Setting a rate floor: how a hotel decides the lowest price it will sell a room for

A rate floor is the lowest price a hotel will accept for a room, fixed before any discount is discussed. Build it from the costs a sold room actually triggers, add the margin the owner chooses, then record who may go below it and why. It is a limit, not a target.

A caller offers to take twelve rooms on a quiet Tuesday in February at a price that is not on any rate sheet. The reservation agent says yes, because nothing on the screen says no. That missing number is the rate floor, and without it every discount decision rests on the judgement of whoever happens to answer the phone. After reading this you will be able to build a floor from real costs, decide where it should differ by channel and season, and set the rules for going below it.

What is a hotel rate floor, and why is it not a target?#

A rate floor is the lowest price at which the hotel is willing to sell a given room on a given night, written down before the negotiation starts. It answers a different question from a target rate: the target is what you ask for, the floor is what you refuse to go under. If you are new to rate setting generally, start with how to set hotel room rates from cost, demand and the guest's alternatives.

Three numbers get confused in practice, and only one of them is a rule. The target rate is what you publish and open with. The budget is the annual average the owner plans around. The floor is the limit below which a sale needs approval.

A floor is a decision rule, not a forecast. It does not predict demand; it protects the hotel from a single bad decision made at 23:40 by someone who wants to close the day. That is why a hotel minimum rate floor policy is worth writing down even in a property with no revenue manager.

Floors drift when they live in people's heads. Two agents quoting the same room type on the same night will land on different numbers unless the number is written and visible at the moment of quoting. The floor is set per room type and per date, not once for the hotel: a sea-view double and a basement single do not share a limit.

A floor is not a pricing strategy. It does not tell you what to charge; it tells you when to stop. Setting a floor is the cheapest part of revenue management, and it is easy to skip. Write the floor down with its inputs and its date, because a floor without a stated basis cannot be defended to an owner and cannot be revised when a cost changes.

Which costs does selling one more room actually trigger?#

Only the costs that appear because that specific room is sold belong in the floor. Everything the hotel pays whether the room sells or not is covered across the year, not by the next booking.

The first block is variable cost per occupied room-night: laundry and linen, amenities, the consumables in the room, the water and electricity an occupied room uses, and the food and drink the guest actually consumes at breakfast or on half board. The second block is the cost-to-serve of the channel: the fee the agency or platform charges, the payment processing fee, and the staff time a manual booking consumes. These differ by channel, which is why the floor differs by channel.

Tax treatment sits alongside both. A stay carries tax obligations that depend on the taxpayer's status and the document issued; describe the duty in general terms and confirm the treatment, and the rate that applies to you, with your accountant.

Fixed costs are not triggered by one more room: rent, insurance, management salaries, the mortgage, the licence. Adding them to a single room-night produces a floor so high that the hotel stops selling in the months it most needs to. The cost of the room being unavailable is a real consideration but it is not a cost: it is the contribution you give up when the room could have sold to someone else at a higher price. Treat it as a separate displacement test, not as a line in the cost build.

Two costs are easy to forget. One is the commission or override paid to a third party who introduced the booking. The other is the cost of a cancellation or no-show on a rate that allows it. Both belong in the channel's cost-to-serve.

How do you add the margin the owner needs to reach a floor?#

Add the margin the owner chooses to the triggered costs, and the sum is the floor. The margin is a business decision, not a calculation, and it should be written down next to the number it produces.

Build the floor in four steps: variable cost per occupied room-night, plus cost-to-serve of that channel, plus tax treatment, plus the margin per room-night the owner wants to protect. The margin is where the owner's strategy enters. A hotel rebuilding cash flow may protect a thin margin; a hotel protecting a position may protect a fat one. Neither is wrong, but the choice must be explicit and dated.

Test the floor against a real offer before adopting it. If the offer sits below the floor, the difference is the contribution you are choosing to give up, and that number is what the approval conversation is about. Compare any below-floor offer with the break-even question: how many rooms at the lower price does it take to earn what one room at the floor earns? The answer is usually more rooms than the caller is offering, and the break-even sum before you cut a rate shows the arithmetic.

These numbers are illustrative only, not a benchmark. Keep the worked example visible to the team, because a floor that arrives as a single number with no arithmetic behind it gets treated as an opinion. Revisit the inputs, not the number: when laundry, food or a channel contract changes, rebuild the floor from the new inputs instead of nudging the old figure.

Should the floor be the same for every channel, season and length of stay?#

No. The floor should differ wherever the cost-to-serve or the displacement risk differs, and it should be simple enough that the person quoting can remember it.

