Know exactly what the Buy Bonus button costs before pressing it

On a slot with the feature enabled, a button marked Buy Bonus or Buy Feature sits beside the reels or in the lower control panel. It displays a price expressed as a multiple of the current stake, so the figure moves whenever the stake changes. Selecting it deducts that amount from the balance in a single transaction and starts the bonus round on the next spin, with no base game play in between.

What arrives after the purchase is the same feature the base game would have produced eventually. The free spin count, the multiplier mechanics, the symbol behaviour and the distribution of possible payouts are unchanged. A studio does not build a separate, more generous bonus for buyers; it builds one bonus and sells an entry ticket to it.

This is the point most often misread. Buying does not improve a player's position against the house. Every outcome inside the round is produced by the same certified random number generator that governs base play, and no sequence of purchases makes the next result more likely to pay. The only thing a purchase removes is the waiting — and on a high-volatility title, that wait is where a meaningful share of the stake would otherwise have been spread across small base-game wins.

Both the price and the certified return for each purchase option are published inside the game itself, in the information or paytable panel. On titles offering several buy options, those panels list a separate return figure for each one. Checking that panel before the first purchase takes a few seconds and is the only reliable source, since figures can differ between game versions and between operators.

Buy pricing is not arbitrary. A studio starts from how often the bonus triggers naturally and works backwards. If a feature arrives on average once every 200 spins, then 200 spins at a given stake represent the expected cost of reaching it the ordinary way, and the buy price is set near that figure. The result is a price that roughly matches what the base game would have taken, adjusted for the value the player gives up by skipping it.

Titles with rarer, richer bonuses carry higher prices; titles with frequent, modest features carry lower ones. Four broad tiers cover almost everything on the market.

Budget buys: 2x to 50x

The cheapest entries deliver a plain feature — a standard free spins round without extra multipliers, preloaded wilds or guaranteed retriggers. ELK Studios sits at the bottom of the range, with X-iter menu options starting around 2x the stake for small enhancements rather than full bonuses. This tier suits frequent, low-cost entries rather than large single events.

Standard buys: 50x to 100x

Most of the market lives here. White Rabbit, Money Train 2 and Gates of Olympus all price their main feature at 100x. At a stake of one unit, that means 50 to 100 units for a single activation — the equivalent of a long base-game session compressed into one transaction, and the reason this tier feels expensive despite being the market norm.

Premium buys: 100x to 500x

Higher prices unlock stronger versions of the same feature: additional starting spins, elevated multiplier values, preloaded wilds or guaranteed retriggers. ELK's Nitropolis charges 300x for its enhanced tier against 100x for the standard one, and Hacksaw's Wanted Dead or a Wild offers 80x, 200x and 400x entries into three different bonus rounds on the same title.

Ultra-premium buys: 500x to 2,000x

NoLimit City anchors the top of the market, with Deadwood priced at 750x for its enhanced feature and San Quentin xWays and Mental reaching 2,000x. One detail here deserves attention because it is an industry response to a real risk: when the 2,000x option is selected on San Quentin xWays, the studio caps the maximum stake at four units per spin. That ceiling exists specifically to limit exposure on a single transaction, and it is worth understanding as a design decision rather than a restriction.

Two cross-references are useful at this point. Readers weighing whether a higher tier is worth its price should read the return figures published for each option, since tiers are certified independently and the most expensive is not automatically the strongest. Readers unfamiliar with the cheapest tier should see how multi-tier menus differ from a single fixed buy.

On most modern titles, yes — modestly. The certified return in buy mode typically sits 0.3 to 1.0 percentage points above the base game figure. The reason is compensation rather than generosity: a player who skips the base game gives up every small win it would have produced, and the studio adjusts the buy variant upward so the option is not mathematically worse than playing normally.

Three published examples show the scale. White Rabbit by Big Time Gaming returns 97.39% in ordinary play and 97.72% when the feature is purchased. San Quentin xWays by NoLimit City moves from 96.03% in the base game to 96.95% on its 2,000x option. Olympus Hades Megaways returns 95.95% in base play against 96.43% inside the Hades free spins reached through a 75x buy.

Those differences are real, but they describe long-run expectation across an enormous number of rounds, not something a player will perceive during a session. A gap of half a percentage point is invisible against outcome swings that routinely run from zero to several thousand times the stake. Treating a slightly higher published return as a reason to buy misreads what the figure measures.

Two further points matter when reading any return figure for this feature. On titles with several buy options, each option carries its own certified return, so the relevant number is the one attached to the specific tier selected — not a single figure for the game as a whole. And the same title can ship with different configured returns at different operators, which is why the in-game panel, rather than any third-party listing, settles the question.

