Revenue management for hotels: the decisions behind the price, not the software

The pattern repeats on the hotel accounts we run with almost calendar regularity: a pace report goes soft for a specific weekend, someone drops the rate to get ahead of it, and in the same week the marketing budget is running a paid campaign built specifically to fill those same nights at full rate. Neither decision is wrong in isolation. Nobody checked what the other was doing, and the guest who found the discounted rate on the OTA that afternoon just learned something about when, and where, to book this hotel.

TL;DR: Revenue management is the set of decisions that determine what a hotel charges for a room on a given night: the price itself, the restrictions attached to it (minimum length of stay, refundable versus non-refundable, which rate plans are open), and the segment mix you're willing to accept at that price. A revenue management system (RMS, the software category built to apply those decisions across more dates and rate plans than a person can track by hand) is a tool for carrying the decisions out; the decisions still have to come from somewhere. A hotel can practise good revenue management with a spreadsheet and a calendar, and plenty of bad revenue management happens inside expensive software. This article covers what the job actually consists of, why revenue decisions and marketing spend routinely contradict each other inside the same hotel, the specific pattern that costs the most (the last-minute discount released through the OTA), what it looks like when the two functions actually talk to each other, and when a system is the wrong answer to a problem that discipline would fix for free.

This is written for the owner or general manager of an independent hotel running somewhere between 50 and 200 rooms, most of whom get pitched a system before anyone has checked whether the underlying decisions are even being made consistently. It's a demystifier, not a software buying guide.

What revenue management actually is, and what it is not

The Hospitality Sales and Marketing Association International defines revenue management as the practice hotels, airlines and car rental companies use to control the supply and price of their inventory to achieve maximum revenue or profit. Nothing in that definition mentions software. It's a set of decisions applied to perishable inventory: a room that doesn't sell tonight is gone forever, so the price and the conditions attached to it need to reflect demand on that specific night, not a flat number someone chose in January and never revisited.

The confusion is understandable. Most of what gets marketed as revenue management is actually the tool: a dashboard, a set of pricing recommendations, an automated rule that nudges rates up or down against pickup. That sits downstream of the job.

The three decisions that make up the job

Strip away the tooling and revenue management is three decisions, repeated for every date on the calendar.

Price by demand. What a Saturday in August costs is not what a Tuesday in February costs, and what this Saturday costs should reflect how this Saturday is actually pacing against comparable weekends, not what last year's Saturday happened to sell for.

Restrictions. The price is only half the rate. Minimum stay, cancellation terms, which channels and rate plans can access it: these move independently of price and often do more to shape who books than the number itself.

Segment mix. How many rooms you're willing to sell to a lower-yielding segment, a group rate or a heavily discounted rate plan, before you close it off and hold the remaining inventory for segments that pay more closer to the date, is a decision made in advance. It should not be a reflex triggered when occupancy looks thin three days out.

None of this is the same conversation as RevPAR, which measures the result of these decisions after the fact. Revenue management is what you do before the night happens; RevPAR is how you find out whether it worked.

Where a system fits, and where it does not

A revenue management system, in HSMAI's own glossary definition, is the computer system a hotel uses as a tool to help maximise revenue, typically holding availability, room type and stay pattern data so a person or an algorithm can act on it. That's a genuinely useful function once a property has enough rate plans, room types and date-level complexity that no one can hold it all in their head. It's good at the mechanical part: crunching pace and pickup across hundreds of dates faster than a spreadsheet ever will.

What it can't do is decide your segment strategy, set your floor, or tell marketing what's happening to price this week. Someone still owns the strategy the system executes. Buying the tool before that person, or that discipline, exists just automates the guessing.

Revenue management and marketing usually run as strangers

Here's the pattern we see on real hotel accounts, not on an org chart. Revenue management and marketing rarely share a calendar, and almost never share a conversation before either one acts. The revenue side looks at pace and pickup and reacts to what's booked. The marketing side looks at cost per click and conversion and reacts to what's converting. Both are optimising a real, correct thing, from a different half of the same date.

