What Is RevPASH? And Why Does It Matter?

Two restaurants both take $4,320 on a Friday night. The first does it with 108 covers across a busy, evenly-paced six hours. The second does it with 108 covers crammed into ninety minutes, with the dining room three-quarters empty either side.

Daily sales cannot tell these apart. Average check cannot either. RevPASH can, and that is the entire reason it exists.

RevPASH (revenue per available seat hour) measures how much money each seat generates for each hour it is available to be sold. It treats a seat-hour the way a hotel treats a room-night: as perishable inventory that either earns or expires. This post covers the formula, how to calculate it from your own POS data, what the number tells you, and the four levers that actually move it.

The Bottom Line

• RevPASH = total revenue ÷ (available seats × hours of operation). It answers how hard your seats work, not how busy the room felt.
• The metric was developed by Sheryl Kimes at Cornell University’s School of Hotel Administration, published in the Cornell Hotel and Restaurant Administration Quarterly in 1998 as part of her work on restaurant revenue management.
• It catches what average check and daily sales both hide: the hours you paid rent and payroll for and sold almost nothing.
• There is no universal benchmark. RevPASH is only meaningful against your own history, by daypart, and against your occupancy cost per seat-hour.
• Four levers move it: meal duration, price and mix, seating configuration, and shifting demand into empty hours. Only one of them involves charging more.

What is RevPASH?

RevPASH is a capacity-based performance metric. It divides the revenue you earned by the seating capacity you had available, measured in seat-hours, so that empty capacity counts against you.

The thinking comes from hotel revenue management, where the central insight is that an unsold room-night is gone forever. Cornell professor Sheryl Kimes adapted that logic to restaurants in the late 1990s, arguing that a restaurant’s real inventory is not food but time multiplied by seats, and that a restaurant manages two variables to maximise it: price and meal duration. Her restaurant revenue management work through Cornell’s School of Hotel Administration is where the metric and the framework around it originate.

The practical consequence is that RevPASH penalises things your other reports reward. A restaurant with a famous two-hour wait at 8pm and an empty room at 6pm looks successful on a sales report and mediocre on RevPASH, because it is paying for seat-hours it never sells.

The RevPASH formula

RevPASH = Total revenue ÷ (Available seats × Hours of operation)

The denominator is seat-hours: your capacity expressed as a quantity of perishable inventory.

Three definitions to get right before you calculate anything, because inconsistency here makes the number useless over time:

• Available seats. The seats you could actually sell, not the seats you own. Exclude the dead tables you never seat: beside the restroom door, in the draught by the entrance. Bar seats count if you serve food at them.
• Hours of operation. The hours the dining room was open and available. Exclude prep and close-down.
• Revenue. Pick one basis and never change it. Food and beverage net of tax, excluding tips, is the usual choice. If you include takeout and delivery revenue, the metric breaks: those orders consume no seat-hours, so they inflate the number and hide a half-empty dining room.

That last point is the most common way RevPASH gets corrupted in practice. If a meaningful share of your revenue leaves the building, calculate dine-in RevPASH separately and track the outbound channels on their own metrics. Mixing them produces a figure that looks excellent while the room is empty.

How to calculate RevPASH from your POS data

Take a 60-seat restaurant open for a six-hour dinner service that took $4,320 from 108 covers at a $40 average check.

• Seat-hours available: 60 seats × 6 hours = 360 seat-hours
• RevPASH: $4,320 ÷ 360 = $12.00

That single figure is a start, but the service-level average is the least interesting version of the metric. The value appears when you calculate it hour by hour:

Now the same service tells a different story. The restaurant earns $20.00 per seat-hour at its peak and $4.00 in its first hour. That first hour is not a slow hour to be tolerated: it is 60 seat-hours of inventory sold at a fifth of what the business can achieve, with the same rent and most of the same payroll attached.

Two hours of the six are carrying the service. The question RevPASH poses, and that no other standard report poses, is what the 5-6pm and 10-11pm hours are for.

Pull the inputs from your POS: net sales by hour, and your own seat count. Most systems will give you hourly sales directly. Track it weekly by daypart and day of week, because a Tuesday lunch and a Saturday dinner are different businesses that happen to share a dining room.

What RevPASH shows that average check and daily sales hide

Each of the common metrics is blind to something specific:

• Daily sales tell you the total and nothing about the capacity it took to earn it. Sales can hold steady while you add hours and staff, which is a declining business reported as a flat one.
• Average check tells you what each guest spent but nothing about how many guests you could have served. Raising average check while covers fall can leave sales flat and RevPASH falling.
• Table turn time tells you how fast tables cycle but nothing about what they earn. Fast turns at a low check can produce less than slower turns at a higher one.
• Seat occupancy tells you how full the room was but nothing about the money. A full room of guests nursing coffee is high occupancy and poor RevPASH.

RevPASH combines capacity, time and spend into one figure, which is why it is the metric that catches a specific and common failure mode: the restaurant whose sales look stable while the edges of service quietly empty out. In the warning-sign list for why restaurants fail, falling RevPASH against flat sales is one of the more reliable early indicators, because it shows the problem is structural rather than a slow week.

What is a good RevPASH?

There is no industry benchmark worth quoting, and anyone who offers one is overlooking that RevPASH varies enormously with format, price point, market and trading hours. A fine dining room at $120 a head and a fast-casual counter at $14 cannot be compared on this metric.