Base and buy-mode return figures, by title

The return figure answers only half of the question a buyer is really asking. What a purchase changes about the shape of a session matters considerably more than a fraction of a percentage point.

The mechanic does not alter the maths of a slot. It alters the rate at which that maths is experienced, and that distinction explains almost everything players observe when they switch from ordinary play to buying.

On a high-volatility title, a bonus arrives naturally somewhere between once every 200 and once every 500 spins. Across those hundreds of spins the swing is absorbed: small base-game wins arrive, the balance drifts rather than lurching, and the eventual feature lands against a background of ordinary play. Purchasing removes that background entirely. The same distribution of outcomes now resolves every thirty to sixty seconds, one transaction at a time, with nothing in between to cushion it.

The outcome distribution inside a feature is wide. Industry estimates place the share of purchased rounds returning less than their own cost at roughly 30% to 50%, with individual results running from nothing at all to several thousand times the stake. A round with an expected value around 80x the stake will, in practice, produce a long series of results well below that figure interrupted by rare ones far above it. Buying ten features in succession on an extreme-volatility title is therefore a materially different experience from letting ten features arrive over an hour of base play, even though the underlying numbers are identical.

The consequence shows most clearly in hourly cost. At a stake of one unit with a 100x buy taken roughly every two minutes, a session consumes in the region of 3,000 units per hour. Ordinary play at the same stake, depending on spin speed, runs closer to 300 to 600 units per hour. Nothing about the return percentage has changed — but the rate at which the balance is exposed to it has increased roughly five to tenfold.

None of this makes the feature a poor choice in itself — it makes it a fast one. A player who understands that the same money is being exposed several times faster can size the stake accordingly. The cost reference below turns that into concrete figures, and bankroll sizing covers what to do with the result.

Regulators rarely intervene in the internal design of a slot. The bonus buy is one of the exceptions, and the reasoning is consistent wherever restrictions appear: the mechanic removes the spacing that base-game spins impose between high-cost events, allowing losses to accumulate faster than in ordinary play. Where an authority has concluded that this materially raises risk of harm, it has removed the option rather than limiting it.

When that happens, the game itself usually remains available. Studios ship a market-specific build in which the buy button is absent and the base game maths is recalculated — commonly with a slightly higher base return, since the value previously held in the guaranteed feature is redistributed. Overall return is broadly comparable, but the experience is not, and the same title can therefore behave differently for two players holding accounts in two markets.

United Kingdom

The Gambling Commission removed the feature from licensed sites in 2023 as a player protection measure, having concluded that the compressed pace of repeated purchases was associated with harmful outcomes. The restriction extends to demo play on licensed sites, not only to real-money accounts.

Sweden

Swedish rules have moved in the same direction, treating the mechanic as part of a broader effort to limit features that accelerate the rate at which stakes are placed. Operators licensed for the Swedish market configure titles without the buy option.

Spain

Spain has applied comparable restrictions within its wider framework on rapid-play features. As elsewhere, availability follows the operator's licence rather than the player's preference, so the option does not appear on locally licensed sites.

Regulated markets in North America

The picture here is uneven because authority sits with individual jurisdictions rather than a single national regulator. Some permit the feature, others have followed the restrictive approach taken in Europe, and rules are revised more often than in established markets. A player in this region cannot assume availability from a game's reputation elsewhere and should confirm with the licensed operator directly.

Most European, Latin American and Asian markets

Across the majority of regulated jurisdictions the feature remains permitted, with individual operators applying their own configuration. Availability of a specific title, and of specific tiers within it, still varies according to what each operator has licensed and enabled.

Crypto and offshore operators

The mechanic is almost universally available on operators outside the major licensed markets, frequently including the highest price tiers. Player protection standards, complaint procedures and dispute resolution differ substantially from licensed jurisdictions, and those differences are worth weighing independently of whether a particular feature is offered.

The titles below are selected for what they demonstrate rather than for how they perform. Between them they cover the origin of the mechanic, every price tier from 50x to 2,000x, single-price and multi-tier designs, and the range of feature structures studios have built around a purchased entry. None is presented as more likely to pay than another — that is not a property a slot has. Full return figures for each appear in the RTP table.

Bonus buy is not a single standardised mechanic. Each major studio has developed a recognisable house style — in how many options it offers, how it prices them, and whether it treats the purchase as a shortcut or as a menu of different games. Recognising the style is often more useful than recognising the title.

