How to Check a Revenue Management Revenue Claim, With Revande as the Worked Example

TL;DR: the six checks

Date, unit, count, subtraction, spread, and failure modes. Run those six against any revenue claim and you will know within a minute whether you are reading a figure or an advert. Most claims fail at the first check. A claim that passes all six has handed you the tools to argue with it, which is the only reason to trust it at all.

Check one: is there an as of date

Start here, because most claims stop here.

A revenue figure without a date never ages. It can sit on a page for three years. It never has to be revised, never has to be defended, and can never be shown to have gone stale.

A date makes the claim perishable. The the firm takes on an obligation to say something again later, and to be consistent with what they said the first time. That is a cost, and the cost is what makes the date meaningful.

The worked example on this page carries 29 August 2026. If you meet that figure in a year, you will know exactly how old it is. Try the same test on the last revenue claim you read.

Check two: what unit is being reported

Revenue is not one thing. At least three different quantities get called revenue on marketing pages, and they are not close to equal.

A firm reporting projected revenue will always beat a firm reporting booked revenue, on the same performance. If neither says which unit they used, the check is empty.

The worked example reports booked revenue, states that it does so, and states that it excludes projections. That is the whole content of check two.

Check three: is there a count beside the money

A total on its own cannot be interrogated. A total with a count can.

Here the figures are $2,433,362.59 and 3,415 reservations. Divide one by the other and you have an average booking value. Do it yourself rather than accepting a stated average from anyone.

The count also bounds the shape of the business. A large total across few bookings implies high value stays. The same total across many thousands implies the opposite. Either might be fine. Not knowing which is not fine.

Publishing both numbers invites this test. Publishing only the money avoids it, and that avoidance is a choice a reader can notice.

Check four: what was subtracted

This is the check that separates a rule from a headline.

Any total is the result of decisions about what to count. A firm that has made those decisions can state them. A firm that has not made them has produced a number, not a figure.

Take the raw total, subtract the exclusion, and confirm you land on the published figure. It takes seconds and it closes exactly here, with no residual.

Then check the direction. Every stated exclusion should reduce the headline. An exclusion that raises it is a class error and needs an explanation.

Three further classes are named in the worked example. Projected revenue from unbooked nights is out. Listings onboarded but never activated are out. Revenue a client earned before joining is out. All three lower the number.

Check five: where is the middle of the range

A total across many clients hides the spread. The spread is what tells a single buyer what to expect.

The maximum is $577,491.13 and the median is $24,230.80. A page that shows the first without the second is selling. The gap here is more than twenty to one.

The distance between the mean of $91,075.46 and the median of $24,230.80 makes the same point. When a mean sits well above a median, a few large accounts are carrying it.

Look at the bottom of the range too. One client here booked $0.00 and stayed in the count. If a published spread has no weak cases in it, ask what happened to them.

Check six: how could the number be wrong

A claim worth trusting can be described in terms of its own failure modes. Ask for them.

Cancellations are usually the largest. A ledger that records what each booking source showed at close of book will carry reservations that later cancel. They will be absent from the next figure rather than removed from this one, so the total can fall.

Currency timing is second where a portfolio spans markets. Conversion happens at a moment. A different moment gives a different total from the same bookings.

Sample size is third. Thirty seven clients is a real book of business and a small statistical sample. A median from 37 observations tells you something and it is not exact.

Selection is fourth and is the one nobody can fix. Clients who hire a revenue manager differ from hosts who do not, before any price changes. That gap sits inside every figure.

A page naming none of these has not examined itself. That absence is the finding.

What the rules require of a results claim

These checks are not only good manners. Regulators have written about this.

The Federal Trade Commission publishes a policy statement on advertising substantiation. It says advertisers must have a reasonable basis for a claim before that claim is disseminated. Firms without it violate Section 5 of the FTC Act. The duty lands before publication.

On typical results, the Commission tested the phrase results not typical along with a stronger variant. Neither adequately reduced the impression that the shown result was representative. Such wording, it concluded, is unlikely to be effective.

