You Can't Know the Right Price. You Can Only Find It.

For the owner or revenue manager who suspects revenue management is mostly guesswork. It is — in the way science is: structured guessing, measured honestly, repeated until the guesses get good.

We say this more often than almost anything else, usually when a hotelier asks what revenue management really is under all the dashboards and acronyms: it’s a continuous series of small experiments and honest measurements. You try a price. You watch what the market does. You keep what worked, change what didn’t, and go again. That’s the craft. Everything else — the software, the reports, the theory — exists to make that loop faster and more honest.

It sounds almost too simple, and the simplicity hides the part that trips people up. Most hotels do the first half enthusiastically and skip the second half entirely. They try things constantly — a rate bump here, a weekend promo there — and almost never go back to ask, cleanly, did that do what I thought it would? The result is motion without learning: a year of activity that leaves you no smarter about your own hotel than you were in January. This piece is about doing both halves on purpose.

Every price is a hypothesis

Start with the uncomfortable truth that makes the whole discipline necessary: you cannot know the right price in advance. Not because you’re inexperienced — because the information doesn’t exist yet. The right rate for a Tuesday in November depends on demand that hasn’t materialized, a competitor who hasn’t opened or closed out yet, weather nobody’s seen, and a dozen decisions your future guests haven’t made. No spreadsheet contains next month. The best you can do is form a good hypothesis and test it.

So look at what a published rate actually is. When you set that Tuesday at €120, you’re not stating a fact — you’re making a bet: “At €120, I believe this date fills at roughly this pace and lands around here.” The market is the experiment that settles the bet. It answers within days, in the only currency that matters: bookings, or the silence where bookings should be.

Once you see prices as hypotheses, the job changes shape. The rate isn’t the deliverable. The learning is. A price that “worked” and a price that “flopped” are worth the same to you — one data point each — provided you actually read the result. The only truly wasted rate is the one you set, watched fill or not fill, and never asked a single question about.

The half everyone skips: measurement

Here’s where hotels quietly lose the plot. Trying is easy and a little bit fun — you change a number, something happens, you feel busy. Measuring is boring, and it demands an inconvenient thing: you have to have written down what you expected before you acted, then go back and compare it to what happened.

Skip that, and every experiment collapses into a story. Bookings picked up after you dropped the rate, so “the price was too high” — except a competitor also closed out that week, and a local event went on sale, and honestly you’ll never know which one moved the needle. You didn’t run an experiment. You made a change and wrote a comforting caption for it afterward. Activity is not evidence.

And measurement has a hard prerequisite most tools quietly fail: memory. To judge the outcome of a decision, you need to remember the state of the world when you made it — the pace you were on, the position you held, what you believed would happen. Most systems, including your PMS, only ever show you today; yesterday’s view is overwritten by this morning’s. Ask your PMS what your pace looked like three weeks ago and it can’t tell you — not because it’s a bad tool, but because it was built to run the hotel, not to remember it.

The fix is the one good revenue platforms are built around: photograph the state every day and keep the album. (We use exactly that picture — a plant on the windowsill, photographed daily, so any day stays comparable to any other.) A price change you can’t measure the outcome of is just an opinion you charged money for.

Change one thing, and write down what you expected

Between “try stuff” and “run experiments” sits a small amount of discipline that costs almost nothing and changes everything.

Change one thing at a time. If you drop the rate, launch an ad, open a new channel, and loosen the minimum stay in the same week, and bookings move — congratulations, you’ve learned nothing, because you can’t attribute the result to any single cause. It’s the same reason you can’t tell which campaign brought the bookings when everything fires at once. Isolate the variable you actually want to learn about.

Write the expectation down before you act. One sentence in a decision log: “Dropping this Tuesday from €120 to €105 because pace is running 10% behind; I expect pickup to accelerate over the next ten days.” That sentence is what separates an experiment from a gamble. A gamble is a change you rationalize whichever way it turns out. An experiment has a prediction you can be wrong about — and being cleanly wrong is how you learn fastest. (It’s also the quiet failure behind so much money left on the table: decisions nobody wrote down can’t be graded, and ungraded decisions drift back to gut feel.)

And one counter-discipline that matters just as much: don’t run the loop faster than the signal. Re-pricing every date every day isn’t more experimentation — it’s noise. You change the rate before the last change had time to produce a readable result, so nothing is ever attributable; and worse, guests notice. A price that yo-yos teaches people to wait for the dip — panic drops train your own demand to book late and cheap. Disciplined experimentation includes patience: a hypothesis worth testing, a window long enough to read it, and the restraint to leave it alone while it runs.