The result is exactly the scenario this piece opened with: a rate gets cut because a specific weekend looks soft, in the same week a paid campaign is actively spending to bring demand to that same weekend at the rate card price. Whoever's measuring the ad performance sees the campaign apparently underperforming, because it's now competing against a cheaper version of the same room on the same dates. Nobody did anything malicious, and neither was wrong given what each could see. A shared calendar would have prevented it.

The failure mode: the last-minute panic discount

The costliest version of this shows up as a single, recognisable pattern: the last-minute discount, released through the OTA because that's the fastest extranet to touch, landing below whatever the direct site is still charging for the same room.

Here's the arithmetic, clearly illustrative rather than a real property's numbers. Say a room's cost floor, the point below which a sold room loses money once cleaning, breakfast, laundry and commission are accounted for, sits at £58. The rate card says £95 for that Saturday. Three days out, pace looks soft, so the direct rate drops to £79 and, because it was quicker to action, the OTA rate drops further, to £71. On paper both numbers still clear the floor comfortably, so nobody flags it as a problem. What actually happened is that the exact segment marketing had been paying to bring to the direct site just found the same room £8 cheaper somewhere else, and learned that the direct rate card is negotiable if you wait.

Working out where your own floor sits before a rate ever moves is a rate-setting job in its own right, covered in a separate piece later in this series. Without a floor, a discount is a guess dressed up as a decision.

Why that discount trains guests to book late and book elsewhere

A discount released through one channel rarely stays contained to that channel, because guests genuinely shop across more than one before they book. Cornell's hospitality research on the so-called billboard effect looked at the opposite direction of this exact behaviour: it found that each OTA booking was associated with three to nine additional bookings on the hotel's own website, evidence that a meaningful share of guests check more than one channel before deciding where to actually book. The same channel-hopping cuts against a hotel the moment the cheaper price sits on the OTA instead of the direct site.

Once a guest finds that cheaper price, two habits form at once. They learn to wait, because waiting produced a better rate last time. And they learn to book on the OTA, because that's where the better rate actually was. Neither habit is dramatic on its own booking. Both are expensive across a season, because they push demand later, which weakens your ability to yield up genuinely strong dates, and they push it onto the channel that costs the most to serve.

What it looks like when revenue and marketing actually talk

The fix isn't a bigger tool. It's a shared calendar and a shared decision. On the hotel accounts we run, three things are true where this works. One person, named, owns the call to discount a date, and that call gets made in advance against a floor, not reactively against a pace report at four in the afternoon. Marketing knows about a rate change before it goes live, not after budget has already been spent defending a price that no longer exists. And when a date genuinely needs a lower rate to move, the direct site gets it first, or at worst simultaneously, never last.

No software is required for any of it. The revenue decision and the marketing spend have to sit in the same conversation, on the same dates, before either one moves.

When formal revenue management is the wrong tool right now

Three situations mean a system, or even a formal revenue management process, should wait.

Demand is genuinely stable and the property is small. A 20-room property with a predictable season and few room types doesn't need dynamic, date-by-date pricing sophistication. It needs a seasonal rate card, built once a year against last year's actual pace, and the discipline to hold it rather than discount on nerves. Buying a system to automate a decision that barely changes week to week solves a problem that doesn't exist yet.

Nobody owns the decision today. A system layered on top of an unowned decision just produces recommendations nobody applies consistently, which is worse than no system, because it looks like the problem is solved when it isn't.

The rate card has never been checked against what a guest actually sees. If the OTA search results page already shows a different price than the one on the rate card, no amount of pricing sophistication behind the scenes fixes the thing the guest is actually reacting to.

Where to start if nobody owns this job

Before anyone buys anything, put one name against the decision to discount, in writing, and put a floor under every date that name has to clear before a rate goes out on any channel. That single piece of discipline, a named owner and a floor that doesn't move on a bad Tuesday, does more for a 50 to 200-room hotel than a system layered onto a decision nobody is actually making consistently.

If you want a second opinion on where your own rate and channel mix is actually leaking money, that's exactly what our free audit looks at.