CPA vs Revenue Share vs Hybrid: iGaming Deal Economics

CPA vs Revenue Share vs Hybrid: iGaming Deal Economics

How to price a partner deal from measured retention, using a four-step break-even window instead of a rate card.

Table of Contents

• What Are CPA, Revenue Share and Hybrid Deals?
• Why Does Deal Structure Decide Who Carries the Risk?
• Which Operators and Markets Does This Cover?
• What Is the Break-Even Window?
• Step One: How Do You Measure Player Decay Before Pricing a Deal?
• Step Two: How Do You Price All Three Deals in the Same Units?
• Step Three: Where Does Revenue Share Overtake CPA?
• Step Four: What Review Point Should a Deal Carry?
• When Does CPA Win?
• When Does Revenue Share Win?
• When Does a Hybrid Make Sense?
• What Contract Terms Protect You on Each Model?
• How Do You Track Players Back to the Deal?
• What Should You Watch for When Agencies and KOLs Quote a Deal?
• What Compliance Limits Apply to These Deals?
• How Does AP Collective Approach Deal Economics?
• Frequently Asked Questions
• Final Takeaway
• The One-Page Action Spine

Between 55% and 65% of iGaming players who make a first deposit never make a second, according to one affiliate-tracking vendor’s player lifecycle guide. That one statistic decides who wins under each commission model. A partner paid at the first deposit never feels those lost players. A partner paid on revenue feels every one of them.

Most guides define the three models and list rate ranges, then stop. The harder question is how many months each deal takes to pay back and who funds the gap until then. Pick the wrong model for your traffic and you either overpay for players who vanish or underpay partners who then take their best traffic elsewhere.

This guide gives you a four-step method, the Break-Even Window, for pricing CPA, revenue share and hybrid deals from retention data. It is written for operators and marketing leads who pay partners per player, including crypto-native platforms such as exchanges, crypto casinos and prediction markets. The numbers in the worked example are illustrative, and we say clearly where our own experience ends.

Key Takeaways

• CPA pays at the first deposit, revenue share pays over time, and a hybrid splits the two. The real choice is a bet on how long players stay.
• One vendor guide reports that 55% to 65% of first-time depositors never deposit again, so retention decides the model.
• In our illustrative model, revenue share overtakes a $250 CPA at month 11 on sticky traffic and month 16 on median traffic, and not at all within 24 months on fast churn traffic.
• A vendor rule of thumb puts the crossover at month 3 to 5. Our model puts it at 11 to 16. Run your own numbers.
• Every deal should carry a review at day 30 and month 6, so a price built on 8% monthly decay is re-tested if the cohort runs at 15%.
• Gambling advertising rules differ by market. Have counsel review the model before the first payment.

What Are CPA, Revenue Share and Hybrid Deals?

CPA, revenue share and hybrid are the three ways an operator pays a partner for a referred player. CPA pays a fixed fee when the player completes a qualifying action, usually a first deposit above a minimum. Revenue share pays a percentage of the player’s net revenue for as long as the player stays active. A hybrid pays a smaller fixed fee up front plus a smaller revenue share.

Published ranges give a sense of scale and they disagree. One vendor lists casino CPA at $100 to $400 per first deposit and sportsbook CPA at $50 to $150, with casino revenue share at 25% to 45% of net gaming revenue. Another quotes €100 to €250 per depositing player and 25% to 55% of net revenue.

Net gaming revenue, or NGR, is the base for most revenue share deals. It starts from gross gaming revenue, which is bets minus player winnings, and subtracts direct costs such as bonuses. Because operators define those deductions differently, the same percentage can pay very different amounts.

Why Does Deal Structure Decide Who Carries the Risk?

Deal structure decides who carries the risk because the three models pay at different times, and the gap between a first deposit and long-term value is where most players disappear. The affiliate paid at the first deposit never feels that loss. The affiliate paid on revenue feels all of it.

