It has a name like Master Tracker v4 FINAL (use this one).xlsx.
It has fourteen tabs. Four of them are used. Two of them are used but nobody remembers what they do. One is called DO NOT DELETE and nobody has tested whether that is true.
It was built in a weekend by whoever was least busy, to solve one problem — probably inventory, probably during a stockout — and then it absorbed everything else, one column at a time, over three years.
It is now the operating system of your company. And it is a file.
How it happens
Nobody decides to run a business on a spreadsheet. The spreadsheet accumulates.
The sequence is remarkably consistent across brands:
Month 1. You need to know how much of each SKU you have. You make a sheet. It has four columns. It is perfect.
Month 4. You add a tab for raw materials, because finished goods stock is meaningless without knowing whether you can make more.
Month 9. A supplier changes a spec. You add a column for "current supplier" and a column for "notes."
Month 14. You start co-packing. You add a tab for production runs, with lot codes, because the co-packer sends you a sheet and you need somewhere to put it.
Month 20. A retailer asks for a CoA. You add a column with a link to the Drive folder. Sometimes.
Month 26. You hire an ops person. You spend nine days explaining the sheet. They ask why the reorder point on tab 3 doesn't match tab 7. You do not have a good answer.
Month 31. Someone sorts a range without expanding the selection.
That last one is the event people remember. It is not the real problem.
The real problem is not that it breaks
Spreadsheets are excellent. They are the most successful piece of business software ever built, they are infinitely flexible, and telling a founder to stop using them is usually bad advice.
The problem is narrower and more specific than "spreadsheets are bad."
A spreadsheet stores values. Your operation runs on relationships.
That is the whole thing. Everything else follows from it.
When your co-packer switches you from Supplier A's ascorbic acid to Supplier B's, that is not a value change. It is a change that propagates:
- The bill of materials changes
- The batch record template changes
- The CoA on file is now the wrong CoA
- The allergen statement may change
- The label copy may change
- The HACCP plan's hazard analysis for that ingredient may change
- Every finished lot made after the switch is linked to different documentation
- The retailer who has your spec sheet on file now has an outdated spec sheet
A spreadsheet cannot know any of that. You change one cell. Nine other things are now silently wrong, and nothing anywhere tells you.
They stay silently wrong until someone external asks a question that surfaces the inconsistency. Usually an auditor. Usually at the worst possible time.
The three failures, in order of expense
Failure one: the reconciliation tax.
Every week, someone spends hours making numbers agree. Shopify says one thing, the 3PL says another, the sheet says a third. Nobody is lying; the three systems just count different things at different moments.
This is a real cost and it is the one founders notice, because it shows up as a person's time. It is also the cheapest of the three failures, which is why fixating on it leads to the wrong solutions.
Failure two: the decisions you don't make.
Harder to see. You do not run a margin analysis by SKU because assembling the data takes a day. You do not catch that a supplier's price crept up 11% across four POs because nobody looks across POs. You do not notice a slow-moving SKU is eating 30% of your working capital until it expires.
The spreadsheet does not prevent these decisions. It just makes each one expensive enough that you keep deferring it. The cost is entirely invisible because it consists of things that never happened.
Failure three: the single point of knowledge.
One person understands the sheet. Sometimes it is the founder. That person cannot take a proper holiday, cannot be hit by a bus, cannot be promoted out of the work, and cannot be replaced without a two-month handover that will not actually work.
This is the one that caps the company. You cannot grow past the throughput of the person who holds the model in their head — and no amount of hiring fixes it, because new people are hired into a system that only makes sense to its author.
The three false exits
When founders decide to fix this, they usually reach for one of three things. All three are reasonable. All three commonly fail for the same brands.
"We'll get an ERP." Real ERPs — the ones built for food manufacturing with proper lot genealogy — start at a level of cost and implementation effort that is disproportionate for a brand under about $5M. The implementation is not the licence fee; it is six months of someone's full attention plus a consultant. Brands at this stage buy them, deploy 15% of them, and end up running the ERP and the spreadsheet.
"We'll get an inventory app." There are good ones. They solve inventory. Inventory is one tab. The batch records, the CoAs, the supplier specs, the compliance documents — the parts that actually generate the change-propagation problem — sit outside the tool, and now you have a spreadsheet, an app, and a new sync problem between them.
"We'll build it on Airtable / Notion." This works longer than people expect and then fails in a specific way: it recreates the same relational looseness with a nicer interface. Nothing enforces that a batch record must reference a valid input lot. You have rebuilt the spreadsheet with better views, and you have added a monthly fee.
None of these are stupid choices. They are just answers to a slightly different question than the one you have.
What actually replaces it
The useful question is not "what software should we buy." It is: which relationships in our operation need to be enforced rather than remembered?
For most food, supplement, and beauty brands, the list is short — usually four to six:
- A finished lot must reference its input lots
- An input lot must have a CoA attached to it, not to a folder
- A bill of materials change must flag every document that references the changed component
- A batch record must be generated from the current BOM, not a copied template
- A supplier must have a current, dated spec on file for every material they supply
- A shipment must reference the finished lots it contains
Six enforced relationships. That is the difference between a document pile and a system.
Everything else — reporting, dashboards, forecasting — is downstream and can stay in a spreadsheet quite happily. Spreadsheets are genuinely good at analysis. They are bad at being the source of truth.
This is why the answer is rarely "buy the big thing." It is usually: identify the handful of relationships that must not be allowed to drift, put those in something that enforces them, and let the spreadsheet keep doing the analytical work it is actually good at.
A diagnostic
Four questions. Answer honestly.
1. If your ops person left tomorrow, how long until someone else could run a production week without calling them?
2. When you last changed a supplier, how did you find every document that needed updating? (If the answer involves the word "remembered," that is your answer.)
3. Pick a finished lot from three months ago. How long to produce every input lot, every CoA, and every customer it shipped to?
4. How many hours a week does your team spend making two systems agree?
If question 1 is more than a week, question 2 involves memory, question 3 is more than a day, or question 4 is more than four hours — the spreadsheet is not a tool you are using. It is a constraint you are operating inside.
The point
The spreadsheet is not the villain. It got you here, and for a company doing $200K it was the correct answer.
But there is a specific threshold — somewhere between the second SKU family and the first serious retail account — where a brand stops being small enough for a document pile and is not yet big enough for enterprise software. Almost every indie CPG brand spends two or three years in that gap.
The brands that struggle in that gap are not the ones with the messiest spreadsheet. They are the ones who assume the gap is a phase to be endured rather than a problem to be solved, and who keep adding columns until something external — an audit, a recall, a resignation — forces the issue on someone else's timeline.
It was supposed to be temporary. Check how long it has been.
Custom Mango builds custom operational software for CPG brands stuck between spreadsheets and enterprise ERPs. If the diagnostic above stung, book a FoundationSprint.
Custom Mango builds operational software for growing food, supplement and beauty brands. If this article hit close to home, book a free discovery call — the first conversation is about your business, not about software.