What is average check (avg ticket)?

Average check (also called average ticket, average sale, or average bill) is total net revenue divided by the number of bills or guests in a period. It is the cheapest, fastest signal in restaurant analytics: it requires no new instrumentation, it moves week-over-week with real changes in behaviour, and a sustained drop of 5-8% is almost always an early warning that something — pricing, upsell discipline, mix shift — has quietly broken.

Definition

Average check is total net revenue divided by a count. Which count you use defines the convention. Per transaction: divide by closed bills. Per cover: divide by guests served. The two answer different questions and operators who mix them get confusing data. A four-top with a €120 bill is a €120 transaction average check and a €30 per-cover average check — both right, both useful, for different decisions.

The denominator is always net revenue — after VAT, card fees and refunds. Using gross inflates average check by 10-25% (the size of the gross-to-net gap) and bakes a false benchmark into your tracking. The numerator excludes service charge or tronc that gets passed through to staff (that money is never yours; see gross vs net revenue).

Two conventions

Use both. Per-cover catches upsell and menu issues (each guest is spending less). Per-transaction catches group-size and seating issues (you are seating smaller groups). A coffee shop with no group dynamics can use per-transaction exclusively; a full-service restaurant should track both. See cover count for the headcount side of the equation.

The formula

Average check (per transaction) = Net Revenue / Transactions
Average check (per cover)       = Net Revenue / Covers

Week-over-week change %         = (This week − Last week) / Last week × 100

Where:
  Net Revenue   = gross revenue − VAT − card fees − refunds
  Transactions  = closed bills (POS count)
  Covers        = guests served (POS or door count)

The week-over-week comparison is where average check earns its keep. Day-over-day is too noisy (one large group distorts a small day). Month-over-month is too slow (a four-week drift is already an unrecoverable problem). Weekly with a four-week rolling average is the sweet spot for catching drift early.

Worked example

Tuesday lunch at a casual bistro, four consecutive weeks.

Headline revenue looks roughly flat at €1.150-€1.250. But average ticket has slipped from €20,00 to €18,06 — a 9,7% drop in four weeks. Per-cover has dropped from €10,00 to €9,03, the same 9,7%. The drop is not group-size; it is genuine per-guest spend. The diagnostic question: what changed four weeks ago? Did you remove a starter? Did a higher-ticket main get re-priced down? Did the server team stop offering desserts? At €1.156 a week and a 9,7% drop, you are leaving €112 a week — roughly €5.800 a year — uncaptured on Tuesday lunch alone. See the menu pricing playbook for the re-pricing sequence and coffee shop KPI tracking for the weekly tracking routine.

Rules of thumb

Average check has no universal 'good' number — a fine-dining cover and an espresso-bar ticket are not comparable, and even two bistros on the same street differ by menu and location — so the useful rules of thumb are about how you read the trend, not the absolute figure. These are standard tracking guidelines, not fixed targets, and they are what turn a raw weekly number into an early-warning signal.

Common mistakes

Average check is simple arithmetic, which is exactly why it gets read wrong so often — the maths is never the problem, the reading is. These five errors are why a genuine drift can hide in plain sight for a month while every headline number looks reassuringly normal.

• Watching revenue instead of average check. If covers rise as spend-per-guest falls, headline revenue stays flat and the drift is invisible. Only average check exposes a per-guest slip when volume masks it.
• Mixing per-cover and per-transaction. A four-top with a €120 bill is a €120 ticket and a €30 cover — both correct, for different questions. Comparing one week's per-cover figure to another week's per-transaction figure produces nonsense.
• Using gross revenue as the numerator. Dividing the gross till total by covers inflates average check by 10-25%. Use net revenue — after VAT, card fees and refunds — or the benchmark you set yourself is a fiction.
• Reading day-over-day. A single large group or a quiet Monday swings a daily average wildly. Average check only earns its keep at the weekly-versus-four-week-rolling level, by day-part.
• Counting pass-through tips. Service charge or tronc that goes to staff was never your revenue, so folding it into the numerator overstates the ticket and hides a genuine drop underneath.

How it shows up in your daily P&L

Average check is the top of the funnel that every line of your daily EBIT hangs off. nouz starts from gross revenue, subtracts tax and card fees to reach net revenue, then subtracts COGS, variable costs and the day's slice of fixed costs to land on same-day EBIT — and average check is simply that net revenue divided by the day's covers or transactions. When average check slips but your cost lines hold, the EBIT thins for a reason the daily P&L makes obvious: you served the same number of people for less money each.

That is why average check belongs on the same daily screen as EBIT rather than in a separate weekly report. A silent 9% ticket drift over four weeks quietly erases EBIT while covers and costs both look normal — so nouz shows average check per session next to the profit line and flags the week-over-week change, catching the leak while you can still act on menu mix, pricing or upsell training. Seen this way, average check is less a vanity metric and more an early-warning light on the same dashboard as the profit number it ultimately drives.

Why it matters

Average check is the metric most often broken without anyone noticing. Three usual causes. Pricing drift: supplier costs rose, you absorbed them, your effective margin per ticket fell. Mix shift: guests migrated to the new lower-priced bowl and stopped ordering the steak. Upsell collapse: a new server cohort stopped offering starters, sides or desserts. None of these show up in headline revenue if covers stay constant, but all three quietly destroy margin.

The operators who catch ticket drift early have a single habit: weekly average check tracked against a four-week rolling average, by day-part, with a 5% threshold for investigation. A 5% drop sustained two weeks running gets a Monday morning review of menu mix, supplier prices, and the previous fortnight's training. nouz shows average check per session on the daily P&L and flags the week-over-week change in the same line.

Related concepts

• Cover count — the per-guest denominator for average check per cover.
• Food cost percentage — moves with average check via menu mix.
• Gross vs net revenue — why the numerator is always net.
• Café menu pricing playbook — re-pricing sequence when average check drifts.
• Coffee shop KPI tracking — daily-tracked KPI list including average check.

Common questions