How to Calculate Labor Cost Per Item in a Scratch Bakery

A bakery owner asked a question that almost every scratch bakery eventually asks:

"We are a scratch pie shop making sweet and savory pies as well as other bakery items, so determining the actual labor it takes to produce one pie has been challenging. We know our COGS, but the labor piece is much more difficult. A couple of team members make the pie dough, another presses the dough, another makes the filling, and then another may bake the item. We have done time studies on different parts of the process, but that can also vary depending on the skill and efficiency of the team member."

If you run a scratch shop, you already know your flour cost to the penny. Labor is the part nobody can pin down.

Here is the honest answer: you are not going to find the labor cost of one pie. You are going to build a labor cost for a batch, then give each pie its share. That is how every factory in America does it, and it is how your bakery should do it too.

Below are six ways to do that, and the questions that tell you which one fits your shop.

Quick answer: what is the formula for labor cost per bakery item?

Most scratch bakeries should cost labor in three moves.

1. Add up the active labor in a normal batch.

Total labor minutes = Σ (minutes per step × people working that step)

2. Divide by what the batch actually sells.

Labor minutes per item = total batch labor minutes ÷ sellable items produced

3. Multiply by what an hour of your labor really costs.

Labor cost per item = (labor minutes per item ÷ 60) × loaded hourly labor rate

Run the pie shop's own numbers through it. A batch of 24 apple pies takes 140 minutes of hands-on work. Loaded labor is $24 an hour.

• 140 ÷ 24 = 5.83 labor minutes per pie
• 5.83 ÷ 60 × $24 = $2.33 of labor per pie
• Add $6.25 of ingredients and packaging = $8.58 direct cost per pie

That is the number. Now the caveats, because that number is only as good as the four decisions behind it: what counts as a batch, what counts as active labor, what your loaded rate really is, and how you split shared dough and fillings.

Is there an industry standard for calculating bakery labor?

Two answers, and they get mixed up constantly.

For the method: yes. It is called standard costing, and it has been the backbone of manufacturing cost accounting for about a century. You set a standard time for a job, cost it at a standard rate, then compare standard to actual and look at the gap. Standard costing deliberately builds in things like setup time, expected raises, and the learning curve as staff get faster. A pie is a manufactured good. The same math works.

For the percentage: no. There is no single labor percentage that is correct for bakeries. Restaurant benchmarks get quoted at bakeries all the time, and they run from 20–25% of sales for quick service up to 35–45% for fine dining, with prime cost (food plus labor) landing near 60%. A scratch pie shop laminating and filling by hand does not have the same labor model as a shop thawing par-baked product. Use those ranges as a sanity check. Do not use them as a target.

So: standard method, yes. Standard number, no.

Why labor is harder than COGS in a scratch bakery

COGS is easy because a pound of butter goes into a pie and stays there. Labor does not behave that way.

Four things get in the way.

Work is shared. One dough batch feeds apple, cherry, and pecan. One filling run covers three days. Nobody makes "one pie."

Work is split across people. Two on the bench, one on filling, one on the oven. The pie never sits with one person.

The clock keeps running when nobody is working. A pie can be four hours from mix to cooled. That is not four hours of labor.

People work at different speeds. Your bench lead presses shells in 18 minutes. The new hire takes 32. Same pie.

That last one is where most bakeries give up. Don't. It is actually the easiest to solve, and the fix is in Method 2.

The words you need (and nothing more)

First, answer these seven questions. They pick your method.

Before you build anything, answer these. Your answers matter more than the formula.

1. Why am I doing this? Pricing, menu decisions, staffing, or chasing a labor number that has crept up? Different goals need different methods and very different levels of detail.
2. How many products drive most of my sales? Five? Fifteen? A hundred?
3. Do I produce in real batches, or one-off custom work? Batch shops and custom cake studios do not cost labor the same way.
4. How many people touch a typical batch, and at the same time or in sequence?
5. Which components are shared across products? Dough, fillings, crumb, glaze.
6. How stable are my batch sizes? If you sometimes run 24 and sometimes run 6, batch size is your real cost driver, not the recipe.
7. What is my bottleneck? Oven deck space? Bench hours? One person who does all the lattice? Whatever runs out first is where labor costing pays off most.

Hold on to answer 1 and answer 7. Those two decide almost everything.

Six ways to cost bakery labor

Nobody uses only one. Most shops end up running two or three of these side by side.

Method 1: Batch labor costing

The starting point for nearly every scratch bakery.

Stop asking "how long does one pie take." Ask "how many hands-on minutes does a normal batch take."

