Make to stock means making products ahead, to a forecast, and selling them from the shelf. Make to order means making a product only once a customer has ordered it. Make to stock gives customers their goods today and leaves you holding inventory and the risk it doesn't sell; make to order holds almost nothing but makes the customer wait. The choice for each product comes down to two things: how long it keeps, and how long the customer will wait. Many small kitchens end up doing both, plus a middle way, assembling to order from parts made to stock. Below are the differences, how to choose, and how one kitchen runs all three.
The short version
- Make to stock (MTS): make ahead to a forecast, sell from the shelf. Fits products that keep and sell steadily.
- Make to order (MTO): make after the order. Fits custom work and things that don't keep.
- Assemble to order (ATO): make the parts ahead, put them together when the order comes. Gift boxes, decorated cakes.
- Choose by shelf life against the wait the customer accepts, then by how predictable demand is.
- In one kitchen, orders take capacity first; stock fills the rest, protected by a safety stock.
- TaroStack plans both from the same orders, stock and history. More below.
Side by side
| Make to stock | Make to order | Assemble to order | |
|---|---|---|---|
| Made when | Before anyone orders | After the order | Parts before, product after |
| Planned from | A forecast | The orders | A forecast for parts, orders for the finish |
| Customer waits | No | Yes, the lead time | A little |
| Stock you hold | Finished goods, plus safety stock | Ingredients only | Parts |
| The risk | Making what doesn't sell, or running out | Losing customers who won't wait | Parts nobody orders |
| Cash | Tied up in finished goods | Paid ahead, often with a deposit | In between |
| Fits | Chips, granola, sauces, soap, candles | Custom cakes, party trays, private label | Gift boxes, decorated cakes, variety packs |
How to choose, product by product
Ask two questions, in this order.
Does it keep longer than the customer will wait? If a product keeps for months and customers want it today, make it to stock: there's no other way. If it keeps for a day and customers will order a few days ahead, make it to order. If it keeps for a day and customers want it today, you make it to stock on a one-day horizon, every morning, which is what a bakery's daily bake is.
How predictable is the demand? Steady sellers are safe to make to stock, because the forecast is good. Lumpy, occasional or custom products are better made to order, because a forecast for them is a guess, and a wrong guess sits on the shelf until it expires.
Then a practical check: the batch size. If the smallest sensible batch makes 360 bags and a customer orders 12, you're making to stock whether you meant to or not, and the other 348 need a forecast to sell against.
One kitchen, all three
Windward Roots, the made-up taro business I use in these examples, does all three in one kitchen:
| Product | Keeps | Customers wait | Strategy |
|---|---|---|---|
| Taro chips, 5 oz | 90 days | No | Make to stock: weekly plan with safety stock |
| Taro sweet bread, shop | 5 days | No | Make to stock, daily, by par level |
| Taro sweet bread, café | 5 days | Standing order | Make to order |
| Kūlolo party trays | A few days | 5 days' notice | Make to order, with a deposit |
| Holiday gift boxes | Parts keep | 2 days | Assemble to order from chips and cookies made to stock |
The chips are planned a month ahead in the production plan, with a safety stock against a bad week. The shop's bread is decided every afternoon on the daily bake sheet. The party trays are made only when ordered, five days ahead, and a deposit covers the taro if the party is cancelled. And the gift boxes are packed when the order comes, from chips and cookies already on the shelf, so the box costs nothing until someone wants one.
Running both without them fighting
The trouble with a mixed kitchen is that the two kinds of work compete for the same oven and the same hands. A few rules keep them apart:
- Orders take capacity first. A party tray for Saturday is a promise; a batch of chips for next week's stock can move. Book the orders, then fill the rest of the week with stock.
- Set a lead time for orders, and keep it. "Five days' notice" lets the order go on the plan before the week fills up. Saying yes to a tray for tomorrow is how stock runs out on Thursday.
- Protect stock with a safety stock. It's the buffer that lets a big order push a stock batch back a day without the shelf going empty. The safety stock formula works it out.
- Make stock on the quiet days. Long-life stock products are the flexible work: put them where the orders aren't, as in the production schedule example.
- Take a deposit on custom work. It turns a made-to-order product's biggest risk, a cancellation after the ingredients are bought, into a small one.
Where this stops working
On a whiteboard, a mixed kitchen works as long as one person holds it all in their head: which orders are booked for which day, how much stock is on the shelf, what the forecast says, and how many oven hours are left once the orders are in. The day that person is off, or the orders double in December, the whiteboard says the oven is free on Thursday and the stock says otherwise. And the gift boxes are worse: each one draws on three products' stock, so a run on boxes can empty the chip shelf the wholesale customer was counting on.
How TaroStack does it
TaroStack plans both kinds of work from the same numbers. Made-to-order products come from sales orders, each with its needed-by date, and planning works backward from that date. Made-to-stock products are planned from a forecast filled from your own sales history, with safety stock and minimum and maximum levels per item and site. Both land in one plan, checked against your work centers and their hours, so the party trays and the chip batches aren't both promised the same oven, and every suggestion is a draft you approve.
A sales order says plainly, while you're typing it, whether the stock to fill it is on the shelf. Shelf life is part of the plan, so stock isn't made so early it expires, and a batch made too early for a customer's minimum days-left is flagged before you make it. A gift box is a kit: it holds no stock of its own, and when it ships, each part is picked from its own lots, so the chips in a gift box are on the recall report like any other chips, and a store is only told it can sell as many boxes as the scarcest part allows.
Sales orders are on every plan, from $49 a month; recipes, batches and planning are on Standard at $99. The first 30 days are free, and your items, recipes and open orders import from a spreadsheet. If you'd like a hand setting up, ask, and we'll do it with you.