Revenue per seat: the number landlords understand

When a restaurant owner sits down with a landlord to renew a lease, or with a lender to fund a second site, the conversation soon turns to one question: what does each seat earn? Covers, average spend and the busy Saturday nights matter to the owner, but the landlord and the lender think in space. They want to know how hard the room works for the room it takes up.

Revenue per seat answers them. It divides the venue's revenue by the number of seats it has, for a night, a month or a year, and it measures the room by its capacity rather than by its guests. Two venues with the same takings can have very different figures, and the difference tells you a great deal about how each room is used.

This guide covers how to work out revenue per seat, what pushes it up and down, what it reveals about the layout and how to use it in a conversation with a landlord or lender. Every venue and figure in it is an example.

Revenue per seat measures how much each seat in the room earned over a period. A venue that took 600,000 in a year from 60 seats earned 10,000 per seat; one that took the same from 120 seats earned half that.

It looks at the room from the opposite direction to covers. Covers count the guests who came; revenue per seat counts what the room's capacity produced. A venue can serve many covers and still have a low figure if it has a great many seats it rarely fills.

Landlords and property people think in space, because space is what they sell. Retail landlords have long judged shops by sales per square foot, and restaurant landlords translate that into seats: how much a unit's tenant takes for the space it occupies.

Some leases include turnover rent, a share of sales paid on top of or instead of a fixed rent, and the landlord then has a direct interest in the figure. Even without one, a landlord deciding whether to renew, or a lender deciding whether to fund, uses revenue per seat to judge whether the business can carry the space.

Divide revenue for the period by the number of seats. Use the same revenue each time, usually food and drink sales before tax and any service charge, and count the seats the room actually sells, not every chair in the building.

Work it out per year for conversations with landlords and lenders, per month for tracking and per night for comparing services. Each answers a slightly different question, but all start from the same sum.

Count the seats available to guests during normal trading. Include the bar stools if guests eat there, the terrace if it is open for much of the year and the private room if it is regularly used for diners.

Be consistent. A terrace counted in the summer and not the winter makes the figure jump with the seasons; counting it every month, or never, keeps comparisons fair. Note on the report which seats are counted, so the figure means the same thing each time it is read.

Seat counts drift. Tables are pushed together, chairs are borrowed for a party and never returned, a high table by the bar gains two stools. Count the seats in use once a quarter, table by table, so the figure is divided by the room you actually have.

Use the seats each table normally takes, not the most it can squeeze in. A table for four that can hold six at a pinch is still a table for four in the count, and the extra two seats only appear on the nights they are used.

Revenue per seat per night shows how hard the room worked on a single service. A Saturday might earn several times what a Monday does for each seat, which is the shape of the week expressed in seats.

Revenue per seat per year is the figure landlords and lenders usually want, because it smooths out the week and the seasons. It also allows a fair comparison between venues of different sizes, as long as both count seats the same way.

The room earns differently at each service, and the figure for the whole week hides it. Work out what each seat earned at lunch and at dinner, on weekdays and at weekends, to see where the room is strong and where its seats sit idle.

A room whose seats earn well at dinner and little at lunch has a clear opportunity: the seats and the rent are the same at noon as at eight, so every guest added at lunch raises what the space earns.

The figure is the product of two things: how many guests each seat serves over the period, and how much each of those guests spends. Raise either and the figure rises.

The first is about use: covers per seat, which depends on how often tables turn and how well parties fit them. The second is about spend: average spend per head, which depends on prices, the menu and how well the team looks after guests. Most improvements come from one or the other.

Covers per table, the guests each table serves in a service, is the link between seats and guests. A table for four that serves a couple and then a four has served six covers in the evening; one that serves a single couple has served two.

Look at covers per table alongside revenue per seat. A table with low covers for its size is usually too large for the parties it gets, in a spot guests avoid or held too long between seatings, and each of those lowers what its seats earn.

Average spend per head is the other half of the figure. A room full of guests who order one course and tap water earns little per seat, however full it looks. A room where guests order starters, wine and dessert earns far more from the same seats.

