How much should come back: putting a fair refund amount on a stay that was partly spoiled

You had two noisy nights out of five. The street outside the room turned into a party after midnight, and twice you barely slept. The rest of the stay was fine — the room was clean, the bed was good, the breakfast was nice. Now you want to ask for money back, and you hit a wall that isn't the property's policy at all. It's a question in your own head: how much? Asking for everything feels greedy and easy to laugh off. Asking for nothing feels like letting them keep money you shouldn't have paid. So you freeze, write nothing, and the trip stays spoiled and unpaid for.

That freeze comes from one wrong belief: that a bad stay is either a full refund or nothing. It rarely is. The fair refund amount for a spoiled stay is a proportion of the price — a slice that matches the share of the stay that failed. This guide is about that slice: how to put a fair, proportionate number on a partly ruined stay, what a realistic sense of scale looks like, and why a measured figure persuades far better than an inflated one. None of this is a promise of an outcome, and none of it is legal advice — it's general information to help you decide what to ask for.

A refund amount is a slice, not a switch

The first thing to unlearn is the on/off switch. Most people picture a refund as a single button: either the whole price comes back, or none of it does. Real outcomes rarely work that way. The honest refund amount for a partly spoiled stay is a proportion — small or large — that lines up with how much of the stay was lost.

Think of it the way you'd think about any service you only half received. If a five-course meal arrived with one course missing, you wouldn't expect the whole bill cancelled, and you wouldn't expect to pay in full either — you'd expect to pay for the four courses that came and not for the one that didn't. A stay works the same way: two ruined nights of a five-night booking isn't a full refund, it's roughly the value of those two nights, adjusted for how badly they were ruined.

This is what people mean by a partial refund or a price reduction: money back in proportion to the failure, not in proportion to your frustration. The frustration is real, but the figure that gets paid is the one tied to the part of the service that didn't work. Once you see the refund amount as a slice rather than a switch, the whole question of "how much" gets calmer — you're no longer arguing all-or-nothing, you're measuring a share.

That last row is worth sitting with, because it runs against instinct: it feels like asking for more should get you more, when in practice the opposite is true. An inflated number invites a flat "no", because it's plainly unfair and easy to argue down. A figure that visibly matches the problem is harder to refuse, because there's nothing to push back on — it's the arithmetic of what you didn't receive.

A rough sense of scale

So how much does a given fault knock off the price? There's no fixed answer, and anyone who gives you an exact number is guessing. But there is a rough rule of thumb that travel professionals have used for years to sketch the scale — a long-standing industry reference sometimes called the Frankfurt table — for "how much of the price a given problem tends to reduce it by". It isn't law, it isn't a promise, and Booking.com is not bound by it. It's a sense of scale, nothing more. Used that way, it earns its place: it stops you guessing wildly and gives you a defensible starting point.

Read the table below as exactly that — a rough guide to how much a single fault tends to be worth, expressed as "roughly X% off the price for the affected nights". Your specific rate's policy and your local consumer rules always sit on top, and the real figure depends on how bad and how long the fault was.

A few honest words about how to read this. The percentages are per fault and per affected night, not for the whole booking. If the noise hit two nights of five, the reduction applies to those two nights, not to the entire price. And the ranges are wide on purpose — "noise at night" covers everything from a single barking dog to building work that ran until 3am every night, and those are not the same claim. Sit your number where the severity sits.

Match the figure to the share that failed

The table tells you how much a fault is worth, and the other half of the calculation is how much of the stay it touched — put the two together and you have a defensible refund amount.

Start by splitting the price into nights: a five-night stay that cost 500 is 100 a night. Then ask two questions: how many nights did the fault affect, and how badly? Apply the rough percentage to the price of those affected nights alone, and that's your proportionate figure.

A couple of worked examples make it concrete:

• Two noisy nights of a five-night stay. Price is 500, each night is worth 100, and two affected nights are worth 200. Night noise sits in that ~10–40% band; say the noise was bad but not nightly torment, landing you around 25%. That's 25% of 200 — about 50 back. Not the whole 500, not nothing: a slice that matches two spoiled nights out of five.
• No hot water for the first day of a three-night stay. Price is 300, a night is 100. The fault touched one night and the no-hot-water band is around 15%. That's 15% of 100 — about 15 back. A small figure, but a fair one — and small figures are where the next section matters.

Notice what the maths does for you. It turns a vague sense of "they owe me something" into a number you can defend in one sentence: "the room had no hot water for the first of three nights, and I'm asking for roughly 15% of that night back." Nobody can call that greedy, because it plainly isn't. That's the quiet strength of a proportionate figure — it carries its own justification.

When a small refund isn't worth chasing

Here's the honest counterweight, because a trustworthy guide has to say it: not every annoyance is worth a refund request, and some figures are simply too small to be worth chasing.

• Trivial annoyances count for nothing. A slightly dated carpet, a lift that was slow, a pillow you didn't love, a receptionist who was curt — these are real irritations, but they aren't a failure of the service you paid for. No fair refund amount attaches to them, and asking for one mostly weakens your credibility for the things that do matter.
• Very minor faults rarely earn a separate claim. As a rough line, a fault that knocks off under ~10% of an already-small night is more effort than it's worth to chase on its own. If the whole reduction works out to a few units of currency, the time and back-and-forth rarely pay off. That's a judgement call, not a rule — but it's an honest one.
• The exception: when minor faults stack. A handful of small problems together can add up to a stay that genuinely fell short, and then they're worth raising as a group. A room that was a bit grubby and had a faint smell and lost its promised view is a different conversation from any one of those alone.