Take a $200 CPA for a player who deposits $25, claims a welcome bonus and never returns. The same vendor describes this case as paying for a player worth negative revenue. Welcome bonuses can consume 15% to 25% of first-month gross gaming revenue, so early revenue is thinner than it looks.

Now reverse the case. A revenue share partner whose players stay for two years earns more than any CPA would have paid. The operator ends up paying more than it expected for traffic it could have bought at a fixed price. Each side is exposed to a different mistake. Operators overpay under CPA when traffic churns. Partners underearn under revenue share when it churns.

The decision is therefore a bet on retention, and the sensible way to make it is to measure retention first.

Which Operators and Markets Does This Cover?

This guide covers operators that pay partners per player and can measure those players’ revenue: online casinos and sportsbooks, and crypto-native platforms that pay affiliates, creators or agencies for referred users, such as exchanges, crypto casinos and prediction markets. The economics are the same in each. A referred player costs something up front and returns revenue over time.

Two limits matter. First, gambling is regulated, and rules on advertising, affiliate promotion and responsible gambling differ by jurisdiction, with some markets restricting parts of it. Check with counsel. We are not lawyers.

Second, a note on our own experience. AP Collective’s published work is in crypto and Web3, including exchange user acquisition and gaming and consumer apps with paying players. We have not published licensed iGaming casework, and nothing here claims it. Where we use iGaming numbers, they come from affiliate-software vendors, and we treat them as ranges to test, not rules.

What Is the Break Even Window?

The Break Even Window is the number of months a deal takes to pay back, set against the number of months the operator and the partner can each afford to wait. It turns the CPA versus revenue share argument into one question: when does each model cross the line, and who funds the gap until then?

We use 4 steps, in the order a deal should be built.

1. Measure decay. Find out how fast your referred players lose value each month.
2. Price in the same units. Value each model per player at months 3, 12 and 24.
3. Find the crossover. Locate the month where revenue share overtakes CPA.
4. Attach a review point. Re-test the deal at day 30 and month 6.

The steps run from data to contract, with measurement first because every later step depends on it. Skip step one and you are choosing a model on opinion.

Need the Break-Even Window run on your own player data? Talk to AP Collective about its user acquisition service.

Step One: How Do You Measure Player Decay Before Pricing a Deal?

Measure decay by tracking each referred cohort’s monthly net revenue for at least 6 months, and check the 30-day second-deposit rate first because it is the earliest signal. One vendor suggests that partners whose players exceed a 40% second-deposit rate are strong revenue share candidates.

Define decay as the share of a cohort’s monthly revenue lost each month. As a rough guide, one calculator groups traffic into fast-churn casual players at around 15% monthly decay, a median mix at 8% to 10%, and sticky VIP-heavy traffic at 4% to 6%. Treat those as starting points. Your cohorts will differ by product, market and traffic source.

Cut cohorts by partner, not only by month. A partner’s January depositors tell you how that partner’s players behave. A blended average hides it, and blended averages are how operators end up paying the same rate to a partner whose players vanish and one whose players stay.

For crypto-native platforms the same cut works with different events: funded account, first trade, and a second trade within 30 days.

Step Two: How Do You Price All Three Deals in the Same Units?

Price the three deals by calculating what each pays the partner per player at months 3, 12 and 24 under one retention assumption. Comparing percentages with fixed fees tells you nothing until they share a unit.

Here is an illustrative model. The numbers are invented to show the mechanics and are not AP client data. Assume a player produces $80 of NGR in month one, already net of bonuses, and that revenue decays at a constant monthly rate. Compare a $250 CPA, a 35% revenue share and a hybrid of $100 plus 20% revenue share, each paid over 24 months.

Read the same table from the operator’s side. CPA takes 48%, 29% and 20% of 24 month NGR across the three profiles. Revenue share takes a flat 35%. The hybrid takes 39%, 32% and 28%.

The model leaves out bonus abuse, negative carryover, chargebacks and taxes. Add your own costs before you sign anything.

Step Three: Where Does Revenue Share Overtake CPA?

