Metrics and profitability

Restaurant KPIs: the numbers to check weekly

A short weekly KPI set for UK restaurants and pubs: net sales, covers, spend per head, GP, labour, prime cost and occupancy, with formulas and an Example.

Ava research teamPublished 6 min read

On this page
  1. Why weekly, and why so few
  2. The eight weekly KPIs
  3. Read sales growth against price inflation
  4. Set a trigger and an owner for each
  5. Add context lines, not more KPIs
  6. Measurement mistakes that mislead
  7. Adjusting the set for pubs and drinks-led venues
  8. Example: one Monday dashboard
  9. Running the weekly KPI meeting
  10. Frequently asked questions
  11. References

Key takeaways

  • Eight numbers, checked every Monday, tell most operators what they need: net sales, covers, spend per head, food GP, drink GP, labour cost, prime cost and occupancy.
  • Always compare sales growth with price inflation; a sales rise smaller than your price rises means fewer guests or smaller orders.
  • Measure everything net of VAT and on the same week definition every time.
  • Each KPI needs an owner and a trigger: the level at which someone acts.

Why weekly, and why so few

Monthly management accounts arrive too late to change the month they describe. Daily numbers swing too much with weather and bookings to guide decisions. Weekly is the rhythm that lets a manager act: change next week's rota, push a dish, chase a supplier price. And a short list beats a long one, because a dashboard of 30 measures is a dashboard nobody reads.

The pressures that make the numbers matter are clear in the national data. In October 2026, UK accommodation and food businesses named economic uncertainty (39.8%), labour costs (39%) and material costs (38.5%) as the top challenges to turnover[1]. Each of those shows up in a weekly KPI before it shows up in the bank balance.

The eight weekly KPIs

A weekly KPI set for a single-site restaurant or pub.
KPIFormulaWhat it tells you
Net sales vs same week last yearThis week's sales excluding VAT ÷ same week last year, minus 1Whether the business is growing, before you look at why
CoversGuests served, by day partWhether footfall is rising or falling
Average spend per headNet food and drink sales ÷ coversWhether guests are spending more or less per visit
Food GP %(Net food sales − food cost) ÷ net food salesMenu pricing, portion control, waste and supplier prices
Drink GP %(Net drink sales − drink cost) ÷ net drink salesPricing, pouring accuracy, stock losses
Labour cost %Total wage cost including employer NI and pension ÷ net salesWhether the rota matches demand
Prime cost %(Food and drink cost + labour cost) ÷ net salesThe controllable cost that decides profit
Seat occupancyCovers ÷ (seats × turns available), by day partWhere empty seats are worth selling

Food and drink cost for GP should come from opening stock plus purchases minus closing stock, not from purchases alone. A weekly stock count of high-value lines, such as meat, fish, spirits and wine, is enough for a reliable figure; a full count can be monthly.

Read sales growth against price inflation

A sales increase is not the same as more trade. Prices in the restaurants and hotels division of the UK Consumer Prices Index rose 3.8% in the 12 months to September 2026[2]. In the same month, like-for-like sales at Britain's leading managed hospitality groups were 2.7% higher than a year earlier, their best growth of 2026[3], and managed restaurant groups grew 2.6%, their best month since November 2024[4]. If your menu prices rose by roughly the same as the index and your sales grew less than that, you are serving fewer guests or selling less to each one.

Like-for-like sales growth at managed hospitality groups, September 2026Year on year. CPI for restaurants and hotels rose 3.8% over the same period. Bar groups fell 1.7% (not charted: this chart type shows positive values only).
Like-for-like sales growth at managed hospitality groups, September 2026Column chart. Pub groups: 3.8%; All managed groups: 2.7%; Restaurant groups: 2.6%.0%1%2%3%4%Pub groups: 3.8%3.8%Pub groupsAll managed groups: 2.7%2.7%All managed groupsRestaurant groups: 2.6%2.6%Restaurant groupsLike-for-like sales growth at managed hospitality groups, September 2026Column chart. Pub groups: 3.8%; All managed groups: 2.7%; Restaurant groups: 2.6%.0%1%2%3%4%Pub groups: 3.8%3.8%Pub groupsAll managed groups: 2.7%2.7%All managedgroupsRestaurant groups: 2.6%2.6%Restaurantgroups

Source: NIQ (CGA by NIQ) with RSM, 2026[3]

Show the data
ItemValue
Pub groups3.8%
All managed groups2.7%
Restaurant groups2.6%

That is why covers and spend per head sit next to sales on the list. Together they explain the sales line: more covers at the same spend, the same covers at higher spend, or a fall in one hidden by a rise in the other.

Set a trigger and an owner for each

A KPI without a trigger is just a number. Agree in advance the level that prompts action, and who acts. The triggers below are an Example for a fictional casual dining restaurant; set your own from your budget and history.

  • Example: food GP below the weekly target by more than 2 points for two weeks running. Owner: head chef. Action: check top ten dishes' recipe costs against current supplier prices, check waste log and portion sizes.
  • Example: labour cost above target by more than 2 points. Owner: general manager. Action: compare rota hours with covers by day part; move hours from quiet to busy shifts.
  • Example: spend per head down for three weeks. Owner: front-of-house manager. Action: check add-on rates for starters, sides, desserts and second drinks by server.
  • Example: occupancy below 40% on a day part for four weeks. Owner: general manager. Action: decide whether to sell that slot, change the offer or reduce opening.
  • Example: covers down year on year for three weeks while bookings are flat. Owner: general manager. Action: look at walk-ins, delivery share and local events.