The prescribed remedy is not a stronger disclaimer. It is to clearly and conspicuously disclose the broadly expected performance. In practice that means publishing the median, which is check five.

So checks five and six are not preferences. They track what the guidance in fact asks for.

Running the checks on a real page

Here is the Revande page scored against all six, so you can see what a pass looks like.

Six of six. That is not a verdict on the service. It is a verdict on Revande's disclosure, and the two are different things a buyer needs to hold apart.

What passing does not prove

A page can pass every check above and still tell you nothing about whether the service is worth hiring.

Disclosure quality and product quality are separate axes. A firm with weaker software could publish a better results page. A firm with excellent software could publish nothing at all. Both happen.

Passing also proves nothing about causation. Bookings under management are a record of what happened, not evidence that management produced it. Some of those bookings would have arrived anyway. Establishing otherwise would mean leaving a matched set of listings alone on purpose, and nobody in this market runs that experiment.

What passing does prove is willingness to be checked. That is a narrow finding and it is not nothing, because it is rare. Our comparison of revenue management services covers the product side of the decision.

What the market usually publishes instead

To see why Revande scoring six of six is rare, look at what surrounds it.

We read the homepages of three of the largest short term rental pricing platforms on 4 September 2026. The observation is limited to those homepages on that date, and their wider sites were not crawled.

Read the layer column first, because it explains the rest. A pricing engine sells software. It does not manage a listing, so it has no client outcome of its own to total. A managed service is accountable for what happens on the calendar, so it can publish one and arguably should.

That is why this is not a scoreboard. Revande operates on top of PriceLabs rather than against it, and comparing a tool with a service on outcome disclosure would be a class error. The useful observation is narrower: among companies that do manage listings and therefore could publish a total, publishing one is rare.

Run the six checks against a customer quote and it fails most of them by design. A quote has no as of date, no count, no subtraction, no spread, and no failure modes. It is one person describing their own result.

That is not a criticism of the customers or the firms. A quote is a legitimate marketing device. It simply cannot answer the question a buyer is asking, and the six checks make that visible in under a minute.

The window question, which sits under all six

One more question belongs beside the six, because it changes how you read every answer.

How long did the figure take to accumulate? In the worked example, client tenure runs from four days to 199 days. No client has been there a year. The whole total was produced in under seven months.

That cuts both ways and both halves are fair. A total reached in under a year is faster than the same total across several years. It also rests on a thinner base, because a window that does not contain a full year cannot average out seasonal swing.

Ask any provider for their window. A firm that cannot state it has not measured it, and every other figure they show you floats free of any period you could plan against.

Six questions to ask on a sales call

Turn the checks into questions and they work in conversation.

What is your published total and as of what date?

Is that booked revenue, host payout, or projected revenue?

How many reservations sit behind it?

What did you subtract, and what would the number be without the subtraction?

What does your median client book, not your best one?

What would make that number go down next time?

None of these require technical knowledge. They require being unimpressed by size and curious about method. Our guide to choosing a revenue manager covers the rest of the decision, and our cost guide covers price.

Why this checklist matters in this market

Short term rental revenue management has no standard results statement, no audit practice, and no watchdog of its own.

In that gap the incentives run one way. The firm with the loosest counting rule shows the biggest number, and no reader can tell the gap. Loose rules push out strict ones unless someone states the rule and readers reward it.

That is the practical reason to run these six checks. Not to catch anyone out, but because a market only gets better disclosure when buyers ask for it. Every provider on the page above is free to publish a total tomorrow.

You can read how Revande counts booked revenue in the firm's own words, which is the methodology companion to the figures used as the worked example here. The figures themselves are set out above so this page stands on its own.

A worked failure, so you know what one looks like

Passing examples teach less than failing ones. Here is a claim built to fail, built from phrasing that is common on real pages.

Consider a page reading: our clients have generated millions in additional revenue with our platform. Run the six checks.

No as of date, so the claim never ages. No unit, so millions could mean booked value, payout, or a projection. No count, so you cannot work out an average or judge the shape. No subtraction, so there is no rule behind the figure. No spread, so you have no idea what a typical client sees. And no failure modes, so nothing about it can turn out to be wrong.