Use three comparisons instead:

• Against your own history. Same daypart, same day of week, period over period. This is the comparison that matters, and the only one that is unambiguous.
• Against your occupancy cost per seat-hour. Take monthly total occupancy cost, divide by seats, divide by trading hours in the month. That gives the rent cost of a single seat-hour, and your RevPASH has to clear it by a wide margin: enough to cover food, labour and everything else. This is the most useful absolute reference point available, and almost nobody calculates it.
• Against your own peak. The gap between your best hour and your average hour is your realistic improvement headroom, as the worked example above shows.

The capacity side of the calculation is also worth sanity-checking, because the denominator is a function of the room itself. If a chunk of your seats are tables guests refuse, your real capacity is lower than your seat count and your RevPASH is being measured against inventory you cannot sell. That is a layout problem rather than a sales problem, and the restaurant floor plan is where the dead zones and the seat mix either work or do not.

The four levers that move RevPASH

Only one of them is price, which is the point.

1. Meal duration

Reducing average meal duration raises RevPASH directly, because it increases the number of covers each seat-hour can hold. The levers are operational, not confrontational: faster ordering, kitchen pacing, prompt bill delivery and payment at the table, and removing the dead time between the last plate and the bill. Kimes’s framework treats duration as one of the two primary variables in restaurant revenue management for exactly this reason.

The caution is obvious. Pushing duration below what the concept supports damages the experience, and in a fine dining room duration is part of what the guest is paying for.

2. Price and menu mix

Raising prices lifts RevPASH if covers hold. Shifting mix toward higher-margin items lifts it without raising a single price, which is usually the better first move: menu design, server recommendations and the position of high-margin dishes all do this work. Re-costing the menu against current invoices matters here too, since a dish priced on last year’s costs is dragging the number down invisibly.

3. Seating configuration

Match the table mix to actual party sizes. A dining room of four-tops seating parties of two wastes half its capacity, and the waste is permanent. Two-tops that combine into fours, a well-used bar or counter, and a sensible number of large tables raise effective capacity without adding square footage.

4. Shifting demand into the empty hours

This is where the biggest gains usually sit, because the 5-6pm hour in the example above has more headroom than the 8-9pm hour will ever have. Early-evening pricing, a limited pre-theatre or early menu, happy hour food, reservation incentives at off-peak times and taking bookings for the shoulder hours all convert seat-hours you are already paying for into revenue.

Tracking RevPASH hourly is what makes this possible to act on, because it identifies which hours are worth the promotional effort and, just as usefully, whether an hour is so weak that closing during it would be more profitable than staffing it.

Where RevPASH does not help

It is a dine-in capacity metric, and it has real limits:

• It ignores cost entirely. RevPASH is a revenue metric. A restaurant can raise it while destroying margin through discounting or an unprofitable mix. Manage it alongside prime cost, never instead of it. On the cost side the equivalent capacity question is how fast stock moves, which is what the inventory turnover ratio measures; the two together tell you whether both your seats and your stock are working.
• It says nothing about purchasing. Seat-hour performance is downstream of what you bought and on what terms. If supplier minimums force you to over-order to hit a threshold, that cost shows up in margin rather than in RevPASH, and understanding minimum order quantity and why it matters is part of keeping the two from working against each other.
• It does not handle off-premise revenue. Takeout, delivery, catering and wholesale consume no seat-hours. Keep them out of the calculation and measure them separately.
• It penalises concepts where duration is the product. A tasting-menu restaurant will always show a lower RevPASH than a bistro, and that is not a problem to fix.
• It needs clean, consistent inputs. Changing how you count seats or hours mid-year destroys comparability, which is most of the metric’s value.

Used properly, it belongs in a weekly set alongside prime cost, food and labour percentages, covers and cash position: the numbers worth reviewing from the first week of trading, which is also the point at which the revenue assumptions in a restaurant business plan finally meet reality and can be checked against it.

In fact the first opportunity to calculate it comes earlier than most owners realise. A soft opening at reduced capacity produces real covers against real hours, which is enough for a first honest read on seat productivity before the grand opening; far more useful than the impression that the room felt busy. And for anyone operating or planning more than one site, RevPASH is the cleanest way to compare locations of different sizes, which is why it tends to appear in the reporting pack that any serious attempt to franchise a restaurant has to put in front of prospective franchisees.

If you have not opened yet, the capacity assumptions that determine your eventual RevPASH (seat count, trading hours, format and check average) are all being set during planning, and how to start a restaurant covers where those decisions fall in the sequence.

Frequently asked questions

What is the RevPASH formula?

RevPASH = total revenue ÷ (available seats × hours of operation). The denominator is seat-hours, so a 60-seat restaurant open six hours has 360 seat-hours of capacity available to sell.

What does RevPASH stand for?

Revenue per available seat hour. It is the restaurant equivalent of RevPAR (revenue per available room) in hotels, and it was adapted from hotel revenue management by Sheryl Kimes at Cornell University.

Should delivery and takeout revenue be included in RevPASH?

No. Off-premise orders use no seat-hours, so including them inflates the figure and masks an underperforming dining room. Calculate dine-in RevPASH on dine-in revenue only, and track off-premise channels with their own metrics.

How often should RevPASH be calculated?

Weekly, broken down by daypart and day of week, with hourly detail for at least your main service. A single whole-restaurant monthly figure averages away the information that makes the metric useful.

Is a higher RevPASH always better?

Not on its own. RevPASH measures revenue against capacity and says nothing about cost, so it can rise while profitability falls: through heavy discounting that fills seats, or a mix shift toward high-price, low-margin items. Read it alongside prime cost.