What keep what you win actually promises

The phrase keep what you win has a specific meaning in the terms of a spin offer, and it is worth pinning down before the arithmetic starts, because the phrase does a lot of advertising work for a fairly narrow promise. In a standard free spins grant, the money the spins produce does not arrive as cash: it arrives as bonus funds, and bonus funds carry a wagering multiplier, so a player who wins two pounds from a grant might need to push eighty pounds of stakes through a slot before any of it converts to a withdrawable balance. A keep-what-you-win grant removes that layer. The winnings land as real money from the first spin, or close to it, and whatever the reels pay out sits in the cash balance rather than a locked one.

What the phrase does not remove is everything else. Two clauses almost always remain, and they do most of the limiting work on this format. The first is the winning cap: a ceiling on how much the spin grant can pay out in total, so a session that runs hot on a volatile slot finds its conversion clipped at the cap no matter what the reels did. The second is the expiry: the spins themselves are credited with a window, commonly between twenty-four hours and seven days at operators of this kind, and a grant that is not used inside that window disappears unused. Neither clause is hidden in the dishonest sense; both live in the bonus terms a click below the banner. Both are simply quieter than the promise.

There is a further subtlety worth naming. Some operators print keep what you win on the banner while keeping a small multiplier in the spin terms, or applying the no-wagering treatment only to winnings above a threshold, or releasing winnings in instalments. The honest versions state plainly that spin winnings are paid as cash with a maximum total and nothing else attached. The way to tell the difference is to read the definition of the winnings in the terms: if the words bonus funds, playthrough, or turnover appear anywhere near the spin winnings section, the promise has a smaller footprint than the banner suggests. If the winnings section says cash and names a cap, the offer is what it claims.

For a reader deciding whether to take one of these grants at all, the practical summary is this: it is the cleanest spin format on the market, it costs nothing to accept, and its realistic value is the raw per-spin value trimmed by the cap and the expiry. On a thirty-spin grant at 10p, that realistic value is a couple of pounds, occasionally more if the session runs well and the cap is generous. That is the entire offer, priced honestly, and it is still a better structure than the multiplier versions that ask for forty times the winnings before anything converts.

The per-spin arithmetic of a thirty-spin grant

Every spin offer is priced by one number before any terms act on it: the per-spin value. Free spins in this market carry a per-spin value commonly between £0.05 and £0.20, and the figure is set by the operator rather than by the slot’s own minimum stake, so the same headline count can be worth four times more or less depending entirely on which end of the band the grant sits in. Thirty spins at 5p is £1.50 of raw grant. Thirty spins at 10p is £3.00. Thirty spins at 20p is £6.00. The count on the banner is identical in all three cases; the value is not close.

Taking the 10p case as the model, which is the assumption this page works with throughout, the gross grant is £3.00. On a no-wagering grant, that gross figure is the starting point of the realistic value rather than a number that decays through a playthrough. The next modifier is the slot’s return rate: a typical online slot returns around 96 percent over the long run, so the expected outcome of £3.00 of stakes at that rate is roughly £2.88, with the missing pence being the house margin on the volume. That expected figure is an average across all possible sessions, not a prediction of any single one, and a thirty-spin sample is far too small to converge on it.

The small sample is what makes this format interesting rather than worthless. Over thirty spins, the results scatter widely: a grant on a low-volatility title might land between £1.50 and £4.50 almost every time, while a grant on a high-volatility title might return nothing at all in a third of sessions and spike past £15 in the occasional one. The average across many grants trends toward the return rate, but the experience of one grant is a lottery ticket with a known expected value. A reader who understands that can treat the grant as a free sample of variance rather than as a small payment, which is a more accurate way to think about it than either the banner’s optimism or the cynic’s dismissal.

The cap then acts on the distribution rather than the average. A £20 ceiling does nothing to the expected value of a 10p grant, because the expectation sits under two pounds and the ceiling sits far above it; what the ceiling does is truncate the rare hot session. A £10 ceiling truncates it more visibly, cutting anything above that line. On the modelled figures, a grant with a £20 cap and no multiplier retains close to its full expected value, while the same grant with a £10 cap loses only the tail. Both are workable terms. The versions to be wary of are the ones where the cap sits at or below the expected value of the raw grant, because there the ceiling stops being insurance against outliers and becomes the offer’s real size.

One more line of arithmetic closes the picture. The winnings from a no-wagering grant sit in the cash balance, which means they are withdrawable immediately, subject to the operator’s minimum withdrawal. Some operators of this class set that minimum above the typical value of a small spin grant, which quietly converts the promise into credit for further play rather than cash out. The withdrawal minimum lives in the payments section of the terms, not the bonus section, and it is the last figure to check before treating a keep-what-you-win grant as money rather than as extra play.