Batch labor cost = (total active labor minutes ÷ 60) × loaded hourly rate

Labor per item = batch labor cost ÷ sellable items

A 24-pie apple run:

140 ÷ 24 = 5.83 minutes per pie. At $24 loaded, that is $2.33.

Now watch what happens when you actually staff it the way most shops do — two people on the bench, two on fill and crimp, plus 20 minutes of setup and cleanup:

215 labor minutes ÷ 24 = 8.96 minutes per pie = $3.82 per pie at $25.60 loaded.

Same pie. Same recipe. Labor is 64% higher. That gap is not a rounding error, and it is the reason "minutes × people" matters more than any other line in this article.

Best for: repeatable batches, a manageable number of hero products, and shops that have never costed labor before.

Weakness: it tells you what happened, not what should have happened.

Method 2: Standard labor minutes (the fix for the skill problem)

This is the answer to the pie shop's real question.

Your bench lead does it in 18 minutes. The new hire takes 32. So which one is "the" labor cost?

Neither. You set a standard.

A standard is the time a competent baker takes, on a normal batch, on a normal day. Not your fastest person. Not your newest. Time three or four normal runs, throw out the weird ones, and pick a number you would defend to your accountant.

Once the standard exists, your price stops moving every time the schedule changes. And you get something better: a variance.

Labor variance = (actual production hours − standard hours earned) × loaded rate

Say last week's production sheet earned 32 standard hours, and your timeclock says the bakers logged 37.

• 5 extra hours × $24 = $120 that week
• About $6,240 a year

Now you have a real question instead of a vague feeling: why did we spend five hours more than the work called for? Usually it is one of five things — batches ran small, somebody was training, a mixer was down, the schedule was built badly, or the standard is simply wrong. All five are fixable. None of them are visible without a standard.

One caution from the cost accounting world: standards can be gamed. If a manager is judged on favorable variances, they will run oversized batches nobody needs to make the number look good. Use variance to ask questions, not to grade people.

Best for: any shop with more than a couple of bakers and any shop where staff turnover changes your production times.

Method 3: Sub-recipe costing (dough, filling, crumb)

If you run a scratch pie shop, this is probably where your labor actually lives.

Stop trying to push dough labor onto pies. Make the dough its own product.

Cost the dough batch once:

Cost per dough unit = ingredient cost + (dough batch labor minutes ÷ 60 × loaded rate) ÷ units yielded

One person, 40 minutes, one dough batch. At $24 loaded that is $16.00 of labor. Then pick how you spread it. You have three fair options, and they give different answers.

Option A — by unit. Batch yields 60 shells. $16.00 ÷ 60 = $0.27 of dough labor per shell. A double-crust apple pie carries $0.54. A single-crust pecan carries $0.27. Use this when your shells are all about the same size.

Option B — by weight. Batch is 30 lb of dough. An apple pie uses 1.2 lb: 1.2 ÷ 30 × $16.00 = $0.64. A hand pie using 0.3 lb carries $0.16. Use this when sizes vary a lot — 5" hand pies next to 10" deep dish.

Option C — by driver. Allocate by whatever actually consumes the labor. If lattice tops eat bench time and plain tops don't, weight is the wrong driver and pieces-of-dough-handled is the right one. This is activity-based-costing in an apron, and it earns its keep exactly where the pie shop lives: "many machines and products, and tangled processes." It also has a real cost — more pools to maintain, more data to collect — so use it on the two or three components where the answer actually changes a decision.

Then do the same for fillings. A 35-minute apple filling run is $14.00 of labor. Spread over 24 pies, that is $0.58 per pie.

Once dough, filling, and crumb each carry their own labor, a finished pie becomes simple addition: dough labor + filling labor + assembly labor + oven labor + pack labor. No more guessing.

Best for: scratch shops where components feed many finished items. Which is most scratch shops.

Method 4: Product family standards

For shops with big catalogs.

You will never time 140 SKUs, and you should not try. Group by how the work feels, not by flavor.

Now a new blueberry lattice does not need a study. It inherits the lattice number on day one. You refine the handful of oddballs later, when you have a reason to.

Best for: 40+ SKUs, heavy seasonal rotation, and anyone whose spreadsheet project died from scope.

Method 5: Top-down labor percentage and prime cost

Start here if the question is "are my prices supporting this business," not "what does this pie cost."

Labor % = total labor ÷ net sales × 100

Prime cost % = (COGS + labor) ÷ net sales × 100

Sales of $100,000, COGS of $25,000, labor of $30,000:

• Labor = 30%
• Prime cost = $55,000 = 55%
• Everything else — rent, utilities, equipment, software, marketing, your own pay, debt, tax, profit — has to come out of the other 45%.