Revenue share overtakes a $250 CPA at month 11 for sticky traffic and month 16 for median traffic, and it never does within 24 months for fast-churn traffic in the model above. The crossover month is the number that decides the deal.

Cash timing shows the other side. At month 3 the partner has received $250 under CPA, about $72 to $81 under 35% revenue share, and $141 to $146 under the hybrid. The hybrid pays early, which is why partners with little cash cushion like it.

Published guidance gives a different crossover. One vendor says that for most programs the crossover between revenue share and CPA falls between month 3 and month 5. Our model gives 11 to 16 months. Neither is an error. The gap comes from the assumed CPA size, first-month revenue and decay, and it is the strongest argument for running your own numbers instead of borrowing a rule of thumb.

Whichever number you land on, state it in the deal. If the crossover is month 16 and the partner cannot wait that long, the partner should take CPA or a hybrid, and the operator should expect to pay more up front.

Step Four: What Review Point Should a Deal Carry?

Every deal should carry a review at day 30 and at month 6, checking second-deposit rate and cohort revenue against the assumption the price was built on. A deal priced on 8% decay that runs at 15% is a different deal.

Day 30 checks early quality: second-deposit rate, bonus-only players, and qualified versus unqualified first deposits. Month 6 checks the curve: has the cohort tracked the assumption, and where is the crossover now?

Build the review into rules. For CPA, set qualification thresholds. Vendors describe minimum deposits of $20 to $50 or a wagering requirement before the partner is credited. For revenue share, agree what happens when a partner’s cohort beats or misses the curve: tier up, tier down or renegotiate. For a hybrid, decide whether the revenue share has an end date.

Keep one ratio on the dashboard. One vendor suggests a healthy program targets 2.5x to 4x player lifetime value to CPA at month 12. Below 1x, the operator loses money on every player. Between 1x and 2.5x, it is buying growth on purpose, and it should say so.

Paying creators or KOLs per player? Ask AP Collective about its influencer marketing service with the review points built into the brief.

When Does CPA Win?

CPA wins when traffic churns fast, when the partner needs cash now, or when the operator needs a predictable acquisition cost. In the model, CPA is the best deal for the partner on fast-churn traffic ($250 against $183 for revenue share) and the most expensive for the operator on that traffic, at 48% of 24-month NGR.

Price a CPA from lifetime value. One vendor suggests setting CPA at 30% to 50% of average player lifetime value, and warns that a $200 CPA on players worth $120 loses money on every acquisition. At the 8% median decay in our model, 24-month NGR is $865, so a $250 CPA sits at 29%, inside that guidance.

CPA also suits new markets, where nobody knows the retention curve yet. A fixed fee caps the operator’s exposure to a number it can forecast, and the operator learns the curve while it pays for volume.

The risk is low-quality traffic and bonus hunters. Set qualification rules before the first payment, and agree clawbacks for players who fail them.

When Does Revenue Share Win?

Revenue share wins when players stay, when the partner can wait, and when the operator wants the partner to care about quality. In the model it pays the partner $303 on median traffic and $437 on sticky traffic, which beats a $250 CPA from month 16 and month 11 respectively.

For the operator, revenue share means cost follows revenue. If a cohort fails, the operator pays little. If it succeeds, the operator pays 35% of a large number, which is the price of a partner who sends better players.

Two traps come with it. The first is the base. GGR and NGR differ, and a vague definition gives the operator room to deduct costs the partner never agreed to. The second is negative carryover. One vendor notes that large jackpot wins can create months of negative carryover in casino, so a partner can earn nothing for months after a single big winner.

Partners should ask whether negative months carry forward and for how long. Operators should state it plainly. A deal that hides these terms usually ends in a dispute.

When Does a Hybrid Make Sense?

A hybrid makes sense when the partner needs early cash and the operator wants the partner to keep caring about retention. In the model, the $100 plus 20% hybrid pays $141 to $146 by month 3, against $72 to $81 for pure revenue share, and it beats a 250CPAat24monthsonmedianandstickytraffic(273 and $350).