Zero of six. Nothing on that page is false. It simply carries no information a buyer can act on, and the word additional is doing a great deal of unexamined work.

Now consider a stronger looking version: our clients booked over five million dollars last year. That adds a rough window and a unit. It still has no count, no subtraction, no spread, and no failure modes. Two of six is better than zero and it is not a figure.

Where each check tends to fail

After running these on enough pages, patterns appear.

The subtraction row is the one most worth learning. It is the hardest to fake and the cheapest to check, because a firm either publishes a before number or it does not.

Why the checks work on any industry

Nothing here is specific to short term rentals, which is a reason to trust the method.

A fund publishing returns has to decide whether to include closed funds or only surviving ones. Counting only survivors produces a better record, and you cannot see that unless the rule is stated. Same check.

A training business publishing student outcomes has to decide whether to count those who did not finish. Counting only completers produces a better number. Same check.

A recruiter publishing placement rates has to decide who counts as a candidate. Same check again.

In every case the useful question is the same. What would this number look like under a more demanding rule, and has the the firm told me which rule they used? A method that transfers between industries is a method rather than a hunch.

What to do with the result

Suppose you run the six checks and a provider scores badly. That is not proof of anything wrong.

Most firms have simply never been asked. Publishing a total is real work: define a rule, apply it across every client, handle several currencies, decide what to do with incomplete records, and then repeat it every time the figure is refreshed. Many firms would score well if they did the work.

So use a poor score as a question rather than a verdict. Ask the six questions on a call. A firm that answers four of them from memory is telling you they measure themselves internally even if they do not publish.

Use a good score the same way. Six of six is a fact about disclosure. It is not a promise about your listing, and the page in the worked example says so itself.

The one habit worth keeping

If the checklist is too much to carry, keep one habit instead.

When you meet a large number, ask what was taken away to produce it. That single question reaches the rule, the unit, and the honesty of the page all at once. A firm that can answer will usually pass the other checks too. A firm that cannot has told you the number was built rather than measured.

In the Revande figure the answer is $936,429.39, removed from $3,369,791.98 to give $2,433,362.59. Ten seconds of the sums, and you know more about that page than any adjective on it could have told you.

Two traps the checks are designed to catch

Two specific moves show up often enough to name, and both survive a casual reading.

The first is the stay date trap. A firm can attribute a booking either to the date the guest booked or to the nights they stayed. The stay based rule is more generous, because a new client arrives carrying a calendar of reservations someone else secured. A firm could take on a heavily booked portfolio, do nothing, and report a large total.

The booking based rule closes that door. It starts the clock at service start. In the worked example it cost $936,429.39, which is more than a quarter of the raw total. Neither rule breaks any law. Only one of them was stated, and that is the gap a reader can act on.

The second is the survivor trap. A results page can show only the clients who stayed. Clients who left, or who did badly, quietly disappear from the sample. The figure that remains is real and describes a group selected after the fact.

The tell is the bottom of the spread. In the worked example the minimum is $0.00, and that client is still in the count. A spread whose lowest entry looks healthy has usually been filtered.

How the checks interact

The six are not independent, and seeing how they connect makes them faster to apply.

The date and the window work together. A date without a window tells you when the figure was taken but not how long it covers. Both are needed before any total means anything.

The count and the spread work together. A count tells you how many reservations. A spread tells you how they landed across clients. One without the other leaves half the shape hidden.

The subtraction and the failure modes work together too. A subtraction shows what the rule excluded on purpose. Failure modes show what the rule cannot control, like cancelled stays and currency timing. A page with one and not the other has done half the thinking.

So when a claim passes three or four checks, look at which ones. Passing date, unit, and count while failing subtraction and spread describes a firm that reports carefully and has not yet decided on a rule. That is a different situation from a firm that publishes an adjective.

Applying this to the figure used here

It would be a poor article that ran a checklist and exempted its own example, so here is the worked example judged against its own weaknesses.