For scale, restaurant prime cost usually runs around 60% of revenue with labor near 30%. If your prime cost is 55%, you are not in trouble. If it is 78%, no per-pie spreadsheet is going to save you, and you should be looking at price and mix before you look at minutes.

Best for: a monthly gut check, and for deciding whether product-level costing is even your problem.

Method 6: Contribution per labor minute (the one most bakeries never run)

Use this when labor, not money, is your constraint.

Bakeries rarely run out of demand. They run out of bench hours and oven decks. So the sharpest question is not "which pie has the best margin percentage." It is "which pie earns the most per minute of bench time."

Contribution per labor minute = (price − ingredients − direct labor) ÷ labor minutes

The apple pie earns four times what the pot pie earns for every minute of your bench. Both have respectable margins. Only one of them deserves your Saturday.

That does not automatically mean cut the pot pie. It means: price it higher, batch it bigger, simplify the build, run it on a slow day, or accept it as a traffic driver you keep on purpose. In cost accounting terms, when products compete for a bottleneck, the one with the highest contribution against that bottleneck should get preference.

Best for: any shop that is selling out, turning down orders, or thinking about hiring.

Which method fits your bakery?

The number most bakeries get wrong: your loaded hourly rate

A $20 baker does not cost $20 an hour. Not close.

Loaded labor cost = wages + employer payroll taxes + unemployment + workers' comp + benefits + other employment costs

Loaded hourly rate = total employment cost ÷ productive hours

Here is a $20/hour baker, built up line by line.

Then one more step almost everyone skips. You pay for 2,080 hours a year, but that baker is not in production for 2,080 hours. Take out PTO, training, meetings, and non-production cleaning — call it 80 hours — and you have about 2,000 productive hours.

$23.05 × 2,080 = $47,944 ÷ 2,000 = $23.97 per productive hour.

Round it to $24. Your $20 baker costs you $24 an hour of real production time — a 20% burden.

Two reality checks on that 20%:

• It could be much higher. Across US private industry, benefits average 30.1% of total compensation — $14.01 of benefits on top of $32.60 of wages per hour worked, which is about a 43% load (BLS, March 2026). That average includes health plans and retirement that a lot of small bakeries do not offer. Most independent shops land somewhere between 12% and 25%.
• You do not need separate rates for everyone. One blended production rate is fine, and it is what most bakeries should use. Split it only when the wage gap is large and the work is genuinely different — a $32 decorator and a $16 packer, for example.

For context on where wages sit: the median US baker earns $17.86 an hour, or $37,160 a year, and the BLS projects 6% job growth through 2035 with about 36,500 openings a year (BLS Occupational Outlook Handbook). Your loaded rate is going up, not down. Build the model once and let it re-price itself.

Active labor vs. oven time: what actually counts

The pie bakes 45 minutes. Do you add 45 minutes of labor?

Almost always no. Ask one question of every block of time:

Do somebody's hands have to be on this product right now?

• Hands on: mixing, pressing, filling, crimping, egg wash, loading, rotating, pulling, moving racks, boxing, labeling. Count it.
• Hands off: baking, proofing, chilling, cooling, resting. Don't count it — as long as that person is doing other productive work.

That last clause is the whole thing. If your baker genuinely stands there watching the oven because there is nothing else to do, that is real money going out the door, but it is not the pie's fault. In cost accounting it is idle time — an employee being paid while not engaged in productive work. Burying it inside the pie's cost hides the actual problem, which is scheduling.

So: cost the pie at active labor. Track idle time separately. Then fix it with the schedule, not with the price list.

The small-batch problem (this is where margins quietly die)

Some of your labor does not care how many pies you make. Mixing setup, scaling, oven fire-up, cleanup, and washing down the bench take the same time for 6 pies as for 24.

Split that 140-minute apple run into fixed steps (about 50 minutes) and per-pie steps (about 90 minutes at a 24 batch), then rerun it:

The six-pie run costs more than double per pie. Nothing changed except the batch.

This is not a bakery quirk. It is textbook setup cost behavior: a $675 setup spread over 500 units is $1.35 each, and over 1,000 units it is $0.68. Cut the number of setups and you free real capacity.

Practical moves, in the order most shops should try them:

1. Batch the component, not the finished item. Make dough and filling in full runs; assemble to order.
2. Consolidate production days. Savory pies Tuesday and Friday, not a little every day.
3. Set a minimum run size per product and hold to it.
4. Set a par level and bake to par instead of to whatever feels right that morning.
5. Price the small run. If a customer wants 6 lattice pies on a Wednesday, that is a different product than 24 on a Saturday.