Add context lines, not more KPIs

Some numbers explain the KPIs without needing their own targets. Keep them as notes under the dashboard:

  • Delivery and takeaway share of sales. Across the market, delivery, takeaway and click-and-collect made up 20.9% of spending with restaurants in September 2026[5]; your own share changes your GP, because commission and packaging sit outside the plate cost.
  • Weather and local events, which explain many swings in covers.
  • Price changes made that week, so later readers know why GP moved.
  • Staff changes, sickness and new starters, which explain labour and service swings.
  • Review score and number of new reviews, as a check on whether savings are hurting the guest experience.

Measurement mistakes that mislead

Most KPI arguments in hospitality are really arguments about definitions. Fix these once and write the definitions down.

  • Changing the week. Pick Monday to Sunday or another fixed week and never move it, or comparisons with last year break.
  • Counting covers differently by channel. Decide whether takeaway orders count as covers. Most operators report them separately, as orders, so spend per head stays comparable.
  • Labour from the rota, not payroll. Rota hours ignore overtime, holiday pay, employer National Insurance and pension. Use payroll cost, or add a fixed uplift that you check against payroll monthly.
  • Including service charge in sales. Tips and service charges belong to staff and should sit outside net sales, or your spend per head and GP will be overstated.
  • Comparing with a closed week. If the venue shut for a refurbishment last year, compare with the nearest normal week instead and note it.
  • Ignoring part-weeks at month end. Weekly KPIs and monthly accounts will never match exactly; reconcile them once a month rather than arguing every week.

Adjusting the set for pubs and drinks-led venues

In a wet-led pub, drink sales may be most of the takings, and covers are hard to count. Swap or add a few measures so the dashboard reflects how the business earns money.

  • Replace covers with transactions, and spend per head with average transaction value, for the bar.
  • Split drink GP into draught, packaged, wine and spirits, because each has a different margin and different loss risks.
  • Add cellar yield on draught: pints sold compared with pints expected from the barrels used.
  • Track food covers separately if you run a dining room, so food GP and occupancy still have a base.
  • Keep stock value on hand as a monthly check; over-stocking ties up cash that a weekly KPI will not show.

For mixed venues, report food and drink separately and together. A rising blended GP can hide a falling food margin if drink sales are growing faster.

Example: one Monday dashboard

Example. A fictional 80-cover neighbourhood restaurant reviews week 44. Net sales were £31,400, up 3.1% on the same week last year. Covers were 1,010, down 1.5%, and spend per head was £31.09, up 4.7%. Food GP was 67.2% against a target of 69%, drink GP 71.5% against 72%. Labour cost was 31.8% of net sales, prime cost 63.1%, and dinner occupancy on Tuesday and Wednesday was 38%.

  1. What it says

    Sales grew only because prices rose; fewer people came. Food GP is under target, and the midweek evenings are half empty.

  2. What the chef finds

    Two suppliers raised prices on beef and dairy three weeks ago and the recipe costs were never updated.

  3. What the manager decides

    Re-cost and re-price the two affected dishes; test a midweek set menu for four weeks; move six rota hours from Tuesday to Saturday.

  4. What gets checked next Monday

    Food GP, midweek covers and Saturday ticket times.

Running the weekly KPI meeting

  1. Same time every week, ideally Monday or Tuesday morning, 20 minutes, standing up if it helps.
  2. The numbers are ready before the meeting, on one page, with last week and the same week last year.
  3. Each owner says in one sentence what moved and why.
  4. Agree no more than three actions, each with an owner and a date.
  5. Start the next meeting by checking those actions.

As labour costs rise, with the National Living Wage at £12.71 an hour from April 2026[6], the weekly view of labour and prime cost matters more each year. For the detail behind each KPI, see gross profit targets for food and drink, how to calculate dish GP and set prices, labour cost percentage and prime cost.

Frequently asked questions

What are the most important KPIs for a restaurant?

For weekly control: net sales against last year, covers, average spend per head, food GP, drink GP, labour cost percentage, prime cost and seat occupancy by day part.

Should restaurant KPIs include VAT?

No. Measure sales and GP net of VAT, otherwise your margins look higher than they are and changes in VAT rates distort comparisons.

How do I compare this year's sales fairly with last year's?

Use the same trading week, adjust for days the venue was closed, and compare growth with your own price increases to see whether trade really rose.

How often should I count stock for GP?

Count high-value lines weekly and do a full count monthly. Weekly GP based on purchases alone is unreliable because stock levels change.

References

  1. Office for National Statistics (2026). Business insights and impact on the UK economy, wave 166 dataset (opens in a new tab) ↑

  2. Office for National Statistics (2026). Consumer price inflation, UK: September 2026 (opens in a new tab) ↑

  3. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (NIQ RSM Hospitality Business Tracker) (opens in a new tab) ↑

  4. NIQ (CGA by NIQ) with RSM (2026). Bright September gives hospitality groups their best growth of 2026 (NIQ RSM Hospitality Business Tracker) (opens in a new tab) ↑

  5. NIQ (CGA by NIQ) (2026). Restaurants' delivery sales strengthen in September after tough summer (Hospitality at Home Tracker) (opens in a new tab) ↑

  6. GOV.UK (2026). National Minimum Wage and National Living Wage rates (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo