How to Track Cafe Food Waste With Your Data

The number nobody can see

You buy 180 gallons of milk in a month. Your sales say you should have poured about 156. Nobody dropped a jug in front of you, nobody wrote anything down. Twenty-four gallons are just gone, and if you asked your team where, you would get shrugs.

That gap is waste, and it is the hardest cost in a cafe to see because nothing announces it. A missed order you feel at 4 PM when the shelf is empty. Waste never sends a signal. It shows up only when you line up what you bought, what you sold, and what is still on the shelf, and notice the three numbers do not agree.

I run a small specialty coffee and matcha shop, and for a long time I had no idea what mine was. I assumed it was low because I am careful. When I finally tracked it, it was not low, and it was not where I guessed. This post is how to find yours from data you already have.

How much are we talking about

You will read a lot of tidy benchmark numbers for cafe waste. I am not going to hand you one, because I did not measure your shop and neither did the people who publish those figures. What I can tell you is the shape of it: in my own shop, once I started comparing purchases to recipe-predicted usage, the gap ran high enough to matter every single month. Call it illustrative and run your own math.

So here is the illustration I will use for the rest of the post. Say you spend $8,000 a month on inventory and 12 cents of every dollar leaks out as waste. That is $960 a month going in the trash, down the drain, or simply unaccounted for. Your real number might be half that or double it. The point is not the percentage. The point is that until you track it, it is a rumor, and you cannot fix a rumor.

Where cafe waste actually comes from

Most owners assume waste means spoiled milk and unsold pastries. Those are real, but in my experience they are rarely the biggest leak. There are five sources, and only one of them is the one you already worry about.

1. Expired perishables

Milk, cream, and pastries have short windows. A gallon that does not get used before its date is a direct cost, and pastries unsold at close are waste unless you donate or discount them. This is the most visible source and the one everyone acknowledges. It is usually not the largest.

2. Over-portioning

A recipe says 2 oz of simple syrup. The barista pours 2.5. That is 25% more ingredient cost on that drink, and across a few hundred drinks a day it stops being rounding error. Over-portioning is almost never on purpose. It comes from loose training, worn-out measuring tools, or a recipe vague enough that "a splash of vanilla" means three different amounts to three people. Tight recipe costs start with recipes specific enough that this cannot happen.

3. Batch dumps

This is the big one nobody tracks. Drip brewed on a timer and dumped when it goes stale. Cold brew that did not sell fast enough. A batch of simple syrup that got contaminated, a pot of chai that sat too long. Batch dumps hide because they feel like normal operations, but they are real beans, water, filters, and time that produced zero revenue. Brew and dump three pots of drip a day and that is three pots a day you paid for and threw out.

4. Breakage and spills

Dropped containers, spilled syrups, a bag that tears in storage. Each one is small. A month of them is not.

5. Untracked usage

Staff drinks, samples for regulars, recipe testing, milk burned on steaming practice, shots pulled to dial in the grinder. This is not waste in the moral sense. It is inventory that leaves the shelf without ringing up, and if you do not account for it, it looks exactly like waste in your numbers.

Cafe waste tracking: the gap between sold and used

Here is the whole method, and it turns on one idea. Your Square knows what you sold. It has no idea what you used. An iced matcha latte with oat on the receipt is a $6.50 line item to Square. To your shelves it was a scoop of matcha, ten ounces of oat milk, a cup, and a lid. Cafe waste tracking is just measuring the distance between those two views, and closing it. The same sold-versus-used gap is the whole argument in connecting your POS to ingredient inventory; waste is what falls into it.

The math is two lines:

Starting count + received orders - recipe-predicted usage = expected ending count

Expected ending count - actual count = shrinkage (waste plus untracked usage)

The first line takes what you had, adds what came in, and subtracts what your sales should have consumed through your recipes. The second line compares that prediction to what you actually count on the shelf. Whatever is missing is your waste plus the untracked usage from source five above.

Every piece of that is data you already generate. The received orders are on your invoices. The recipe-predicted usage is your Square sales run through your recipes, modifiers and all. The only input you have to supply is an honest count on both ends, which is why consistent counting is the whole foundation. Skip the count and the gap is unmeasurable, and waste stays a rumor.

You do not need software to run this once. A spreadsheet and one careful count will show you the shape of it. Software earns its place when you want the recipe-predicted usage computed for sixty items automatically instead of by hand, and when you want the gap tracked week over week without redoing the math each time.

Reading waste by category

Not all waste is equal, and not all of it is fixable at the same rate. Breaking the gap down by category tells you where to spend your attention. The ranges below are what I have seen in my own shop, not benchmarks to hold yourself against. Track your own and the real numbers will surprise you in at least one category.

Milk and dairy

In my shop, dairy is one of the leakier categories, and spoilage plus over-steaming drive most of it. Track it by comparing milk purchases to recipe-predicted pours, which is really just how much milk your cafe actually uses measured against what came in the door. Sell 200 lattes at 10 oz of milk each and you should have used about 156 gallons. Buy 180, count what is left, and if the shelf confirms it, 24 gallons went to waste. Now it is a number you can chase, not a shrug.

Coffee beans

Espresso waste tends to be low, because every shot is made to order. Drip and batch brew is where it climbs, because of the dump cycle. The metric to watch is brewed-but-not-sold: pots you made against pots' worth you actually rang up.

Paper goods

Cups and lids rarely go to waste unless they are damaged in storage. If this category shows a real gap, look for a storage problem or cups walking out for non-customer use, not spoilage.

Syrups and specialty ingredients

Contamination, expiration, and over-portioning are the drivers here. These items cost a lot per unit, so even a small percentage of the gap turns into real dollars fast. Watch the expensive bottles.

Three fixes with data behind them

Fix 1: Tighten batch schedules

If your dump data shows you brew ten pots of drip a day but only sell seven pots' worth, brew less. Move to on-demand during slow stretches instead of a timer. This is the single biggest lever on bean waste for a drip program, and it needs nothing more than hourly sales mapped against your brew times. Your 30 days of sales data already holds the hourly shape.

Fix 2: Set a waste target by item

Once you have a baseline gap for an item, set a number to beat. If milk is running 13%, aim for 10% next month. Put the number where the team can see it and check it weekly. People move the number they can see. When a shift lead knows dairy waste dropped from 13% to 9% on their shifts, it sticks.

Fix 3: Order to demand, not to feeling

Over-ordering is the upstream cause of most spoilage waste. Order 20 gallons on Friday because "weekends are busy," use 16, and the extra four are at risk of expiring before anyone pours them. Day-of-week usage fixes this. When you know Saturday burns 9 gallons and Sunday burns 7, you order 16 and a small buffer, not 20. That is exactly what par levels are for: a trigger built on what the shop actually uses.

The compound effect

Take the illustration from the top. Move waste from 12% to 8% on $8,000 a month in inventory and you keep $320 a month, or $3,840 a year. For a single-location cafe, that is real money, and it is the low end of what tracking usually surfaces once you start looking.

The bigger win compounds. Better waste data makes for better ordering. Better ordering means less overstock. Less overstock means less sitting past its date. Each turn feeds the next, and the numbers underneath it all get more honest. When you know your true waste, your recipe costs stop lying to you, your profit reports match reality, and even a decision like your batch-brew schedule has something under it besides a hunch.

The starting move never changes. Count consistently, compare the counts to what your sales should have used, and measure the gap. Track that one number and it will point you straight at where to work.