The winning cap and the expiry clock

The two surviving clauses deserve their own examination, because between them they decide almost everything about how much a clean grant delivers. Start with the cap. Operators of this kind commonly set the ceiling on spin winnings somewhere between £10 and £50 for a no-deposit grant, with the smaller figures attached to the larger spin counts and the more generous figures attached to taster-sized grants. The cap works as a maximum total conversion from the grant: once the cumulative winnings from the spins reach it, the grant ends and any further spins in the bundle are either forfeited or continue spinning for nothing, depending on the wording.

The interaction between the cap and the slot choice is the part most readers miss. If the spins are tied to a high-volatility title, the cap is doing real work, because high-volatility games are precisely the ones that produce sessions far above expectation; the operator is effectively buying the upside of variance and leaving the player the average. If the spins are tied to a low-volatility title, the cap rarely binds at all, and the grant behaves almost exactly as the raw per-spin arithmetic predicts. Where the terms let you see which slot the spins run on, the volatility of that title tells you how much the cap will actually matter in practice, which is more useful information than the size of the ceiling in isolation.

The expiry is the quieter clause and the one that forfeits more value in aggregate, because it does not trim winnings, it removes unused spins entirely. Windows at operators of this kind commonly run from twenty-four hours for registration grants to seven days for the more relaxed versions, and the clock usually starts at crediting rather than at first use. A grant claimed on a Monday for a weekend session can expire before the session happens. The habit that avoids this is mechanical: claim the grant when you intend to use it, not before, and treat the expiry date printed at crediting as the deadline rather than an afterthought.

There is a third clause that occasionally reappears on this format even where the wagering has genuinely been removed: a maximum bet rule that applied to the spins themselves. It matters less on a fixed-value grant, since the per-spin value is set by the operator anyway, but some terms carry over language from their wagered offers, including stake limits on how the winnings may be played after crediting. Where the winnings are cash, such a limit is unusual; where the winnings sit briefly as bonus funds before release, it is common. The distinction is visible in the terms within a minute of reading, and it is the last check worth making before the grant is spent.

Set against the multiplier formats, the trade is clear. A grant of the same size with a forty-times multiplier on winnings converts a hot session into a long playthrough that usually ends in nothing, but it often carries a higher cap and a longer window. A no-wagering grant converts immediately but lives inside a tighter ceiling and a shorter clock. Neither is a trick; they are two price points on the same product, and the keep-what-you-win version is the one where the stated number and the experienced number stay closest together.

Where the small print re-enters

A no-wagering grant is the format where the terms matter least, which is not the same as the format where they matter not at all. The first place the small print re-enters is the definition of the winnings. The cleanest offers pay spin winnings to the cash balance at the moment each spin resolves. Others accumulate the session’s winnings and release them once the grant finishes, which introduces a short delay and, in a few cases, a manual review step before the balance moves. Neither arrangement costs money, but the second one means the balance is not immediately withdrawable, and a player expecting instant access should know which version they hold.

The second re-entry is the account itself. A no-deposit grant still requires a registered account, and operators of this class verify identity at some point in the journey: sometimes at registration, sometimes at the first withdrawal, occasionally only when the withdrawal crosses a threshold. Where verification happens late, the discovery that documents are needed arrives after the winnings exist, and the withdrawal pauses until the checks clear. This is not unique to free grants; it is how the whole class operates. The practical step is to expect the request and to have the standard documents to hand rather than treating a verification pause as a dispute.

The third re-entry is the promotional fine print around the grant: one account per household, one grant per person, exclusions for duplicate registrations, and the operator’s general right to void promotions where abuse is suspected. These clauses are standard across the market, licensed and offshore alike, and they exist because no-deposit grants are the most abused promotion in the industry. A reader taking one grant honestly has nothing to worry about from any of them; a reader planning to open several accounts across a household will find the terms written specifically against that, and the voiding of winnings under such clauses is enforced more often than any other term in the bonus book.

The fourth re-entry is the payment rail. Even where the winnings are cash, the rail they travel out on has its own terms: minimums, fees on some methods, and the review-plus-transit windows that operators of this class publish. Crypto rails move fastest and sometimes carry a network fee that matters more on a £3 withdrawal than on a £300 one; card and bank rails carry day-scale windows. A grant that survives the cap and the expiry intact can still lose a noticeable slice to a fixed fee on a tiny withdrawal, which is the last cost to price before deciding whether to cash out or play the balance through.

Read together, these four re-entries do not change the character of the offer. They trim it, delay it, and occasionally frustrate it, but they do not convert a real grant into an advertising device the way a heavy multiplier does. That is the honest ranking of formats: the keep-what-you-win grant sits at the clean end of the spectrum, and the small print on it is genuine small print rather than the main event.

