the new bag went into a batch before it went into your records, you also just created a lot in your finished product that you can't fully account for. And you won't find out until the day someone asks you to.
This is the quiet version of a traceability gap. Nothing broke. No batch failed. The product is fine. The hole is in the paper, and the paper is the only thing that answers for you when a customer, a buyer, or an inspector asks a pointed question six weeks from now.
Why the substitute is the dangerous one
Your normal supplier is easy. You've bought from them a hundred times. Their lot codes look familiar, their Certificate of Analysis lands in the same inbox, the entry is muscle memory. The record more or less writes itself because you've done it so many times you can't skip it.
The substitute is different, and it's different in exactly the way that matters:
- It shows up under pressure. You went looking for it because something was already going wrong. Pressure is when steps get skipped.
- It's unfamiliar. New supplier, new lot format, new CoA layout, maybe a different pack size. Nothing about it is automatic yet.
- It's temporary in your head. You're telling yourself this is a one-time thing, so it doesn't feel like it deserves the same treatment as a "real" supplier. But that one-time bag is going into a real batch that ships to real customers.
So the exact ingredient most likely to skip your recordkeeping is the one bought in a hurry, from someone new, that you don't expect to buy from again. That's not a coincidence. That's the shape of the problem.
What actually protects you: entry before use, every time
Here's the version that goes right, and it's not complicated.
We ran into this ourselves at Off The Deck. Canning salt for one of our products: regular supplier sold out, we scrambled, found it somewhere else, and it happened to be cheaper. The kind of small win that makes you feel clever for an afternoon.
The reason it stayed a small win instead of turning into a future headache is boring: we've ritualized supplier data entry. New supplier, new lot, the Certificate of Analysis. All of it gets entered before the ingredient goes into anything. Not at the end of the day. Not "I'll catch up on it this weekend." Before it touches a batch.
That's the whole discipline. It sounds almost too simple to matter, but the entire difference between a clean trace and a bad afternoon lives in the word before.
Because once that new salt lot exists in your system, tied to its supplier, its CoA, and its receiving date, every batch that uses it inherits the link automatically. You don't have to remember anything. The batch record knows. The trace is built at the moment of use, not reconstructed under stress later.
Skip the "before," and you're left doing archaeology: digging through order emails, matching delivery dates to production dates, trying to remember which bag you opened on which day. And you're doing that archaeology on the exact day you can least afford to, when someone's already asking.
The test that tells you if you're actually covered
Forget the theory. Here's the only question that matters, and you can ask it of yourself right now:
If someone asked you today which finished batches used your substitute supplier's lot, not your regular one but the backup you bought in a hurry, how fast could you answer, and would you trust the answer?
Before we tightened this up, honest answer: it would've taken a while, and there'd have been a nagging worry about whether it was accurate. Now it's a minute or two, and the worry is gone. That's the entire gap between "I think so" and "here it is."
The founder who can answer that question in two minutes with confidence is in a completely different position than the one who needs an afternoon and still isn't sure. Same product. Same salt. The only difference is whether the record got made at the right moment.
A spreadsheet can hold this. It just won't defend you.
You can absolutely track substitute suppliers in a spreadsheet. Plenty of people do. The problem isn't capacity. A spreadsheet will hold every lot you type into it.
The problem is that a spreadsheet only knows what got typed, when someone remembered to type it. There's no link between the salt lot and the batch that used it unless a human built that link by hand, correctly, every time, including the frantic day the substitute came in. A spreadsheet doesn't catch the batch you forgot to tag. It doesn't flag the lot with no CoA attached. It sits there looking complete, and you don't discover the hole until you go looking for something that isn't there.
That's the founder's version of the spreadsheets-are-a-liability problem: it's not that the spreadsheet is wrong, it's that it can't prove it's right. And "prove it's right" is the entire job when the question comes.
The takeaway
Second-sourcing an ingredient is normal. Doing it to save money is smart. Doing it because you got caught short is just Tuesday. None of that is the risk.
The risk is letting the substitute skip the step your regular supplier never skips, because it's new, because you're busy, because you told yourself it was temporary. Enter it before it's used. Tie it to its lot and its CoA at the door. Then the reorder that saved you money stays a win, instead of turning into the one lot you can't account for on the day it counts.
Know what's in every batch. Even the batch you scrambled for.
See how FourFoxes ties every supplier lot to every batch, automatically →