Production Planning for a Small Bakery: A Practical 2026 Guide

Why production planning fails in small kitchens

Production planning in a small bakery rarely fails because the plan was wrong. It fails because of where the plan lives and who can see it. There are two classic failure modes, and almost every small food business has lived through at least one of them.

The plan lives in your head. This works remarkably well right up until it doesn't. One person holds the whole week — what is proving, what is going in next, which order is being collected at eleven, which customer changed their mind on Thursday. It is efficient, because there is no overhead of writing anything down. It is also completely fragile. The business cannot open without that person. Nobody else can be usefully delegated to, because delegating requires explaining the plan, and explaining the plan takes longer than doing the job. And when that person is ill, or on holiday, or simply overloaded on a Friday in December, the plan stops existing.

The plan lives on a whiteboard nobody updates. The whiteboard is an improvement, in principle. It makes the plan visible. But whiteboards decay: they are written at the start of the week and then reality diverges from them by Tuesday afternoon, and once a whiteboard is known to be wrong, everybody stops reading it. A whiteboard that is 80% accurate is arguably worse than no whiteboard, because it produces confident mistakes. Someone bakes the wrong batch size because they trusted a number that was three days stale.

The fix is not more discipline. It is a plan that updates itself where possible, that is reachable from a phone as well as a wall, and that keeps a distinction between what you intend to do and what you actually did.

Plan backwards from collection, not forwards from the mixer

The most common structural mistake in bakery production planning is planning forwards. You start with what time you get in, work through what you can make in the time available, and hope it lands before the customers do. This puts all the slack at the end of the day, which is exactly where you cannot afford it, because the end of the day is where the deadlines are.

Plan backwards instead. Start with the fixed points in the day that cannot move:

• Collection and delivery times. A wedding cake collected at 2pm, a wholesale drop that must be on the van by 7am, a market stall that opens at 9am.
• Counter open. What has to be on the shelf when the door unlocks, as opposed to what can arrive by mid-morning.
• Cooling and finishing time. A cake cannot be iced hot. Work back from the collection slot through decorating, through cooling, through baking, through proving, to the mix.

Once you have those fixed points, everything else fills in around them. The important consequence is that the flexible items — the things with no committed customer attached — become genuinely flexible. They are the shock absorber for the week. When something goes wrong, you know immediately what you can drop, because it is the only thing on the plan without a name and a time next to it.

See how FoodCore plans production and orders together →

Batch by oven temperature and proving time, not by product

Orders arrive by product, so it is natural to plan by product: do the sourdough, then do the brownies, then do the sausage rolls. In a small bakery with one or two decks, this is close to the worst possible sequence, because it forces the oven to change temperature between every group.

The constraint in a small bakery is nearly always the oven, then the prover, then the bench. Plan around the constraint:

1. Group by temperature band. Everything that bakes hot goes first — bread, pizza bases, anything wanting 220–240°C. Then step the oven down: pastry and enriched doughs in the middle band, then cakes and traybakes, then anything low and slow like meringue at the end. The oven falls through the day rather than see-sawing.
2. Sequence within a band by proving time. The item needing the longest prove is mixed first, even if it bakes last within the band. This is what keeps the prover full and the oven fed rather than one of them idle.
3. Put the bench work where the oven is busy. Decorating, filling, packing and labelling do not need the oven. Schedule them deliberately into the windows where the oven is occupied, rather than letting them pile up at the end.

The output is not more product per hour in some abstract sense. It is fewer oven cycles, less energy, less standing around waiting, and a plan that a second pair of hands can follow without you narrating it.

Build a weekly rhythm you can repeat

Small bakeries that run calmly almost always have a rhythm rather than a plan. The same broad shape of week repeats, and only the quantities and the special orders change. That is what makes it possible to hire, to take a day off, and to know on Monday what Thursday looks like.

A workable rhythm usually assigns each day a primary character:

• A prep day for the things that keep — doughs to freeze, pastry blocks, fillings, ganache, dry mixes.
• Two or three bake days carrying the bulk of counter and wholesale volume.
• A decorating day ahead of the weekend order peak, when nothing else competes for the bench.
• An admin and ordering slot — supplier orders, costing review, next week's plan — protected as a real block, not squeezed into the gaps.

Rhythm also makes forecasting easier. If Wednesday is always a bake day, you have thirty Wednesdays of comparable data rather than thirty different days. Our companion guide to production scheduling for small food businesses goes further into sequencing and bottlenecks.