It also costs the operator less than CPA on churning traffic: 39% of 24-month NGR against 48%. On sticky traffic it costs more, 28% against 20%, which is the price of sharing the upside.

Hybrids fail in 3 ways. The fixed part is set so high that the revenue share is decoration. The revenue share has no end date, so the operator pays on a cohort for years. Or the terms are so flexible that two partners with the same traffic hold different deals. One vendor guide notes that hybrid terms are typically negotiable, which is true and is also why they need a written rule.

Set the fixed fee to cover the partner’s acquisition cost and the revenue share to reward retention. If the fixed fee alone covers more than the partner’s cost, you have rebuilt CPA with extra steps.

What Contract Terms Protect You on Each Model?

Five terms matter on every model: the revenue base, bonus and negative carryover treatment, the attribution window, the qualification rules and audit rights. Get all five in writing.

• Revenue base. Name NGR and define every deduction. Without that, a percentage means nothing.
• Bonuses and carryover. State whether bonus costs reduce the base, and whether negative months carry forward and for how long.
• Attribution window. Vendors describe 30-day windows for CPA and lifetime attribution for revenue share. Agree what happens when a dormant player returns through another partner.
• Qualification rules. Minimum deposit, wagering and the time allowed to qualify, plus clawback rules.
• Audit rights. The partner should be able to check the revenue figures, and the operator should be able to check traffic quality.

For crypto-native platforms, swap the events. A funded account and a first trade replace the first deposit, and fee revenue replaces NGR. The five terms do not change.

Want to know what comparable platforms pay before you sign? Ask AP Collective about its competitive intelligence service.

How Do You Track Players Back to the Deal?

Track players by passing a partner ID through registration, storing it on the player record, and joining deposits, second deposits and revenue to it. Every step in this guide depends on that join.

Capture first-touch and last-touch partner IDs at registration and key every later event to the account. Track 4 things per partner: first deposit, second deposit within 30 days, cumulative NGR at months 3, 6 and 12, and churn date. Our guide to {LINK: crypto marketing attribution} covers the capture steps.

Two failure points deserve attention. Identity verification often happens on a third-party page, so pass your own player ID through it and back. And partner-reported numbers will not match yours, so reconcile weekly and agree which system is the record.

Show each partner’s players as a curve, not a total. One vendor’s example puts two partners on a $200 CPA, one returning 1.31x its cost and the other 0.31x, although both look productive on first-deposit volume alone.

What Should You Watch for When Agencies and KOLs Quote a Deal?

Watch for quotes that move the risk onto you without saying so. Agencies and creators usually quote a flat fee, a per-player fee or a revenue share, and each carries the same trade-off as an affiliate deal.

A flat fee pays for activity whatever the result. A per-player fee pays for qualified players, and the definition of qualified is the whole negotiation. A revenue share pays for retention and needs data access to verify it.

Ask 4 questions. What event triggers payment? Who owns the tracking data? What happens to payment if players churn? Does the partner have a stake in quality, or only in volume? A quote that cannot answer all four is a quote for volume.

Creator deals add one more issue: disclosure. Many markets require paid promotion to be labelled, and gambling promotion carries extra rules. Ask your counsel, and make the creator’s contract name the rules they follow.

What Compliance Limits Apply to These Deals?

Compliance limits depend on the market, and for gambling they can decide whether a model is allowed at all. Run 3 checks before the first payment: the partner’s licence or registration status where one is required, the rules on what the promotion may say, and the player protection duties that follow the player.

Some jurisdictions restrict or prohibit affiliate promotion, limit how bonuses and offers can be described, or attach responsible gambling duties to marketing partners. A revenue share can raise extra questions in some markets, because the partner earns more from players who lose more.

Have counsel review the model for each market. Keep records of who promoted what, when, and under which terms. We are not lawyers, and this section is a prompt for questions, not advice.