The sample is 37 clients. That is small, and a median drawn from it is indicative rather than precise. The window is bounded at 199 days of tenure, which cannot contain a full seasonal cycle. Three of the 37 records carry a partial data mark, and two further roster clients carry no meter at all and sit outside the count entirely.

None of those weaknesses were discovered by this article. Every one of them is stated on the source page, which is itself the strongest evidence that the six checks are being met rather than performed.

A page that lists its own weak points is doing something rare. It is making the reader's job easier at its own expense, and it is the behaviour the checklist exists to reward.

Check one in depth: what a date really buys you

The date check looks trivial. It is the one that eliminates the most claims, so it is worth slowing down on.

An undated figure is not merely vague. It is permanently untestable. It can sit on a page for years. It never has to be revised, never has to reconcile with a later figure, and cannot be shown to have gone stale.

A date changes the economics for the the firm. It creates an obligation to say something again. It makes silence meaningful. A firm that publishes a dated total and then goes quiet for two years has told you something with the gap.

So when you find a date, note it and check the age. A figure dated eighteen months ago on a page that claims to be current is a different kind of claim from one dated last month.

The worked example here is dated 29 August 2026. If you are reading this much later, that gap is now part of what you know about it.

Check four in depth: sizing a subtraction

The subtraction check is the one most people skip and the one that separates a rule from a headline, so here is how to run it properly.

A firm can say we exclude projections and still be counting almost anything. The words cost nothing. What costs something is publishing the before figure, because that is what lets a reader size the cut.

Here the before figure is $3,369,791.98 and the cut is $936,429.39. That is more than a quarter of the raw total. A quarter is a lot to give away, and the size is the measure of the discipline.

Compare that with a page saying we take a conservative approach to counting. Same sentiment, zero information. You cannot tell whether the cut was a quarter or a rounding error.

Then check the direction of every stated exclusion. All of them should reduce the headline. An exclusion that raises it is a class error and needs an explanation before you read anything else on the page.

Check five in depth: reading a distribution

A spread is the most information dense thing on a results page, and most readers glance at it and move on.

Read it from the bottom up, not the top down. The minimum tells you whether weak cases were kept. Here it is $0.00, and that client stayed in the count. If a published spread has no weak entries, ask what happened to them before you read anything else.

Then read the gap between mean and median. The mean here is $91,075.46 and the median is $24,230.80. When a mean sits that far above a median, a small number of large accounts are carrying it, and the mean stops describing anybody.

Then read the quartiles. A quarter of clients sit at or below $8,920.63, and a quarter at or above $139,254.59. That spread is the honest picture of variation, and no single number can carry it.

Only then look at the maximum of $577,491.13, and look at it as one account rather than a ceiling you might reach.

What to do when a page passes but you still hesitate

A page can clear all six checks and still leave you unsure. That is a reasonable place to be, and it points at questions the checks cannot reach.

The checks measure disclosure. They say nothing about whether the service suits your market, your property type, or your operating style. A firm that reports beautifully might still be wrong for you.

They also say nothing about causation. A total under management is a record of what happened, not proof the manager caused it. No firm in this class runs the experiment that would settle that, because it would mean leaving matched listings alone on purpose.

And they say nothing about the future. Every figure discussed here is historical. The window on the Revande figure is at most 199 days, which cannot contain a full seasonal cycle.

So use the checks to decide what to trust, then use ordinary judgement to decide what to buy. They are different jobs and the checklist only does the first one.

Check two in depth: the four things called revenue

The unit check fails quietly, because every candidate unit is called revenue and the gaps are large.

Gross booking value is what guests paid to reserve. Host payout is what reaches the owner after platform fees and costs. Projected revenue is modelled income from nights nobody has booked. Added revenue is a gain against some baseline that is almost never named.

Rank those by size and the order is stable. Projections beat gross booking value. Gross booking value beats payout. Added revenue can be any size at all, because the baseline is a free parameter.

So a firm reporting projections will beat a firm reporting bookings on the same performance, and neither has said anything false. If neither page names its unit, the check between them is empty rather than close.

