Expected loss is a function of turnover — not deposit
A 96% RTP game does not mean a player who deposits €100 loses €4. RTP describes what happens to money as it is wagered — turned over — not what happens to a deposit. The operator keeps roughly 4% of every euro staked, and the same money gets staked many times over before it is gone: a win gets rebet, a loss gets chased, and the deposit passes through the game far more than once. That gap between “money deposited” and “money wagered” — turnover — is where the whole economics of a CRM campaign actually lives.
Because a player who plays until their bankroll is exhausted has, on average, lost the entire bankroll by the time they stop, and every euro of turnover costs the house edge in expectation, the two must be equal: expected turnover × house edge = starting bankroll. Rearranged, that gives the single most useful formula in this section:
Expected turnover to ruin = bankroll ÷ house edge.
This ignores overshoot — the technical detail that the losing spin which finally exhausts the balance usually takes it slightly below zero rather than landing on it exactly (a result known as Wald's identity). That correction shifts the answer by a rounding error, not by an order of magnitude, so the formula above is the one worth carrying around.
Worked through: a €100 bankroll at 96% RTP — 0.04 house edge — gives an expected turnover to ruin of €2,500. A €100 deposit is a €2,500 revenue event for the operator, not a €4 one.
That number stays abstract until it is turned into a session. At €1 a spin, €2,500 of turnover is 2,500 spins. The UKGC's own remote technical standards set a minimum game-cycle time for a slot of 2.5 seconds (RTS 14D — other casino games get a 5-second floor, RTS 14G). At that floor, 2,500 spins is a minimum of 1 hour 44 minutes of continuous play — a real session length, not a number on a spreadsheet.
How an operator actually makes money
Every deposit an operator takes in becomes turnover, and turnover is what the rest of the P&L is built from. The waterfall below runs top to bottom in the order money actually flows — from the gross number the games generate down to what the business keeps — and each line names the CRM decision that actually moves it, so a campaign brief can be pointed at the right target instead of a vague “drive revenue.”
| Line | What it is | What moves it |
|---|---|---|
| Turnover | Total amount staked, including rebet winnings — not deposits. | Acquisition volume, game mix, and session length: the whole point of CRM. |
| GGR | Gross Gaming Revenue = turnover × house edge, aggregated across players. | Which games get promoted or weighted, and what RTP mix a segment plays. |
| − Bonus cost | What promotions actually pay out in expectation, once wagering, caps and busts are counted. | Directly, by campaign design — wagering basis, weighting, caps, eligibility (Pillar 3). |
| − Payment processing fees | Card and PSP fees on deposits and withdrawals. | Payment method mix, and indirectly, how often a campaign drives repeat deposits. |
| − Gaming duty | A jurisdiction-set percentage of GGR, paid to the regulator. | Not a CRM lever at all — but it shapes which markets and products are worth promoting hardest. |
| − Affiliate revenue share | A cut of revenue paid to acquisition partners on the players they bring in. | Channel mix — a joint CRM/marketing decision about which acquisition sources to lean on. |
| NGR | Net Gaming Revenue — what is left after every deduction above. | The number a campaign should actually be judged against, not GGR or bonus face value. |
Most bonus campaigns are briefed and reported against GGR, or worse, against turnover itself — but the line a CRM team's bottom-line impact actually shows up on is NGR, three deductions further down. A campaign that generates real turnover and real GGR can still be NGR-negative once its bonus cost is subtracted, which is exactly why bonus cost — not bonus face value — is the number the rest of this Academy keeps coming back to.