How Does AP Collective Approach Deal Economics?

AP Collective’s position is that a deal should be priced from measured retention and re-tested on a schedule, whatever the model. Our published work is in crypto growth, and two case studies show the kind of paying-user problem this guide addresses.

For MEXC, AP ran a micro-influencer and UGC operation for a leading exchange, activating 380+ creators across selected markets and reporting 2.6M+ KOL impressions. For Sport.fun, AP supported the go-to-market for a consumer app on Base that reports $123M+ in trading volume and 20K+ unique paying players, alongside a token sale that closed at $10M with 3.38x oversubscription.

Neither case study discloses commission structures or payback data, so they show execution, not deal economics.

One inconvenient fact. The ranges in this guide come from vendors that sell tracking software, and your own cohort data will beat any of them. If you cannot measure retention by partner, fix that before you negotiate.

Planning a launch with partner deals in the mix?

Book a call with AP Collective about go-to-market strategy and bring your retention data.

Frequently Asked Questions

What is the difference between CPA and revenue share?

CPA pays a fixed fee when a referred player completes a qualifying action, usually a first deposit. Revenue share pays a percentage of the player’s net revenue for as long as the player stays active. CPA shifts retention risk to the operator, and revenue share shifts it to the partner.

What is a hybrid commission deal?

A hybrid pays a smaller fixed fee up front plus a smaller revenue share. In our illustrative model, $100 plus 20% revenue share pays $141 to $146 by month 3, against $72 to $81 for a 35% revenue share, so it gives the partner early cash while keeping a stake in retention.

Which model is best for a new operator?

CPA is usually the safer start, because nobody knows the retention curve yet and a fixed fee caps exposure to a number you can forecast. Measure cohorts for at least 6 months, then test revenue share or a hybrid with partners whose players show a strong 30-day second-deposit rate.

What is NGR and why does it matter?

Net gaming revenue is gross gaming revenue minus direct costs such as bonuses. It is the base for most revenue share deals. Two operators offering 35% can pay very different amounts if they deduct different costs, so the contract should name every deduction.

How long does revenue share take to beat CPA?

It depends on your numbers. One vendor says the crossover usually falls between month 3 and month 5. In our illustrative model, with a $250 CPA, it falls at month 11 for sticky traffic, month 16 for median traffic, and not within 24 months for fast-churn traffic.

How much should CPA be relative to player value?

One vendor suggests 30% to 50% of average player lifetime value, and a 2.5x to 4x lifetime value to CPA ratio at month 12. Treat both as starting points and test them against your own cohorts.

What is negative carryover?

Negative carryover means a partner’s losing months, such as a big player win, are carried forward and offset against later revenue before any commission is paid. It can leave a partner with no earnings for months. Agree whether it applies and how long it lasts before the first payment.

Can crypto platforms use the same deal models?

Yes. Exchanges, crypto casinos and prediction markets can pay per funded account or first trade, a share of fee revenue, or a hybrid. The Break-Even Window works the same way, with funded-account and trade events in place of first deposits and NGR.

How do you stop bonus hunters in a CPA deal?

Set qualification rules before the first payment: a minimum deposit, a wagering or trading requirement, and a time limit to qualify. Add clawbacks for players who fail them. Vendors describe minimum deposits of $20 to $50 as common starting points.

How do you compare agency and KOL quotes?

Compare what triggers payment, who owns the tracking data, what happens if players churn, and whether the partner has a stake in quality. Price each quote per funded or qualified player at months 3, 12 and 24, so flat fees, per-player fees and revenue shares share one unit.

Final Takeaway

A deal chosen from a rate card gets renegotiated at the first bad quarter, usually by whoever feels the loss first. A deal priced from measured retention, with a crossover month written into it and a review at day 30 and month 6, survives the quarter. The model takes a day to build once the cohort data exists, and it decides who funds the next 16 months.

The One Page Action Spine

Reviewed periodically. If you spot something outdated, write to info@apcollective.io.