None of this means you should talk yourself out of a fair claim. It means you should aim the energy where the failure was real. A clear request about a genuine problem lands far better than a long list that mixes one real fault with five gripes — the gripes drag down the part that deserved to be paid.

When faults stack — and the ceiling on adding them up

That brings us to the trickier arithmetic: what happens when more than one thing goes wrong, or when a single fault drags across several nights of the booking.

When a stay has more than one real fault, the reductions do add up. A room with no hot water and persistent noise is two separate failures, and a fair refund amount can reflect both. You don't have to pick the worst one and ignore the rest. Work each fault out against the nights it affected, then add them together.

But there's a ceiling, and it's important to know it so your figure stays credible. When faults pile on top of each other, a single accommodation stay's combined reduction is generally treated as not running past about half the price — roughly a ~50% cap on the whole booking. The logic is plain: even a fairly bad stay was still, on some level, a place you slept. A figure that climbs toward or past the full price is the territory of a room that was unusable — bed bugs, no heating in deep winter, a room you couldn't stay in — which is a different case from a stay that was poor but liveable. If your stacked reductions push past half, that's a sign you've either double-counted or you're in full-refund territory and should frame it that way instead.

So the discipline is plain: add the real faults, apply them to the right nights, and sanity-check the total against that rough half-the-price ceiling. A number that respects the ceiling reads as fair. A number that ignores it reads as a grab.

The right number depends on the right remedy

One more thing decides the figure, and it's easy to miss: the size of the refund depends on what you're owed, and a refund is only one of the things a spoiled stay can earn. The fair outcome isn't always money back — it can be a move to a working room, a credit, or compensation for a real cost the failure forced on you. Getting the figure right starts with getting the right remedy for the problem — because the refund amount only makes sense once you know whether you're asking for a price reduction, a replacement, or a real out-of-pocket cost on top.

Keep those two things separate in your head:

• A price reduction is the slice of the room price you didn't get value for — the partial refund this whole guide is about. It's measured against what you paid for the stay.
• A real cost the failure forced on you is separate money — a night you had to book elsewhere because the room was unusable, a taxi to it, laundry for clothes an infestation ruined. That sits on top of the price reduction and is measured by your actual receipts, not by a percentage.

Mixing the two up is the most common way a request goes wrong. If you fold a replacement-room receipt into a "percentage of the price" claim, the number stops making sense and gets easy to challenge. Keep the price reduction as a clean proportion of what you paid, and list any real extra costs separately with their receipts. Two tidy numbers beat one muddled one.

Write the figure down — and let the request carry it

Once you have your number, the request almost writes itself, and a written request with a clear figure is far harder to wave away than a vague complaint. A good one says, in plain order: what you booked, what went wrong, which nights it touched, and the proportionate amount you're asking for and why. That last "and why" — the one sentence linking the figure to the share that failed — is what makes a measured number look as fair as it is. That's the part we can take off your hands — 2refund turns your answers into a clear request that's yours to send.

A short checklist for landing the figure before you send anything:

• Split the price into nights, so you know what a single night was worth.
• Decide which nights the fault touched, and don't claim against the nights that were fine.
• Pick a percentage from the rough sense of scale, sitting it where the real severity sits — not at the top of the range by default.
• Add separate faults, then sanity-check the total against the rough half-the-price ceiling.
• List any real extra costs separately, each with its receipt, never folded into the percentage.
• Write the one-sentence "why" — the link between the figure and the share that failed.

How much should I actually ask for after a partly spoiled stay?

A proportion of the price that matches the share that failed — that's the honest answer to "how much". Work out the value of the affected nights, apply a rough percentage for how bad the fault was, and that's your partial refund. A small, well-explained figure is far stronger than a big one you can't justify.

Two of my five nights were ruined by noise. Can I get the whole stay refunded?

Usually not the whole stay, no — the three good nights were a service you did receive. The fair refund amount is a price reduction for the two noisy nights: their value, reduced by a rough percentage for how bad the noise was. That proportionate figure is the one a property finds hard to refuse.

Several little things went wrong. Do they add up, or do I just claim the worst one?

They can add up — a fair refund amount can reflect every real fault, each measured against the nights it touched. But there's a rough ceiling: a single stay's combined reduction is generally treated as not running past about half the price. If your total pushes past that, you've probably double-counted, or the room was actually unusable and you should frame it that way.

The fault was tiny. Is it even worth claiming?

Often not on its own. A trivial annoyance earns no refund, and a fault worth under roughly 10% of an already-small night is frequently more hassle than it's worth to chase alone. The exception is when several small faults stack into a stay that genuinely fell short — then they're worth raising together.

The whole thing comes down to one calm idea: the fair refund amount for a spoiled stay is a slice, not a switch. Measure the share that failed, put a proportionate figure on it, keep any real extra costs separate, and write it down plainly. A number that visibly matches the problem is the one that's hardest to argue with — and the one most likely to bring the right share of your money back.

The cases in this article are illustrative composites, not real client records. We build them from the patterns we see again and again across the disputes we help with, because we can’t share real customers’ booking details. The stories are invented; the way they play out is true to our experience.

This article is general information, not legal advice. We’re a self-help tool, not a law firm. Rules, fees and deadlines change and vary by country, so always check the policy attached to your own booking and your local consumer protections.