Plan against actual sales, not optimism

The quantities on a production plan are where most waste is created, and the cause is almost always the same: planning from hope rather than from history. A useful method:

1. Pull four to six weeks of sales by product by day of week. Not by week — by day. Saturday and Tuesday are different businesses.
2. Use the median, not the average. One exceptional Saturday drags an average upwards and then you bake to that number every Saturday for a month.
3. Add confirmed orders as a separate line. Committed volume is not forecast volume and should never be blended into it.
4. Adjust for known events. School holidays, market dates, a bank holiday, a big local event, and — for anything sold outdoors — the weather.
5. Review monthly. Consistent waste on a product on a particular day is not bad luck. It is the plan being wrong, repeatedly.

If you are not yet confident that your recipe costs are accurate, the quantities question is secondary — you cannot judge whether over-baking is expensive until you know what a unit costs. The recipe costing guide covers that groundwork.

A worked weekly production plan for a small bakery

Here is a realistic week for a small bakery with a counter, a small wholesale round and weekend celebration cakes. Quantities are illustrative — the structure is the point.

Note what the fourth column does. Every run points at something it feeds. A run with nothing in that column is a run you should question, because it is production without a destination — which is another way of saying waste you have not noticed yet.

How the FoodCore production calendar works

FoodCore's production calendar is available on the Growth plan (£40/month inc. VAT) and the Core plan (£65/month inc. VAT). It is designed around exactly the failure modes above — a plan that is visible, current, and honest about the difference between intention and history.

Week and month views, with runs and orders together

The calendar offers a week view for working the plan and a month view for seeing the shape of a period — which weeks are already heavy, where a wedding or a bank holiday sits. Critically, it shows production runs and order fulfilments on the same calendar. That is not a cosmetic choice. The question a baker actually needs answered is not "how much production is scheduled on Friday" but "does Friday's production fit alongside Friday's collections". Two separate screens cannot answer that; one calendar can.

Drag a job to another day to reschedule

Plans change. A supplier is late, an order is moved, someone is off. On the FoodCore calendar you drag a planned job to another day and it is rescheduled — no re-entry, no deleting and recreating.

Completed and cancelled work is not draggable

This is the detail worth pausing on. Once a run is completed or cancelled, it cannot be dragged. It stops being a plan and becomes a record of what happened.

Why that matters: a production record is evidence. It tells you what you actually made on the fourteenth, how much stock that consumed, which batch a customer's order came from. If a stray drag on a phone screen could quietly move last Tuesday's completed bake into this Thursday, the record would be worthless — and you would have no way of knowing it had been altered. Making history immovable is not a limitation; it is the whole reason the record is trustworthy. Your future is editable, your past is not.

Stock deducts automatically, and orders create runs

Two connections do most of the heavy lifting:

• Completing a production run automatically deducts its ingredients from stock, based on the recipe and the batch quantity. Your stock figures follow production rather than depending on a separate counting exercise, and your shopping list reflects what the plan is going to consume.
• Moving an order into production creates the production run. A confirmed customer order does not have to be re-typed as a baking job — the order becomes the run, and the link between them is kept.

Together with the orders and collections diary (also Growth and above), this is what closes the loop between "a customer asked for this" and "we made it, and here is what it cost us in stock". For a bakery-specific walkthrough of how these pieces fit together, see FoodCore's bakery management software page.

See FoodCore for bakeries →

What to do when the plan breaks

It will. A plan's value is not that it is always right — it is that when it goes wrong, you can see immediately what the damage is and what your options are.

• Protect committed orders first. Anything with a customer name and a time attached is untouchable until everything else has been considered.
• Cut speculative volume, not quality. Reducing the counter batch by a third is recoverable. Rushing a prove is not.
• Reschedule rather than abandon. If a job can move to tomorrow without breaking a commitment, move it — and move it on the calendar, so tomorrow's plan is honest about being fuller than usual.
• Record the cancellation. A cancelled run that is simply deleted teaches you nothing. A cancelled run that stays in the record tells you, three months later, that Thursdays keep failing for the same reason.

Bakery production planning: frequently asked questions

Start a 7-day free trial — no card required →

Further resources

• FSA: food hygiene guidance for your business
• GOV.UK: food labelling and packaging
• Production scheduling for small food businesses
• Managing Christmas orders in a small bakery
• Stock management for UK food businesses
• FoodCore bakery management software
• FoodCore product overview