The vocabulary that carries the P&L
Ten terms, each tied to a specific CRM decision rather than offered as trivia. If a briefing document or a vendor pitch uses one of these loosely, that is the moment to ask which of the columns below it actually means.
| Term | What it means | The CRM decision it drives |
|---|---|---|
| RTP | Return to player — the long-run share of turnover a game pays back out, as a percentage. A 96% RTP means the operator keeps 4% of everything wagered, over the long run. | Which games are cheap or expensive to run a bonus campaign on — moving the RTP a campaign is weighted toward changes bonus cost more than any realistic change to the wagering requirement. |
| House edge | e = 1 − RTP. The operator’s structural margin on every unit staked, expressed as a fraction. | The multiplier in every expected-loss and expected-cost calculation a campaign brief should include. |
| Volatility | How spread out a game’s payouts are around its RTP. Low volatility pays small amounts often; high volatility pays rarely but big, for the same long-run RTP. | What a session actually feels like to a player, and therefore retention, bankroll needed to survive a promotion, and how a bonus offer’s completion odds and payout both move (the RTP pillar covers this in depth). |
| Hit frequency | How often a spin returns any win at all, regardless of size — separate from RTP and separate from volatility. | How "alive" a game feels in a short session, independent of what it actually costs the operator. |
| Turnover | The total amount staked across a session, including money re-wagered out of previous wins — not the amount deposited. | The base every expected-loss and every wagering-requirement calculation is built on. This is the single term that most CRM copy gets wrong by substituting "deposit" for it. |
| GGR | Gross Gaming Revenue = turnover × house edge, aggregated across players — the top of the P&L waterfall. | The top-line number a campaign should be judged against, before any deduction. |
| NGR | Net Gaming Revenue = GGR minus bonus cost, payment processing fees, gaming duty and affiliate revenue share. | The real bottom line — a campaign that looks positive on GGR can still be NGR-negative once bonus cost is subtracted. |
| Bonus cost | What a bonus offer actually pays out in expectation once wagering, caps and player busts are accounted for — not its face value. | The number a campaign P&L should be built on instead of the headline bonus amount credited. |
| Wagering requirement (WR) | The turnover multiple a player must clear on a bonus, or on bonus plus deposit, before it converts to cash. | How completable an offer is on paper — but it is one of several dials operators hold, not the only one (basis, weighting and the cashout cap all move it further). |
| Game weighting | What fraction of a stake on a given game counts toward clearing a wagering requirement. | Real completability, more than the wagering requirement figure itself — a 10% weighting turns an offer that looks easy into one that is arithmetically almost impossible to clear. |
What the regulator requires you to tell players
This is not background reading — it is a product and compliance obligation CRM copy runs into constantly, on game pages, promo terms and lobby content alike.
Before the player commits
The UKGC's Remote Technical Standards, RTS 3C require that information enabling an informed decision about the chances of winning is easily available before the customer commits to gamble. That requirement is satisfied by any one of three forms: the average theoretical return-to-player percentage, the house edge, margin or over-round, or the probability of each winning event occurring. It does not require all three — but it does require one of them to be genuinely easy to find, not buried in a terms page a player has already committed money before reaching.
Losses cannot be dressed up as wins
The same standards, RTS 14F, are explicit and worth quoting exactly: “The gambling system must not celebrate a return which is less than or equal to the total stake gambled.” A payout that gives back less than what was staked is a loss, whatever the animation on screen does — a rule that matters here because CRM messaging about “wins,” near-misses and streaks runs into exactly the same principle: dressing up a net loss as a win is a regulatory problem before it is a copywriting choice.
Where this goes next
RTP and house edge set what a euro of turnover costs. They barely touch what a session actually feels like to the player sitting through it — that is volatility's job, and it is the subject of the next pillar, RTP and volatility, in depth. Once turnover, house edge and volatility are load-bearing vocabulary rather than jargon, the bonus conversion math in Pillar 3 — the 35x-bonus-only-versus-35x-deposit-plus-bonus distinction, conversion caps and the real mechanics of bonus hunting — will make a lot more sense on a first read.
If your team is running bonus campaigns without a shared answer to “what does this actually cost against turnover, not face value,” that gap is exactly what a CRM growth audit is built to close.
A note on responsible gambling
Everything above describes aggregate economics — expected values, averaged across many players and many sessions. No real player is an average: someone can bust in the first fifty spins or run far past the expected turnover to ruin before they stop, and the maths in this pillar predicts the population, not any one person's night. That matters for how it gets used. A grind ratio above 1 (Pillar 3) is, by construction, an offer most players cannot complete — a fact worth pricing into term design and affordability thinking, not just into a campaign's expected NGR.