If you take one thing from this whole article, take this: for most scratch bakeries, batch size moves labor cost per item more than recipe, skill, or wage rate does.

How to build this in your bakery, in one week

Day 1 — Pick five products. The ones that pay your rent. Not the whole case.

Day 2 — Write down your normal batch. For each one: how many do we actually run at a time, and how many sell out of that run? Trim and scrap come off the top. You cost against sellable units.

Day 3 — List the steps. Dough, divide, press, filling, assemble, finish, oven, cool, pack, clean. Mark which are fixed per batch and which scale per pie.

Day 4 — Time three normal runs. Not the fastest. Not a training day. Record minutes and how many people. If two people work 20 minutes, that is 40 labor minutes.

Day 5 — Build your loaded rate. Wage plus taxes, comp, benefits, divided by productive hours. One blended number is fine.

Day 6 — Set standards. Round to something sane. 6 minutes, not 5.83. Write them where the team can see them.

Day 7 — Compare and decide. Standard hours earned vs. actual hours worked. Contribution per labor minute by product. Then make one decision: a price, a batch size, or a production day.

Then leave it alone until something changes.

How often should you redo this?

Not weekly. Standards are meant to be stable — that is the point of them.

Revisit when something real changes:

• wages or your burden move
• a recipe or portion size changes
• batch sizes or par levels change
• you buy equipment (a sheeter, a depositor, a bigger deck)
• packaging changes
• a product's volume shifts a lot
• your variance stays ugly for a month

That last one matters most. A standard you keep missing is either a process problem or a wrong standard. Both are worth knowing.

The mistake to avoid: chasing false precision

You do not need to know that an apple pie holds $2.8374 of labor.

You need to know that apple is about $2.33 and pot pie is about $4.40, and then do something about it. Raise the pot pie price. Double the run. Drop a step. Move it to Tuesday. Buy the tool that halves the bench time.

A rough model your team actually uses beats a perfect model nobody opens. Get to 90% accurate, then go make decisions with it.

The full question list

Print this. Answer it with your team. The answers are worth more than the spreadsheet.

Why are you costing labor?

• Setting or defending prices?
• Deciding what stays in the case?
• Finding out why labor is up?
• Planning staffing or a hire?
• Quoting a wholesale account?

About production

• What is a normal batch for this product?
• How many sellable units come out of it, after trim and scrap?
• Which steps are fixed per batch, and which scale per unit?
• How many people work each step, at the same time?
• Which steps are hands-on, and which are just the clock running?
• Which components are shared across products?
• How much setup and cleanup does a run take?
• How often do we run undersized batches, and why?
• What is our bottleneck — bench, oven, cooler, or one person?

About people

• What is our loaded rate per productive hour, not per paid hour?
• How much do our times swing between the fastest and newest baker?
• Should we cost at standard times rather than actual times? (Usually yes.)
• How much of last month's production labor was training?
• Is a $30 decorator doing $16 work?

About the money

• What is our labor %, COGS %, and prime cost %?
• Which products bring in the most gross profit dollars — not percentage?
• Which products eat the most labor minutes?
• Which products earn the most per labor minute?
• Which products generate the most waste, remakes, or rework?
• Which products only lose money in small batches?

Where the numbers should live

One last practical note. Every method above dies the same death: it lives in a spreadsheet on one laptop, and nobody updates it after month two.

The parts that need to be automatic are the boring ones. What did we actually produce yesterday? How many did we scrap? How many hours did the bakers log against that production? What does tomorrow's order book say we need to bake? If your POS, your custom orders, and your production list all live in different places, somebody has to hand-assemble that every week, and eventually they stop.

That is the part BakeStreet is built around — orders, production lists, and what actually got made sitting in one system, so the labor question becomes a report instead of a project. But you do not need software to start. You need a stopwatch, a batch, and an honest loaded rate.

Bakery labor cost FAQ

What is the formula for labor cost per bakery item?

Labor cost per item = (total batch labor minutes ÷ sellable items) ÷ 60 × loaded hourly labor rate

For example, 140 labor minutes for 24 pies is 5.83 minutes per pie. At a $24 loaded rate, that is $2.33 of labor per pie.

How do you calculate labor when several employees make one bakery product?

Add up each person's active minutes on each step, then divide by the sellable units in the batch. Minutes must be multiplied by people: two bakers working 20 minutes on the bench is 40 labor minutes, not 20. This is the single most common mistake in bakery labor costing.