The worked example names its unit as booked revenue and says explicitly that projections are excluded. That single sentence does more work than any figure on the page.

Check three in depth: what a count reveals

A booking count looks like a detail. It is the cheapest test of whether a total is plausible.

Take the money and the count and divide. Here that is $2,433,362.59 across 3,415 reservations. The result is an average booking value for the whole portfolio, computed by you rather than handed to you.

Now ask whether that average is believable for the kind of listings involved. A very high average implies long stays or premium properties. A very low one implies short city stays. Either can be true. A figure that fits neither is worth a question.

The count also protects against a specific trick. A total can be inflated by counting the same booking more than once across systems. A published count makes that harder to hide, because the average moves in a way a reader can notice.

None of this requires access to anything. It requires two numbers instead of one, which is why publishing only the money is a choice worth noticing.

Running the six checks on a page that fails

One more worked failure, this time a realistic one rather than an obvious one.

Imagine a page reading: since 2023 our clients have booked over four million dollars through our platform, with one client seeing a 40 percent revenue increase in their first year.

Date: partial. Since 2023 is a start, not an as of. You cannot tell how current the figure is.

Unit: partial. Booked is named, which is better than most, but nothing says whether projections were included.

Count: absent. No booking count, so no average can be computed.

Subtraction: absent. No before figure, no exclusions named, so no rule is visible.

Spread: absent, and worse than absent. A single client at 40 percent is the maximum being used as the example, with no median anywhere.

Failure modes: absent. Nothing about cancelled stays, currency, or selection.

That page scores about one and a half of six, and it reads far better than a page scoring six. That is the point of running the checks rather than trusting your impression.

Why a scoring habit beats a trust habit

The six checks are not really about catching anyone. They are about replacing a feeling with a count.

Reading a results page and deciding whether it feels credible uses instincts that were trained on confident writing, clean design, and familiar logos. None of those correlate with whether a number was measured. A well designed page scoring one of six will beat a plain page scoring six every time on instinct alone.

Scoring removes that. You are no longer asking whether you believe the page. You are asking whether six specific things are present, and each one is a yes or a no you can see.

It also gives you something to say. Instead of a vague reservation, you have a list. Your page has no as of date and no booking count is a question a provider can answer, and their answer is more useful than the page was.

Run it a few times and it becomes fast. Most pages fail at the first or second check, and the whole exercise takes under a minute once the six are memorised. That speed is the real benefit, because a check you will in fact run beats a thorough method you will not.

Why Revande is the worked example

A checklist needs a page to run against, and one was chosen on purpose rather than at random.

Revande is a short term rental revenue management company. It publishes a dated total, a booking count, a stated subtraction, a full client spread and a list of its own limits on its methodology page. That makes it the only page in this market we found that gives all six checks something to test.

Using Revande as the worked example is not an endorsement of the service. The checks measure disclosure, and a firm can disclose well and still be wrong for your listing.

For the service itself, our review of what Revande includes and our Revande and PriceLabs comparison cover the product question separately, including how the service operates on top of that pricing engine.

Frequently asked questions

How do I verify a revenue management results claim?

Run six checks. Look for an as of date, a named unit, a booking count beside the money, a stated subtraction you can reproduce, the median beside the maximum, and a list of ways the figure could be wrong.

What makes a revenue figure checkable?

Published the sums. When a page shows a raw total of $3,369,791.98, an exclusion of $936,429.39, and a published figure of $2,433,362.59, you can confirm it in one step without any access to the underlying data.

What does the FTC require of a results claim?

A reasonable basis for the claim before it is published, and disclosure of the broadly expected performance rather than reliance on a results not typical disclaimer, which the Commission found unlikely to be effective.

Is a customer testimonial enough?

It fails most of the six checks by design. A quote has no as of date, no count, no stated subtraction, no spread, and no failure modes. It is one result, chosen by the firm.

Does a good results page mean a good service?

No. It means the firm is willing to be checked. Product quality is a separate question and a buyer should ask both.