Brewery and taproom inventory: three yield clocks and one quarterly stock review

What to take away

The three clocks

A restaurant buys and sells the same thing. A brewery buys one thing, converts it over weeks, and sells something else. That conversion is what makes inventory here awkward.

Raw materials sit on the first clock. Malt, hops, yeast, cans, crowns, labels and cleaning chemicals all arrive on lead times you do not control, and the lead times are not equal. Some are ordered weeks out and some months out. Running out of one stops the brew day regardless of how much of the others you hold.

Beer in process sits on the second clock, which is set by biology rather than by purchasing. Fermentation and conditioning take the time they take. You cannot expedite a batch to cover a Friday.

Finished beer sits on the third clock, which runs down. A keg has a freshness window and a can has a shorter practical one. Beer that is technically sellable but past its best is inventory you own and would rather not pour, which is a different problem from inventory you do not have.

Plan against all three or you will solve one at the expense of another. Buying malt in bulk to save on the first clock does nothing for the third.

Yield, in named quantities

Work this out in your own units rather than accepting a rule of thumb. Define these for your operation and write them where the numbers live:

One gallon is 128 fluid ounces, so the theoretical pours in a full keg are V times 128, divided by G. That is the number your point of sale will never reach, and the distance between the two is the thing worth measuring.

Then define the four losses:

Measure each one separately for a quarter. Do not average them into a single yield figure, because the fixes are unrelated. Cellar loss is a process question for the brewer. Packaging loss is a packaging-line question. Draft loss is a glassware, temperature and training question for the taproom. Unsold loss is a forecasting question for whoever decides what to brew next.

Your own numbers are the only valid ones here. Any published benchmark was measured on somebody else's brewhouse, glassware and tap layout.

Where draft loss actually comes from

This is the loss that touches payroll, because most of it is human and trainable.

Common sources include foam from temperature or pressure problems, overpours, spills during keg changes, beer discarded during line cleaning, samples and training pours. Record comps and samples; it is the only way to tell a training problem from a gas problem from a theft problem. Those three have nothing in common except that all show up as beer that left the keg without a sale.

Sourcing: what to lock and what to leave open

Split your purchasing into three tiers by how much notice the vendor needs and how badly a gap hurts.

Contracted purchasing fits critical items with predictable demand and long lead times. Recurring purchasing suits steady-use items ordered on a regular schedule. Spot purchasing is for uncertain or specialty needs when supply is available; relying on it for critical items risks higher costs or late deliveries.

The mistake is treating everything as spot because it feels flexible, then discovering the flexible approach costs more per pound and arrives late. The opposite mistake is contracting a specialty ingredient for a beer whose demand you have never tested.

Packaging materials need their own attention. They carry approval steps that ingredients do not. A label is a regulated item, not just a design.

Artwork approval lead time is separate from printing lead time. Decide what you are packaging as part of the draft and packaged lineup before you order anything with a printed date on it. Where federal label approval applies, submit the label through TTB's COLAs Online system and obtain a Certificate of Label Approval (COLA) before interstate sale; check whether your product qualifies for an exemption.

The Alcohol and Tobacco Tax and Trade Bureau: Starting a TTB-Regulated Business: Brewery page gives the federal starting point for a brewery, including the permit and the reporting a brewer owes.

Treat it as the frame for your record keeping, not a checklist filed once. The federal side tracks beer produced and beer removed, which your inventory system already counts. The TTB Brewer's Report of Operations reports brewery activity; filing is monthly, quarterly, or annually depending on your filing status and eligibility.

Records that serve two purposes at once

Your inventory numbers and your tax numbers should come from the same source. Keeping two sets is how they diverge.

The Internal Revenue Service: What kind of records should I keep? guidance lets a business choose its own recordkeeping system. That system must clearly show income and expenses, and the supporting documents must survive.

Applied here, the count sheet, the packaging log, and the sales export must reconcile with each other. A person has to be able to do that reconciliation on a Monday morning without heroics.

Build it so that one person, in under an hour a week, can answer: what did we make, what did we package, what did we sell, and what is the difference. If the answer takes longer than that, the system will stop being used within two months.

Where you also serve food, the storage and handling rules come from whichever code your state has adopted, which the U.S. Food and Drug Administration: State Retail and Food Service Codes and Regulations page identifies state by state. That affects how you store, label and date things in the cooler that beer also lives in.

If your counts live in a spreadsheet or a cloud system, they are also a security question. The National Institute of Standards and Technology: NIST Small Business Quick-Start Guides cover the basics for a small operation, and the relevant part here is simple: know who can change a count, and keep a copy you can restore from.

Physical constraints you inherit from the building

Cold storage is the limit that surprises people. Your keg capacity, not your fermenter capacity, often decides how much finished beer you can carry. Your existing walk-in fixes it. Dry storage for malt and empty cans works the same way.

Raw material and finished goods capacity belong in the site evaluation, not discovered afterward. If you are choosing a building, treat storage volume as a first order question in how the site itself is evaluated.

Waste is the other physical constraint. Spent grain, spent yeast, trub and wastewater all leave the building.

How they may leave is a local matter tied to your permits and to what your license and local approvals require. Confirm the arrangement before your volume changes, because the route that works now may not be permitted at twice that output.

A quarterly review that is worth the hour

Doing this four times a year gives you a genuine picture. Doing it once, after a bad month, gives you an anecdote. The habit is worth more than the precision, and it sits naturally alongside the wider startup and market picture when you are deciding what to brew next.

Common questions

How much beer should I keep on hand?

Enough to cover your longest realistic gap between packaging runs plus the variability of your busiest week, and no more, because the third clock is running. Derive it from your own weekly depletion rather than from a target number of kegs.

Is a keg count the same as an inventory system?

No. A keg count tells you what is in the cooler. An inventory system tells you what is in the cooler, what is in tank, what is on order and what you are about to run out of. The first is a snapshot and the second is a forecast.

What is a reasonable loss figure to aim for?

Aim to know yours before aiming to improve it. Any figure quoted without the brewhouse, the packaging line and the tap layout it came from is not comparable to your operation, and chasing somebody else's number usually means measuring differently rather than losing less.

Should staff pours and comps be recorded?

Yes, every one, by name of the person who authorized it. Not because you suspect anyone, but because unrecorded pours are the noise that hides every other signal in your yield numbers.

Who should own the count?

One named person, with a second person checking it monthly. Shared ownership of a count means nobody owns it, and the count is the input to everything else on this page.