Big Time Gaming

The originator, and the studio that took the most unusual position on its own invention: after establishing the mechanic with White Rabbit in 2017, it stopped adding the option to new releases. Its designs favour a single moderate price alongside token accumulation, so base play gradually reduces the cost rather than being skipped outright.

Pragmatic Play

Standardisation. A large share of the catalogue prices its main feature at 100x with a single option, and the studio pairs it with ante bet as a lower-commitment alternative on the same titles. The result is predictability across a very wide library.

NoLimit City

The most aggressive pricing in the market, with tiers reaching 2,000x on San Quentin xWays and Mental and 750x on Deadwood. Notably, it also ships the clearest exposure control: a stake cap on the highest tier, which limits what a single transaction can cost.

ELK Studios

Menus rather than buttons. The X-iter system replaces one price with four to six, starting near 2x for minor enhancements and climbing to several hundred, each certified independently. It is the design that gives a player the most granular control over volatility.

Hacksaw Gaming

Tiering by feature type. Where most studios sell stronger versions of one round, Hacksaw commonly sells entry to genuinely different rounds — three distinct bonuses at three prices on Wanted Dead or a Wild being the clearest example.

Relax Gaming

Self-contained bonus rounds built around persistent symbols and respins, with the Money Cart series as the template. Pricing is typically two-step: a standard entry and a premium one that pre-activates a modifier, which makes the value of that modifier unusually transparent.

Four distinct designs sit under the same label. They differ in how many choices a player is given, whether the purchase is final once made, and whether the price can move. Understanding which one a title uses matters more than knowing its headline price.

Multi-tier menus and ELK's X-iter

Instead of one button, the player is shown a menu of four to six options at different prices, often including a mystery entry with a randomised outcome. Each option carries its own certified return, which produces a counter-intuitive but important consequence: the most expensive tier is not automatically the strongest in expectation. Some tiers also guarantee a minimum return on the round, a property an ordinary single-price buy does not have. The in-game panel lists the figures per option and is the only place to compare them.

Bonus gamble

A small number of titles allow the purchased feature itself to be staked before it is played. Extra Chilli is the reference case: eight spins are bought, then gambled up to three times toward a maximum of 24, with the probability of success rising from 50% on the first attempt to 60% afterwards. A failed gamble reduces the spins or returns the player to base play — meaning a completed purchase can be diminished before it produces anything.

Token accumulation and dynamic pricing

Introduced with White Rabbit, this design lets base play reduce the cost of the feature. Gold coin symbols landing during ordinary spins lower the displayed price incrementally, and a sufficient accumulation brings it to zero. Related dynamic pricing systems adjust the figure according to game state, so the price shown is a function of what has already happened rather than a constant.

Ante bet as the middle option

Ante bet is not a purchase and does not guarantee anything. It raises the stake by 25% per spin and roughly doubles the frequency at which scatter symbols appear, improving the chance of a natural trigger while leaving the outcome to the base game. Gates of Olympus offers a direct comparison: 200 spins at 1.25 units costs 250 units and produces around two natural triggers on average, against 100 units for one guaranteed entry. Which is preferable depends entirely on session length and budget, and neither is superior in the abstract.

The tables below show what a purchase costs at common combinations of stake and buy multiplier, and what continuous buying consumes over an hour. They describe expenditure only. Nothing here predicts returns, and no figure should be read as an indication of what a session will produce.

The second table applies the same figures to time. It assumes one purchase every two minutes, which is roughly the pace at which a feature round resolves and the next decision is made — approximately 30 purchases in an hour of continuous play.

Both columns below carry the same weight, and each point is stated with the figure that supports it.

Demo play uses virtual credits, so purchases cost nothing and winnings cannot be withdrawn. For this particular mechanic that makes it unusually useful: the properties worth knowing before spending money — how often multipliers appear, how long unproductive runs last, what a typical round actually returns — only become visible across a series of purchases, and a series is expensive at real stakes.

A meaningful test needs repetition. Five to ten purchases on the same title give a rough sense of how widely results spread and how frequently a round fails to return its own cost — the two things that matter most and the two least visible from a single attempt. On multi-tier titles, running the same number of purchases on two different tiers is more informative than running twenty on one.

Two limitations apply. On sites licensed in markets where the feature is restricted, the buy button is absent from demo play as well as real-money play. And demo results carry no predictive weight for a subsequent real-money session — the sample is far too small, and each round is independent of every other.

The interaction between purchased features and promotional balances is where avoidable losses most often occur, and it is poorly documented across the industry. The rules are set by operators rather than studios, so they vary — but the patterns are consistent enough to be worth checking before, rather than after, a purchase.