The revenue management process is a loop, not a line

Written out, the revenue management process is four repeating steps: see what’s happening, decide what to try, do it, and measure what came back. At Peaqplus we call it Signal → Decision → Action → Outcome, and the arrow from Outcome bends straight back to Signal — because the result of this week’s experiment is the starting information for next week’s. It isn’t a line with a finish. It’s a loop you keep running.

Which means the thing people quietly dread — “I’ll never get it finished, I’ll never nail it” — isn’t a failure. It’s the actual shape of the work. Last October’s perfect rate is this October’s hypothesis, because the compset moved, your reviews moved, the events moved, and the booking window moved. There’s no solved state to arrive at. Revenue management isn’t a puzzle you complete; it’s a practice you keep.

This is, honestly, the most encouraging thing about the discipline — especially if you don’t think of yourself as a “revenue person.” You don’t need to be a statistician or a forecasting savant to do it well. You need two ordinary things: curiosity about why a date behaved the way it did, and a system that remembers, so your curiosity has something honest to chew on. The maths can be assisted. The habit of testing a price and reading the result is the part that’s genuinely yours.

Tighter loops win

Picture two hotels on the same street, both experimenting honestly. One reads its results once a month, from a spreadsheet someone rebuilds by hand the week it’s due. The other sees the outcome of last week’s rate change the morning it becomes readable. Over a season, the second hotel runs the loop many more times than the first. More loops mean more learning, and learning compounds into better prices — not because their people are smarter, but because their loop is tighter.

That’s the honest case for tooling, and it’s a narrow one. A platform doesn’t run the experiment for you and shouldn’t pretend to — the judgment about what to try, and what the result means, stays yours. What it does is take the friction out of the loop: it remembers every state so outcomes are actually measurable, it surfaces the result quickly so you’re not waiting a month to learn, and it lifts the data-assembly chore off your desk so the time goes into the thinking instead of the spreadsheet. A forecast graded against actuals every month, a rate change you can trace and defend later, a history you can rewind — these aren’t features for their own sake. They’re what let your loop turn faster than the hotel next door.

Everything on the Peaqplus platform is built for that one motion: see the signal, make the call, act, and — the part everyone else skips — measure what came back, so the next call beats the last. That’s the whole job. We just make the loop quick enough to run often.

Frequently asked questions

What is the revenue management process? At its core it’s a repeating loop rather than a fixed procedure: you read the current demand signals, decide on a pricing or inventory move, make the change, then measure the outcome against what you expected — and that outcome becomes the starting information for the next round. Formalized, the steps are often labeled Signal → Decision → Action → Outcome. The discipline isn’t in any single step; it’s in closing the loop by measuring, so each cycle teaches you something the last one didn’t.

Is revenue management just guessing? It’s structured guessing — the same way science is. You can’t know the right price in advance because the demand hasn’t happened yet, so you form a hypothesis (a rate), test it in the market, and read the result. What separates it from plain guessing is that revenue management writes down what it expected and measures whether it was right, so the guesses get steadily better over time.

How do I know if a price change actually worked? You need two things: a prediction you made before the change — what you expected pace or pickup to do — and a record of the state at the time, so you can compare. Judge the change against that expectation, not against a vague sense that things feel better. And change one variable at a time: if you moved the rate, ran an ad, and opened a channel in the same week, no honest attribution is possible.

How often should I change my rates? Often enough to respond to real demand signals, but not so often that a change has no time to produce a readable result — and not so often that guests learn to wait for the price to fall. Let a hypothesis run long enough to measure it. Constant re-pricing usually adds noise, erodes trust, and trains your own demand to book late and cheap.

Do I need to be a data expert to do this? No. You need curiosity about why dates behave the way they do, and a system that remembers what happened so you can check. The statistics can be automated; the habit of testing a price and honestly reading the result is the part that matters, and anyone who runs the hotel can learn it.

Where to go from here

If this is the mindset, the complete guide to hotel revenue management is the map — the metrics, the strategies, and where the loop fits. Two neighbors go deeper on the measurement half: Your Forecast Is Always Wrong applies the same “wrong by a known amount” honesty to forecasts, and The Question Your PMS Can’t Answer is about the memory that makes measuring possible at all. Or book a demo and ask to see a decision from three weeks ago measured against what actually happened — that one request tells you whether a tool closes the loop or just decorates it.

You’ll never find the one right price and be done. You’ll find a better price than last time, measure it, and go again. Done right, that isn’t the frustrating part of the job. It’s the whole game.