Should baking time be counted as labor?

Only the hands-on part. Count loading, rotating, checking, pulling, and moving racks. Do not count the 45 minutes the pie sits in the oven if your baker is working on something else during it. If the baker truly has nothing to do, that is idle time and belongs on the schedule, not in the pie's cost.

Should a bakery use actual labor time or standard labor time?

Standard time for costing and pricing. Actual time for management. Standards keep your prices stable when your staffing changes, and the gap between standard hours earned and actual hours worked is the number that tells you whether production is running well.

How do you handle a team member who works faster than everyone else?

Set the standard on a competent baker at a normal pace — not your fastest and not your newest. Then treat the difference as a variance you can act on: train, re-sequence, or move that person to the step where their speed is worth the most. Do not let your product costs move every time the schedule changes.

Should payroll taxes be included in bakery labor cost?

Yes, if you want the real number. A loaded rate adds employer Social Security and Medicare (7.65%), federal and state unemployment, workers' compensation, and any benefits. Most independent bakeries land 12–25% above base wage; the US private-industry average benefit load is closer to 43% of wages.

How do you cost labor for dough or filling used in several products?

Treat the component as a sub-recipe. Cost the whole batch — ingredients plus its own labor — then allocate it by units, by weight, or by whatever actually drives the work. Once dough and filling each carry labor, the finished pie is just addition.

What is prime cost in a bakery?

Prime cost = COGS + labor, expressed as a percentage of net sales. It is the fastest read on whether your product and labor economics leave enough behind to cover rent, utilities, equipment, marketing, debt, and profit.

What is a good labor cost percentage for a bakery?

There is no single right number. Restaurant benchmarks run roughly 20–25% of sales for quick service and 30–35% for casual, with prime cost near 60%. A scratch bakery with heavy hand work will sit higher on labor and lower on COGS than a shop finishing par-baked product. Compare your labor % against your own COGS, prices, and overhead — not against someone else's shop.

How do you decide which products to cut?

Look at contribution per labor minute, not margin percentage. Divide (price − ingredients − direct labor) by labor minutes. The lowest earners per minute are your candidates — but the answer is often re-price, re-batch, or re-schedule rather than remove, especially if the item brings people in the door.

What is the biggest mistake bakeries make costing labor?

Two, tied. Forgetting to multiply minutes by the number of people on a step, and costing labor at a batch size they rarely actually run.

Bottom line

Nobody makes one pie, so stop trying to cost one pie.

Cost the batch. Split it fairly. Set a standard so a new hire does not change your price list. Give dough and filling their own labor. Use a loaded rate, not a wage. And check what each product earns per minute of bench time, because minutes — not dollars — are what your shop actually runs out of.

Start with your five best sellers this week. You will find one product that is quietly paying for everything, and one that only works when you make a lot of it at once. That is the whole payoff.

Sources

1. US Bureau of Labor Statistics — Employer Costs for Employee Compensation, March 2026. Private industry: wages and salaries $32.60 per hour worked, benefits $14.01, benefits 30.1% of total compensation.
2. US Bureau of Labor Statistics — Occupational Outlook Handbook: Bakers. Median pay $17.86/hour ($37,160/year), 2025; projected 6% growth 2025–2035; about 36,500 openings per year.
3. Internal Revenue Service — Topic No. 751, Social Security and Medicare Withholding Rates. Employer share 6.2% Social Security + 1.45% Medicare = 7.65%; 2026 Social Security wage base $184,500.
4. Internal Revenue Service — Topic No. 759, Form 940, Federal Unemployment Tax Act. FUTA 6.0% on the first $7,000 of wages, credit up to 5.4%, effective rate 0.6%.
5. AccountingTools — Standard costing. How standard labor cost is set (including setup time and learning-curve effects), rate vs. efficiency variances, and the risk of managers gaming favorable variances.
6. AccountingTools — Set up cost. Setup cost per unit falls as run size rises: a $675 setup is $1.35 per unit over 500 units and $0.68 over 1,000.
7. AccountingTools — Activity based costing. Cost pools and activity drivers; works best in complex environments with many products and tangled processes; maintenance cost is the main drawback.
8. AccountingTools — Contribution margin. When products compete for a bottleneck resource, give preference to the highest contribution against that constraint.
9. AccountingTools — Idle time. Time an employee is paid but not engaged in productive activity.

Wage, tax, and benchmark figures are current as of September 2026. State unemployment and workers' compensation rates vary by state and by class code.