What to look for in a maker costing tool

There is no best costing tool, and anyone who tells you otherwise is selling one. There is a tool that fits the way you make things, and the way to find it is not to read feature lists but to ask the same nine questions of every candidate and compare the answers you get back.

This page is that list. It is written to be used against us as much as against anyone else, and it ends with the makers who should not buy a costing tool at all.

Coming from Stocky? Shopify retires Stocky on August 31, 2026, and there is a version of this page written around exactly that move: the Stocky alternatives roundup. Start there if Stocky is the tool you are leaving.

First, decide which job you are buying for

Two jobs get lumped together under the word inventory, and the right tool depends entirely on which one is yours.

• You resell finished stock. You buy products and sell them, and you mainly need counts synced across a point of sale and an online store. That is a retail inventory job, and a retail inventory tool does it well.
• You make products from raw materials. A candle is wax, fragrance, a jar, a wick and a label, and its cost moves every time any of those prices move. You need true per-product cost, not stock counts. That is a costing job.

Most tools do one of these well and the other badly. Sort yourself first and the shortlist gets short fast. If you are a retailer, stop here: the built-in inventory tools in whatever platform you already sell on are almost certainly the right answer, and nothing below applies to you.

The nine questions

1. Does it understand units, or only quantities?

You buy wax by the kilo and use it by the gram; you buy jars by the case and use them one at a time. A tool that stores only a quantity forces you to do that conversion in your head every time, which means you will do it wrong eventually. Ask what happens when a purchase unit and a usage unit differ, and watch whether the answer involves the phrase "just enter it as".

2. Can it hold a recipe, and can a recipe contain a recipe?

A finished product is a list of materials with quantities. Any costing tool will do that much. The second half is what separates them: if you make a scent blend, or a soap base, or a dough, and then use it in three products, you need a sub-recipe whose cost flows upward. Without it you end up maintaining the same blend cost in three places, which is the same as maintaining it in none.

3. Which cost method does it use, and does it say so on the report?

Moving average, latest price paid, and dated first-in-first-out layers all give different numbers from the same purchases. None is wrong. What is wrong is not knowing which one you are looking at, because you cannot explain a margin to an accountant, or to yourself in March, if you cannot say how the cost was computed. The best answer is that the method is named on the report itself.

4. What happens to a product's cost when a material price changes?

This is the question that separates a costing tool from a spreadsheet with extra steps. When your supplier raises the price of jars, does every product using that jar re-cost itself, and do you find out? A tool where you have to remember to go and update things is a tool that will be quietly wrong for months.

5. How does data get in on an ordinary Tuesday?

Not the migration. The daily reality. If entering a supplier receipt means sitting at a desk and typing fourteen lines, you will stop doing it, and a costing tool nobody updates is worse than a spreadsheet nobody updates because it looks authoritative. Ask what capture looks like away from a computer, and whether anything the tool reads for you is reviewed by you before it counts.

6. Can you get your data back out, free, today?

Ask this before you put a year of history in, not after. Some tools in this category put CSV export behind a paid plan, and the pricing page is where you find out. The question to ask plainly: can I export materials, products, purchases and cost history, on the plan I am on, at no extra charge, right now? Anything less than a clear yes is an answer. Ask the harder half of it too, which is what happens when you leave for good: exporting and deleting your account is our answer, and it is a fair thing to demand in writing from anyone on your shortlist.

7. How does it bill, and what happens in your busiest month?

A flat published price and a metered price behave completely differently in December. Read the pricing page for the words "per transaction", "per order" and "per sale", and work out what your biggest month would cost. Also ask what happens when you cross a plan limit: a hard stop mid craft-fair season is a different product from a gentle prompt.

8. Does it fit your batch reality?

Some makers produce one item at a time; others pour forty candles from one melt or cut thirty bars from one loaf. If you work in batches, a tool that only knows single units will make you do arithmetic it should be doing, and yields, waste and part-batches are where the cost error creeps in. Ask what happens when a batch of forty produces thirty-seven sellable units.

9. Where does it stop, and does it admit it?

Every tool has a boundary. Costing is not accounting, and neither is bookkeeping, tax filing, or label compliance. A vendor who claims all of it is either describing a plan as a product or has not thought about it. Rules for labeling and food safety differ by state and by product, so check your own state's requirements rather than a vendor's summary of them, and read labeling and regulations for the honest shape of that question before you assume software solves it.

Answers that should give you pause

• "It syncs your stock levels", offered as an answer to a question about cost. Those are different jobs.
• A cost method that is not named anywhere. If nobody will tell you how the number is computed, you cannot defend it.
• Export available on a higher plan. That is a lock, described politely.
• Pricing that needs a call. For a business your size, a price that is not published is a price that varies by how much you seem able to pay.
• A migration promised as one click. Column mapping is real work in every tool; a vendor pretending otherwise will disappoint you on day one.
• Any claim about compliance that does not tell you to check your own state.

How to test a shortlist in an afternoon

1. Take your five most-used materials, with a real purchase price and a real purchase unit for each.
2. Take one finished product that uses at least three of them, and work out its cost on paper first, so you have an answer to check against.
3. Enter exactly that into each candidate and compare its number against your paper one. Any disagreement is a units question, and finding it now is the point.
4. Then change one material price and see what the tool does on its own. This is question four, answered by the tool rather than by a salesperson.
5. Finally, export and open the file. If your five materials cannot come back out, stop there.

Do that with your own numbers rather than the demo data. Demo data is chosen to make a tool look good, and your material list is the only test that matters.

Who should skip this whole category

• Anyone with a handful of products and stable costs. A spreadsheet is genuinely enough, and there is no shame in it. Revisit when updating the sheet becomes the chore you avoid, because that is the real signal, not a product count.
• Retailers. You need a retail inventory tool, and you probably already have one.
• Anyone shopping before they have priced anything. Work out one product's cost by hand first. It takes an evening, it tells you what you need, and it makes every demo afterward legible.

If you have never done that arithmetic, start with COGS for makers and how to price handmade products. They are the same math a tool would do for you, done once by hand.

Where Batchnook fits, and who should skip us

Since this is our site, here are our own answers to the nine, in the same order and without dressing: units are first-class, with a purchase unit and a usage unit per material; recipes hold materials, and a recipe cannot yet contain another recipe; cost is moving average and the method is disclosed on every report; a price change re-costs every product that uses the material; capture is a phone photo of a supplier receipt, whose lines you approve before anything counts; export is free on every tier, forever, as JSON or CSV; pricing is flat, published, and never changes without at least 30 days' notice; batches are a first-class record with yields; and we stop at costing, deliberately.

And who should not pick us:

• Retailers who mainly need point-of-sale and online stock sync.
• Makers whose hardest problem is labeling or food compliance. That is a specialist's job and we do not build it. Check your state's requirements, and run a specialist tool alongside us if you need one.
• Anyone happy on a spreadsheet. If the sheet works and updating it does not hurt, keep it.
• Anyone who needs full bookkeeping in one place. We produce tax-ready cost numbers, never tax advice, and we are not a ledger.

FAQ

When you have picked one, the move itself is a known quantity: switching costing tools without losing your numbers covers the whole running order, and the export guide covers getting your data out of wherever it is now.