RevPASH: revenue per available seat hour

Two restaurants each take the same amount on a Saturday night. One has forty seats and is open for four hours; the other has eighty seats and is open for six. On the sales report they look identical. In truth the first is working its room twice as hard as the second, and the second has a great deal of capacity it is not selling. Revenue alone cannot show the difference.

RevPASH, revenue per available seat hour, can. It divides the money a service earned by the seat hours the room had to offer: every seat, for every hour it was open. The result shows how hard the room is working, and it rises when guests spend more, when more of the seats are used or both.

This guide covers what RevPASH measures, how to work it out from figures most venues already have, how to read it hour by hour and how to raise it without rushing a single guest. Every venue and figure in it is an example.

RevPASH measures how much revenue each seat produced for each hour it was available to sell. It joins the two things a dining room sells, food and drink and time at a table, into one figure, so a venue can see whether it is earning well from the capacity it has.

It answers a different question from the usual measures. Covers say how many guests came. Spend per head says how much each spent. RevPASH says how much the room as a whole earned for the seats and hours it had, which is the question that matters when the room is the thing you are paying for.

The formula is simple: revenue for a period, divided by the number of seats, multiplied by the hours open in that period. A room with sixty seats open for five hours has three hundred seat hours. If it takes 4,500 in that time, its RevPASH is fifteen per seat hour.

Use the same revenue each time, usually food and drink sales before any service charge or tax, and the same hours, usually from the time the first table can be seated to the time the last can be. What matters most is that the figure is worked out the same way every week, so the changes in it are real.

Hotels have long measured revenue per available room, known as RevPAR: room revenue divided by the rooms available. It replaced simpler measures because a hotel that filled every room at a low rate, or charged a high rate for a half-empty building, could look successful on one measure while earning poorly on the other.

Researchers in hospitality adapted the idea for restaurants, where the unit of capacity is not a room for a night but a seat for an hour. A restaurant's seats perish just as a hotel's rooms do: an hour of an empty chair can never be sold again.

Covers are the figure most venues watch, and they mislead in two ways. A night with many covers may have been earned at a low spend, with discounted menus or guests ordering little. And covers say nothing about the time each guest spent, so a busy night with long, slow tables can serve more guests than a quiet one and still waste much of the room.

RevPASH counts both spend and time. A night that filled the room at a good spend scores well; a night that filled it at a poor spend, or left it half empty at a good one, scores worse, which is how it should be.

Spend per head rewards a venue for guests who spend well, regardless of how many came or how long they stayed. A room where four guests spent a great deal each, while fifty seats sat empty, has a fine spend per head and a poor night.

Raising spend per head is worth doing, but not at the cost of filling the room. RevPASH keeps both in view: a higher spend helps only if the seats are used, and a fuller room helps only if guests still spend.

The bottom of the formula, the seat hours available, needs care. Count the seats the room can actually sell during that service, not every chair in the building. Seats in a private room that was not open, a terrace closed for the weather or a table taken out of use do not count that night.

Hours need the same care. A dining room that opens at six but seats nobody before half past six in practice has fewer hours than its door suggests, while one that seats guests until ten and lets them stay until midnight may have more. Choose a rule that fits the venue and keep to it.

Count the hours in which a guest could be seated and served, from the first possible booking to the last. Include the quieter early and late hours, because they are part of the capacity the room offers, and leaving them out hides the very hours where RevPASH is lowest and most can be gained.

If the venue serves lunch and dinner with a break between, count each service separately and leave out the closed hours in between. Mixing them into one daily figure blurs two services with very different demand into a single number.

An invented bistro has fifty seats. Dinner runs from six until eleven, five hours, so each night offers 250 seat hours. On one Thursday, dinner takes 3,000, giving a RevPASH of twelve. On the Saturday, dinner takes 5,500, a RevPASH of twenty-two.

Spend per head is almost the same on both nights. The difference is use: on Saturday, nearly every seat was used for most of the evening, while on Thursday much of the room sat empty after nine. That is where the bistro's opportunity lies, and the sales report alone did not show it.

The most useful view breaks RevPASH down hour by hour. Divide the revenue taken in each hour by the seats available in that hour. The pattern that results is usually steep: very high in the peak hour, much lower in the early and late hours.