The licence gap behind a free grant

A free grant with no multiplier still sits inside the licence question, and the licence question does not shrink because the money involved is small. An operator holding no licence from the Gambling Commission sits outside British regulation entirely, and the practical consequences for a reader in Britain are the same on a three-pound grant as on a three-hundred-pound deposit, scaled to the sums at stake. The first consequence is self-exclusion: GAMSTOP, the national online scheme, is compulsory for Commission-licensed operators and does not reach offshore ones, so a registration there does not by itself block an account at a site of this kind. Anyone who has self-excluded should treat that gap as the first fact of this market, ahead of any offer arithmetic.

The second consequence is the dispute route. A player of a licensed operator who disagrees with a withheld payout escalates through the operator’s complaints procedure and then to an independent Alternative Dispute Resolution body approved by the Commission. A player of an offshore operator has no access to a Commission-approved ADR for those transactions, so the operator’s own support desk is, in practice, the final word. On a small grant the sums rarely justify a dispute, but the structural point stands: the terms are enforced by whoever holds the money.

The third consequence is the softer layer of statutory safeguards that a Commission licence requires: customer funds held separately or protected to a stated level, reality checks on online accounts, account-level limit tools, and recognised safer-gambling information displayed throughout. Offshore operators follow whatever their own licence requires, which the reader must verify on the operator’s register, and the everyday experience is that limit tools and reality checks are thinner or absent. On a free grant this matters least in pounds and most in behaviour, because a frictionless no-deposit registration is the easiest account to open quickly and the hardest to keep deliberate.

None of this changes the arithmetic of the grant itself, which works identically wherever the operator is licensed. What it changes is what stands behind the promise when a promise meets a problem: the expiry that lapsed a day early, the cap that was read differently, the verification that stalls a small withdrawal. The licensed market has machinery for those moments. This end of it does not, and pricing an offer honestly means pricing that absence alongside the per-spin value.

A worked value model

The table below runs the modelled grant through every stage from banner to balance so each intermediate figure is visible. The assumptions are stated in the caption and repeated here: thirty spins at a 10p per-spin value, no wagering multiplier on the winnings, a £20 winning cap, and a typical 96 percent return rate on the underlying slot. Change any assumption and every downstream figure moves with it, which is what the calculator further down this page exists for.

Reading the table honestly means noticing which rows do work and which do not. The wagering row is zero, and that is the whole selling point of the format: nothing in the terms asks the winnings to be replayed. The cap row is also inert at these figures, because a £20 ceiling only acts on sessions that run several times above expectation, which happens occasionally on a volatile title and essentially never on a steady one. What remains is the return rate and the friction, and on a three-pound grant the friction, not the terms, is the largest remaining variable.

The same table run at the other end of the per-spin band tells the mirror story. Thirty spins at 20p produce a £6.00 gross grant, an expected return near £5.76, and the same inert cap; thirty spins at 5p produce £1.50 gross and an expected return under a pound and a half, where a fixed withdrawal fee of even a pound consumes most of the outcome. The spin count on the banner never changes across those three cases, and neither do the terms; the entire difference in value comes from the per-spin line, which is the reason this page keeps returning to it.

The variance around the expected figure is worth restating once more because thirty spins is a small sample. A grant on a low-volatility title will land within a pound of expectation almost every time. A grant on a high-volatility title can return nothing, or can spike past the cap and be trimmed at exactly £20, and both outcomes are consistent with the same model. The expected value prices the average across many grants; the experience of one grant is variance, freely given, and that is how the format is best understood rather than as a small wage.

How the market compares, row by row

The comparison below sets the British licensed baseline against a spread of offshore operators of the kind this page discusses. The columns are the ones that matter when costing a spin offer: the licence label, the shape of the welcome range in this category, the typical multiplier, the minimum deposit, the payout window, and the rails the cashier tends to carry. Rows are text rather than links, and the figures are category ranges, not quotes of any single promotion.

The figures in this table are category-typical ranges compiled from operator-published terms at the time of writing. They are not quotes of any single promotion. The UKGC rows are the British baseline, shown for comparison; none of them is claimable at the offshore offers this page sits beside.

The pattern worth reading out of the table is that the no-wagering structure this page costs is not confined to one licence class: some licensed operators have built whole brands on winnings-paid-as-cash formats, while offshore operators of this kind mostly attach spin winnings to their standard multiplier, with occasional no-wagering tasters used as acquisition. The difference in the surrounding machinery is larger than the difference in the spin terms. The licensed rows carry contractual payout windows, full statutory protections and the national self-exclusion scheme; the offshore rows carry faster crypto rails and larger printed numbers, with the protections described earlier in this page absent. Read the rows as a trade between terms and machinery, not as a ranking of generosity.