Raising spend per head through the menu, recommendations and service raises revenue per seat without adding a single cover. It is often the easier of the two levers, because it does not depend on more guests coming.

Worked out table by table, the figure shows how well the layout works. Some tables earn far more per seat than others, and the reasons are usually visible on the floor plan: the window tables everyone asks for, the corner by the kitchen door nobody wants, the large round that couples are seated at.

A room where a few tables earn most of the revenue and others earn little is a room with layout problems. The tables that earn little are using space that could earn more in a different size, a different place or a different use.

Rank the tables by what each seat earned over a month. The top of the list shows the room's best positions: protect them for the bookings that use them well. The bottom shows its weak spots.

For each weak table, ask why. Is it too large for the parties it gets? Is it in a draught, by the toilets, beside the service station? Is it rarely offered for booking? Each answer suggests a change: a smaller table, a different position, a screen, better lighting or a different role.

The table mix, how many twos, fours and sixes the room has, has a large effect on revenue per seat. A room of fours serving mostly couples leaves half its seats empty at the busiest times, and every empty seat lowers the figure.

Matching the mix to the party sizes that book raises revenue per seat without any change to the menu. Our guide to table mix: how many twos, fours and sixes covers how to work out the right mix.

Grouping tables into zones, such as the window, the main floor, the back room and the bar, shows which parts of the venue earn most per seat. A back room that earns half what the main floor does for each seat may be badly lit, too far from the kitchen or only used when the front is full.

Each zone with a low figure is a decision: improve it, give it a different purpose, such as private dining or groups, or reduce its seats so the space feels better and the remaining seats earn more.

Bar and counter seats usually earn differently from tables: shorter stays, smaller spends, more walk-ins and more drinks. Their revenue per seat can be high, if they turn often, or low, if they are mostly empty outside the peak.

Work out the figure for the bar separately from the dining room. A bar that earns well per seat may deserve more stools; one that earns little may work better as a waiting area or with a few small tables instead.

Terrace seats are seasonal in most places, and their figure should be read with that in mind. A terrace that is full every warm evening and closed for half the year may earn well per seat while open and little across the year.

Compare the terrace's revenue per seat for the months it is open with the indoor room's for the same months. If the terrace earns less per seat even in summer, the extra seats are spreading the same guests more thinly rather than bringing new ones.

Private rooms are often the hardest seats to judge. A room used for a large party twice a week may earn little per seat on average and a great deal on the nights it is used. A room used rarely is space the venue pays for and seldom sells.

Track the private room's revenue per seat separately, and its use: how many nights a month it is booked. If both are low, the space may earn more as part of the main room.

Capacity utilisation, the share of the room's seat time that guests used, explains much of the difference in revenue per seat between services and between tables. A room used for most of its open hours earns more per seat than one that is busy for an hour and empty for the rest.

Our guide to capacity utilisation: how full is full? covers how to measure it, why full is never 100% and what a realistic ceiling looks like.

RevPASH, revenue per available seat hour, takes revenue per seat one step further by counting the hours each seat was open as well. It shows how hard each seat worked for each hour of trading, which matters when comparing services of different lengths.

Revenue per seat is simpler and easier to explain to a landlord; RevPASH is more precise for decisions inside the venue. Our guide to RevPASH: revenue per available seat hour covers how to work it out.

Rent per seat, the annual rent divided by the seats, set against revenue per seat, shows how much of each seat's earnings goes to the landlord. It is the same idea as rent as a share of sales, seen from the seats.

When rent takes too large a share of what each seat earns, the venue struggles however busy it is. Our guide to why restaurants fail, in the numbers covers the share of sales rent can safely take.

Bring the figure to a lease renewal or a rent review, alongside covers and sales, worked out for at least the last full year. It shows the landlord how the business uses the space in terms they recognise, and it supports the case for the rent the venue can afford.

If the figure is rising, it shows a tenant growing into the space. If it is falling, explain why and what you are doing about it. A landlord who understands the business is more likely to agree terms than one who sees only the rent.