Channel first. A booking that arrives through an agency carries a fee, a payment cost and often a cancellation risk that a direct booking does not. The floor for that channel is higher by exactly those costs, not by a round guess. Season next: in a week you expect to sell out, the floor should reflect what the room would otherwise earn; in a week you will not sell out, the displacement risk is close to zero and the floor can sit nearer the triggered cost.

Length of stay matters because a fixed cost per booking spreads over more nights in a five-night stay than in a one-night stay, so the per-night floor for the longer stay can be lower without giving anything away. Occupancy basis matters too: a single-occupancy night in a double room triggers less variable cost than a double-occupancy night, and the floor should follow the cost, not the room's published price.

Group and contract business is priced as a block, but each night still has a floor. Approve the block against the floors of its weakest nights, not its average. If you are wrapping the decision into a campaign, the discount campaign checklist for prices, dates and fences covers the rest of the structure.

Cap the complexity. Three to five floors per hotel, defined by channel tier and season tier, is workable. Twenty floors is the same as none, because nobody applies them correctly under pressure.

Who is allowed to go below the floor, and how is it recorded?#

Name the roles that may approve a below-floor sale, set the size of gap each may approve, and require every exception to be logged before the booking is confirmed.

Approval levels should be written as amounts, not adjectives. The reservation agent may not go below the floor; the front office manager may approve a small gap; the revenue manager or general manager approves a larger one; anything beyond that goes to the owner. The exception is approved before the booking is confirmed, never after: a retroactive log records what already happened, and it does not control anything.

Every exception carries a reason code, so the monthly review can group them: competitor undercut, group volume, last-minute unsold inventory, loyal guest, error correction, or a system price that loaded wrong. The log records the date, room type, floor, quoted price, the gap, who approved it, how many rooms, and what the approval expected to gain. Without the last field the review has nothing to test.

Give every exception an end: a number of rooms, a date, or both. An exception without a limit becomes the new floor for that account. Publish the floors where quoting happens: the rate screen, the channel manager, the agency portal and any assistant that quotes on the hotel's behalf. A floor that lives in a document nobody opens at the desk is decoration.

How do you count floor breaches, and what does the count tell you?#

Count a breach as any sold room-night below the floor with no logged exception, then split the count by cause, because each cause has a different fix.

A system cause means a rate was loaded or mapped wrongly, and the fix is a check before publishing. A process cause means someone forgot, and the fix is training or a screen that shows the floor. A strategy cause means the hotel chose to go below and did not log it, and the fix is the approval rule itself.

Measure money, not only count. Ten breaches of a small gap matter less than one breach of a large one, and the monthly figure should show both the number and the contribution given up. Watch the quiet leak as well: a channel or an agency selling below the floor without an exception usually appears as a rate that does not match what the hotel published, and it is a contract and systems question, not only a pricing one.

Review monthly with the owner, on the same page as the exceptions that were approved. The interesting number is not how many breaches happened but how many were approved with no recorded reason. Change one thing per review: if the count is high because the floor is wrong, fix the floor; if it is high because the rule is unknown, fix the communication. Doing both at once tells you nothing.

When is a rate floor wrong?#

A floor is wrong when it stops the hotel selling rooms that would have added contribution, or when it lets the hotel sell rooms it could have sold higher. Both failures are visible in the breach and exception data.

Too high in a demand collapse: if the floor sits above what the market will pay, the hotel sells nothing and covers none of its triggered costs. The floor should never be defended out of pride; it should be rebuilt from its inputs and, if needed, lowered deliberately and in writing. Too low in strong demand: a floor that never moves upward becomes the price. When the hotel expects to sell out, the floor should rise with the displacement risk, or the cheapest rooms will be gone before the demand arrives.

A floor built for last year's costs is wrong this year. Laundry, food, energy and channel contracts move, and a floor that is not rebuilt drifts below the cost it was meant to protect. A floor used to justify a discount already given is not a floor: if the number moves whenever a sale needs approving, it is a formality and the team will treat it as one.

Elasticity decides how much a floor costs you. If a small price change moves bookings a lot, a high floor is expensive; if it barely moves them, a low floor is. Test the assumption rather than asserting it, using price elasticity for hoteliers: what happens to bookings when the rate changes. A floor is a guardrail, not a forecast. It cannot tell you what demand will do; it can only stop a decision that no one would defend the next morning.

What to do next#

Write the floor down this week: the triggered cost per occupied room-night, the cost-to-serve of each channel, the margin the owner chooses to protect, and the name of the person who may approve a gap. Then log every exception for one month and read the log before you change the number.