Comps & Voids Calculator
Every venue comps food and voids items — it is a normal part of looking after guests and fixing mistakes. The question is not whether it happens, but what it costs you a year and whether it is concentrated on one person. This works out both.
1. What comps, voids and discounts cost you
Totals for one trading period, from your POS reports.
2. The part that actually tells you something
A venue-wide comp rate means very little on its own. What matters is whether adjustments are spread across the team or concentrated on one person. Enter at least three people to compare each against the team median.
Everything is worked out in your browser. Nothing you type is sent anywhere or stored.
What is a normal comp and void rate?
Honest answer: there is no credible industry number, and you should be wary of any page that gives you one. A neighbourhood bistro that comps a dessert when a main runs late, a hotel restaurant absorbing charges onto a guest folio, and a high-volume bar with a heavy happy-hour promo will produce completely different rates, and all three can be well run.
What is genuinely diagnostic is your own baseline and the spread across your team. A stable rate that has not moved in a year is a policy. A rate that doubled last month is an event. And a rate that looks fine in aggregate but comes almost entirely from one till is the pattern worth understanding.
Comps and voids are not the same thing
They get lumped together in conversation and they mean different things:
- A comp is food or drink that was made and given away — a service recovery, a regular looked after, a staff meal. The cost is real: you bought the ingredients and paid someone to cook them.
- A void removes an item before it is settled — a mis-key, a changed order, a duplicate. If it was voided before the kitchen made it, it costs nothing. If it was voided after, the food left the kitchen and the sale did not.
- A discount is a reduced price on a completed sale — promotions, staff rates, loyalty offers. Usually intentional and budgeted.
The distinction that matters operationally is timing: adjustments made before the item is fired are administrative, and adjustments made after are a real product loss. If your POS does not let you separate those two, that is the first thing worth fixing — it is the difference between a typing error and food walking out.
Patterns that deserve a closer look
None of these prove anything on their own. All of them are worth understanding rather than ignoring:
- Concentration. One person consistently well above the team median, across enough shifts that it is not a sample-size artefact.
- Post-settlement edits. Bills reopened and amended after payment. There are legitimate reasons; there should not be many.
- Repeat reprints on the same table. Especially late in a shift.
- Adjustments clustered in one hour — often the quietest one, or right at close.
- Cash-heavy tables with a high void rate. The combination matters more than either alone.
- Manager approvals that never vary. If one code authorises everything, the control is not a control.
Before you act on any of this
An unusual number is a question, not a finding. High-value sections, training shifts, a server who always gets the difficult tables, or a policy nobody wrote down all produce exactly the same signature as something worse — and the innocent explanation is the more common one. Ask before you conclude. Getting this wrong costs you a good member of staff and the trust of everyone who watches it happen.
How to bring the rate down without making service worse
The goal is not zero. A venue that never comps anything is a venue that never fixes a guest's bad night. What works:
- Write the policy down. Most comp variance is not dishonesty, it is twelve people improvising twelve different rules.
- Require a reason code. Not to police staff, but so that in three months you can tell whether comps are a kitchen speed problem or a menu problem.
- Separate pre-fire voids from post-fire voids in reporting. Only one of them costs you food.
- Review weekly, briefly. A five-minute look every week beats a forensic audit once a year, because the trend is the signal.
- Make approval mean something. Individual manager logins, not one shared code.
Or have it watched for you
This page needs you to pull the numbers and type them in. FullGuest's Leak Watch reviews comps, voids, discounts, reopened bills and refunds continuously across every till and outlet, and surfaces the ones that stand out with the evidence attached — framed as something worth a look, not an accusation.