Lenders funding a refurbishment or a second site want evidence that the space earns enough to repay them. Revenue per seat, alongside the costs and the profit, shows whether the existing room works hard enough to justify more borrowing.

For a second site, use the first venue's figure carefully. A new venue in a different area, with a different room, may not reach the same figure for a long time, and a plan that assumes it will is a plan a careful lender will question.

Adding seats raises revenue per seat only if the new seats fill. A few extra tables squeezed into a room that is full on busy nights can add revenue at the peak; the same tables in a room that is rarely full just spread the guests more thinly.

Before adding seats, check how often the room is full and guests are turned away. If it rarely is, the figure will fall, and the room may feel more crowded for the same number of guests.

Sometimes revenue per seat rises when seats are removed. A room with too many tables packed close together feels cramped, slows service and puts guests off. Taking out a few tables can make the room more pleasant, speed up service and raise spend per head.

The total revenue may stay the same or even rise, while the figure per seat rises with it. It is a counterintuitive change, but one worth testing in a room that is rarely full.

The figure makes venues of different sizes comparable, which is why groups and investors use it. A small bistro and a large brasserie can be set side by side, because the figure already allows for their size.

Compare venues of similar kinds, though. A café, a fine dining room and a bar earn very differently per seat, for good reasons. Each venue is best compared with itself over time, and with venues of the same kind.

Work it out monthly, by zone and by table as well as for the room. Compare each month with the same month last year, because seasons move it a great deal.

Watch the trend more than the level. A figure that slips for several months, while covers hold, points to falling spend; one that slips while spend holds points to fewer guests. Either is worth acting on before it reaches the accounts.

The team does not need the figure, but it helps them to understand the idea: every seat in the room costs rent whether or not anyone sits in it. A server who knows that a couple at a table for six leaves four seats earning nothing is more likely to suggest a better table.

Share the zones and tables that earn least, and ask the team why they think guests avoid them. The people who work the floor often know exactly what is wrong with the corner by the door.

An invented brasserie has 90 seats and takes 1.08 million a year, which is 12,000 per seat. Its landlord proposes a large rent rise at renewal. The owner works out the figure table by table and finds the back room's twenty seats earn less than half what the main floor's do, while the window tables earn the most.

Before the renewal meeting, the owner turns the back room into a bookable space for groups and adds three tables for two by the window in place of a large round. Within six months, the brasserie's revenue per seat rises by about a tenth. The owner takes the figures to the landlord, who agrees a smaller rise than first proposed.

servearound's table analytics show revenue per seat for every table, for any dates: the revenue paid on the table's checks divided by its seats, beside its seats, seatings, covers, average party, minutes with a party, unsold minutes, utilisation and revenue per hour. The All tables row gives the room's table revenue, so the room's figure is that total divided by the seats in the list; checks not tied to a table, such as takeaway, sit outside it. The floor plan can be shaded by revenue, and results by zone show each part of the room.

The dashboard shows spend per cover against the previous period, and reservation analytics show covers by day and hour and by party size, the other half of the sum. See Analytics.

Revenue for a period divided by the number of seats the room sells.

Because they judge a business by what it earns from the space it occupies.

The seats available to guests during normal trading, counted the same way each time.

Fill more seats more often, with tables matched to parties, and raise spend per head.

No. RevPASH also counts the hours each seat was open. Revenue per seat is simpler.

Only if the room is often full and turning guests away.

Yes, in a cramped room that is rarely full, if service and spend improve.

Monthly, by table and zone as well as for the room, compared with last year.

Revenue per seat divides a venue's revenue by the seats it sells, for a night, a month or a year. It measures the room by its capacity rather than its guests, it lets landlords and lenders judge how hard the space works, and it rises with more covers per table, a higher average spend per head or both.

Work it out by table and zone to see what the layout is doing, match the table mix to the parties that come and treat bars, terraces and private rooms on their own terms. Read it with capacity utilisation and RevPASH for decisions inside the venue, bring it to every lease renewal and watch the trend as closely as the level.