Three restrictions appear most frequently. Many operators exclude feature buys from wagering contribution entirely, so a purchase does not advance a bonus toward release. Others count them at a reduced rate, commonly in the region of 10% to 20% of the amount staked. A separate group blocks purchases from bonus balances altogether, so the button becomes unavailable until the promotional funds are cleared or forfeited.

A fourth issue is less obvious and more damaging. Most bonus terms impose a maximum stake per spin while a bonus is active. Because a buy price is a multiple of the stake, a single purchase can exceed that ceiling even when the underlying stake does not — and a breach of the maximum stake rule is a standard ground for voiding winnings. This is the mechanism behind a significant share of disputes about purchased features.

On withdrawals, identity verification is required before funds are released, and it is separate from any bonus condition. Verification is routine and often completed quickly, though large payouts relative to the deposit history are more likely to be reviewed manually, which extends the timeline. Purchases at the highest price tiers can produce exactly that pattern. Completing verification when the account is opened, rather than at the point of a withdrawal request, removes the delay entirely.

One point should be established before any of the rest: no method of dividing a budget changes the mathematics of a slot, improves the chance of a winning round, or affects the certified return. What budget structure does control is how long a session lasts and whether it ends at a planned moment or an abrupt one. That is the entire purpose of what follows.

The widely used reference point is to keep a single purchase within 5% to 10% of the session budget. The reasoning becomes clear alongside the distribution figures: if 30% to 50% of purchased rounds return less than their cost, a budget covering only two or three purchases has a substantial probability of being exhausted before any meaningful result appears. A budget covering eight or ten gives the distribution room to express itself.

In practice this means sizing the stake from the budget rather than from a desired payout. A player with 200 units and a 100x purchase has two attempts. The same player at a stake of 0.20 has ten. The maximum win multiplier is unchanged in both cases; only the number of opportunities to reach it differs, along with the size of any result.

Setting the session limit as a number of purchases rather than a sum of money is a small change with a disproportionate effect. A count is unambiguous and easy to track during play, whereas a monetary limit blurs as the balance moves. And raising the stake after a series of unproductive rounds does not recover anything: each round is independent, past results carry no influence over future ones, and a larger stake changes only the rate of expenditure.

The cost reference tables extend these figures across other stakes and price tiers.

The structural feature of this mechanic is that each purchase is a separate, deliberate decision made at speed. That is precisely the pattern regulators have pointed to when restricting it, and it is worth taking seriously regardless of which market a player is in.

A single fact carries more weight here than any general warning. Big Time Gaming — the studio that invented the feature and built the first title around it — stopped adding the option to new releases. That decision was made by the company with the most commercial reason to keep it.

Practical measures are straightforward. Deciding the number of purchases before a session begins, and treating that count as fixed regardless of results, removes the moment where a decision has to be made under pressure. Money committed to the feature is the price of entertainment, not an investment with an expected return. And a purchase made to recover a previous loss is the clearest available signal that a session should end, because it is a decision driven by the last result rather than by a plan.

Licensed operators provide deposit limits, loss limits, session reminders, time-outs and self-exclusion. These tools are free, take a minute to configure, and work better set in advance than in response to a problem.

Every licensed operator publishes the helpline funded in the market it serves, together with the account controls described above. Both are found in the account settings and the site footer, and neither requires a conversation with customer support to activate.

Terms used throughout this page, with the sense in which each applies to purchased features.

Every figure on this page comes from one of two places: the information panel inside the game itself, or documentation published by the studio that built it. Third-party listings are used to identify titles worth checking, never as a source for a number. Where a studio publishes a single return figure covering all play, this page records that rather than presenting an estimated buy-mode figure as though it were certified.

Titles are tested in demo mode before inclusion, with repeated purchases across each available tier so that price, feature structure and the behaviour of multi-tier menus can be confirmed rather than reproduced from secondary sources.

Where sources disagree, the disagreement is published. The maximum win on White Rabbit is reported between 10,000x and 13,000x across provider and catalogue documentation, and the table reflects that range rather than selecting one figure and presenting it as settled.

Selection for the notable titles section is based on what each title demonstrates about the mechanic — historical significance, coverage of the full price range, and representation of the four design types — rather than on performance, popularity or commercial arrangement. No operator or studio influences inclusion, ordering or the figures in any table.

Figures are re-verified quarterly and after any known game update, with the verification date shown beneath each table. Prices and returns can differ between operators and game versions, so the in-game panel remains the authoritative source for any individual account.