The hour-by-hour view shows which hours to protect and which to fill. An hour with a high figure is earning well and should not be discounted; an hour with a low figure is where invitations, early menus and set suppers can add revenue from seats that would otherwise earn nothing.

Day by day, the figure shows the shape of the week. Most venues find the weekend nights several times stronger than the early-week ones, and a lunch that is far weaker than the dinner on the same day.

Use the daily view to decide where effort goes. A small rise on the quietest nights can be worth more over a year than squeezing the busiest ones, because there is far more unused capacity in the quiet nights to work with.

RevPASH can be split into two parts: what guests spent for each hour a seat was used, and the share of the available seat hours that were used. Multiply the two and you get RevPASH again. A low figure is always one, or both, of those halves.

Splitting it this way tells you what to work on. If spend per seat hour used is healthy but the share of seat hours used is low, the room needs more guests at quieter times. If the room is well used but spend is low, the menu, the prices or the service need attention.

The second half, the share of seat hours used, is capacity utilisation. It shows how full the room really was across the whole service, not just at the busiest moment. A room that feels full at eight may be well under two thirds used across the evening, once the early and late hours are counted.

Utilisation is the half most venues can move most easily, through booking rules, table mix and quiet night trade. Our guide to capacity utilisation: how full is full? covers how to measure it and what a realistic ceiling looks like.

Revenue per seat, the revenue a table earned divided by the seats it has, shows which parts of the room earn well for their size. A table for two by the window may earn more per seat than a table for six in a corner that is often taken by four.

Look at it over a month rather than a night, because single nights vary too much. Tables that earn well per seat are worth protecting at the peak. Tables that earn little per seat may be in the wrong place, the wrong size or offered to the wrong parties.

Yield per table takes revenue per seat a step further, by counting the time each table was available as well as its seats. It is RevPASH worked out for a single table: its revenue, divided by its seats multiplied by the hours it was open for service.

It shows the tables that pull their weight and the ones that do not. A large table that seats couples most nights, or a small one squeezed by the door that guests turn down, will have a low yield per table, and both are worth looking at again.

The table turnover rate, how many seatings each table hosts in a service, is one of the main drivers of RevPASH. More seatings in the same hours, at the same spend, means more revenue from the same seats.

It should rise through a better flow of the meal, not by hurrying anyone: prompt greetings, orders taken without delay, the bill brought when it is wanted and tables reset quickly. Our guide to table turnover rate: how to raise it politely covers how.

Every couple seated at a table for four leaves two seats earning nothing for the length of their meal. On a full night, those empty seats are a large share of the room, and they pull RevPASH down even though every table looks taken.

Matching parties to tables of the right size is one of the quickest ways to raise the figure. Hold fours for fours where the diary allows, combine twos when a larger party books and look at whether the room's mix of tables matches the party sizes that actually come.

At the peak, the room is already busy, so the gains come from using it more precisely. Booking lengths that match real turn times, parties seated at tables of the right size, a quick reset between seatings and alternatives offered when the busiest time is full all add revenue from seats that would otherwise sit idle between bookings.

Avoid discounting the peak. Guests who would come anyway need no reason, and every discount at the peak lowers spend on capacity that would have sold at full price.

In the quiet hours, the gains come from filling seats. An early menu, a set supper on a weeknight, an invitation to regulars, a tasting or an event each give guests a reason to come when the room would otherwise be empty.

These hours have the most capacity to gain, and every guest in them adds revenue from seats and staff already paid for. Our guide to revenue management for a dining room sets the quiet hours alongside the peak in a wider plan.

RevPASH measures revenue, not profit. A night with a high figure built on heavy discounting, or on dishes with thin margins, may leave less behind than a quieter night with a better mix. It also says nothing about how guests felt about their evening.

Read it alongside profit per guest and guest feedback. Our guide to profit per cover beats revenue per cover covers the profit side, and a rising RevPASH with falling feedback is a warning that the room is being pushed too hard.

RevPASH lets venues of different sizes be compared fairly, which is why groups find it useful. A small venue and a large one can be set side by side, because the figure is already worked out per seat and per hour.

Compare like with like, though. A neighbourhood bistro and a city centre site serve different guests at different prices, and a lower figure at one may still be a healthy figure for its trade. The most useful comparison is usually each venue against itself over time.

Compare each month's figure with the same month last year, rather than with the month before, because seasons move demand a great deal. A summer month and a winter one are rarely comparable.

If prices have risen since last year, part of any rise in RevPASH is simply the higher prices. Look at utilisation alongside it to see whether the room is also being used better, or only charging more for the same use.

Bar and counter seats are capacity too, and they often earn differently from tables: shorter stays, smaller spends and more walk-ins. Counting them with the dining room can hide how each part of the room is doing.

Work out the figure for the bar and the dining room separately if they serve different trade. A bar with a high figure late in the evening and a dining room with a high figure at eight may each need different booking rules and different ways to fill their quiet hours.

Seats that are only open in good weather, such as a terrace or a garden, should be counted only when they are open. Counting them every night makes the winter figure look worse than it is and hides how well the indoor room is working.

When the terrace opens, the seat hours available rise sharply. If revenue does not rise with them, the extra seats are spreading the same guests more thinly rather than adding new ones, and that is worth knowing before adding more outdoor tables.

A target works best as a rise on last year's figure for the same month and service, rather than a figure borrowed from another venue. A small, steady rise, from better table use in the peak and more guests in the quiet hours, adds up over a year.

Set separate targets for the peak and the quiet hours, because they move for different reasons. The peak target is about precision; the quiet-hours target is about giving guests a reason to come.

The team does not need the formula, but it helps to share the idea: an empty seat at eight on a Saturday is revenue lost for good, a couple at a table for four costs two seats, and a slow reset delays the next guests.

Share the hour-by-hour view with the floor managers, so they can see where the room is strongest and where it is quiet. A team that understands why the booking rules exist applies them with more care, and with more warmth towards the guests they seat.

An invented brasserie with seventy seats works out RevPASH by hour for its dinner service. The figure peaks between eight and nine and falls by more than half before seven and after half past nine. Couples often sit at fours at the peak, and the hour before seven is almost empty most weeknights.

The manager sets booking lengths from the brasserie's real turn times, holds fours for larger parties at the peak and starts an early menu from six on weeknights. Over three months, the peak hour earns slightly more, the hour before seven fills steadily and RevPASH for the whole dinner service rises, while spend per head barely moves.

servearound does not show a RevPASH figure, but its table analytics hold the parts. Every table is listed with its seats, revenue, booked minutes (the time it had a party), unsold service minutes, utilisation, revenue per hour and revenue per seat, so a table's figure for the period is its revenue divided by its seats multiplied by its booked and unsold minutes added together, in hours. The page also shows revenue per seated hour across the room, results by zone, the gap between seatings, a floor plan shaded by revenue or utilisation and a heatmap of when the room fills.

Utilisation there is the time each table had a party, from its stays where tables are seated in servearound and its bookings' lengths where they are not, against the service time the venue's shifts make available, capped at 100%, so it works as capacity utilisation rather than a count of guests in seats. Payment analytics show money collected by hour, and reservation analytics show covers by day and hour, which together show where the quiet hours are. See Analytics and the reservations module.

Revenue per available seat hour: revenue divided by the seats multiplied by the hours the room was open.

Take a service's revenue, then divide it by the number of seats multiplied by the hours that service was open.

Yes, if drinks are sold at the tables. Use the same revenue each time, before service charges and tax.

The hours in which a guest could be seated and served, for each service separately.

Usually, but check it against profit and feedback. A figure raised by rushing or discounting will not last.

Spend per head ignores empty seats and time. RevPASH counts both.

Use the peak more precisely and fill the quiet hours, rather than raising prices or hurrying guests.

Monthly, by service and hour, compared with the same month last year.

RevPASH, revenue per available seat hour, divides a service's revenue by its seats multiplied by its hours, showing how hard the room is working. It combines spend and capacity utilisation in one figure, and it rises when guests spend more, when more of the room is used or both.

Work it out by service and by hour, compare each month with last year and split it into spend and use to see what to fix. Raise it with tables matched to parties, booking lengths that fit real turn times, a better table turnover rate without hurrying anyone and guests in the quiet hours. Watch revenue per seat and yield per table to see which tables earn their place.