CPA vs RevShare vs Hybrid: How Agencies and iGaming Brands Agree on Payouts
The rate is on the table from day one; baseline, hold and carryover only surface in the first report. Here's how the three payout models work and which contract terms to settle before launch.
Contents
- CPA: a fixed fee per player — and the baseline decides everything
- RevShare: a share of NGR, carryover and deal duration
- Hybrid: a fixed fee plus a percentage
- Which model works for agencies and brands — and when
- An illustrative calculation
- What to put in the contract
- How to raise the payout model in negotiations
- FAQ
The payout model is the first thing an agency and an iGaming brand get serious about. The rate comes up on the very first call, but the terms around it — baseline, hold, carryover, payment schedule — often surface only a month later, when the first report lands and the numbers don't match.
Below: how CPA, RevShare and Hybrid deals work, who each model suits and when, and what to put in writing before the first click rather than after the first dispute.
CPA: a fixed fee per player — and the baseline decides everything
CPA (cost per acquisition) is a fixed payout to the agency for every player it brings in — or, more precisely, every qualified player. That word does a lot of work. The agency isn't paid for a sign-up, and not always for any first deposit (FTD), but for a player who has met the baseline.
The baseline is the minimum threshold a player has to reach to count. It can be a first-deposit amount, wagering volume, a number of deposits or active days within a set period, or completed verification. The stricter the baseline, the fewer players qualify from the same traffic.
For the agency, CPA means predictable money on a short cycle: media spend pays back as soon as players clear the hold. For the brand, it's a predictable cost per player — along with the risk of paying for players who deposit once and disappear.
RevShare: a share of NGR, carryover and deal duration
RevShare is the agency's share of the revenue the brand makes from referred players. The percentage is usually calculated on NGR rather than GGR. GGR is player stakes minus winnings. NGR is what's left of GGR after deductions — bonuses, taxes, payment fees and whatever else the contract lists.
There's no single NGR standard: every brand has its own list of deductions. So the same RevShare percentage at two operators can mean very different money. Ask for the list of deductions in writing, before you sign.
Negative carryover
If the NGR from an agency's players goes negative in a given month — say, one player lands a big win — that negative balance may carry over into the following periods. Until it's covered, the agency earns nothing. Amounts already paid usually aren't clawed back, but the carryover eats into new earnings.
There are several options: unlimited carryover, no carryover (every month starts from zero), carryover capped at a set amount, or a reset after an agreed number of months. Which one applies should be written into the contract, not worked out over chat.
RevShare duration
The share may be paid for as long as the player stays active, or for a fixed period from registration. Also pin down what happens to payouts if the partnership ends: does the agency keep its share on players it has already referred, and for how long?
Hybrid: a fixed fee plus a percentage
A hybrid deal combines both models: the agency gets a fixed payout per qualified player plus a share of NGR. The logic is to split the risk — which is why both components of a hybrid are usually lower than in a pure CPA or pure RevShare deal.
Hybrids are easy to get tangled up in. Does carryover apply only to the RevShare component, or can a negative balance reduce CPA payouts too? Is there one baseline for both the fixed fee and the start of the revenue share? Those answers belong in the contract, not in a chat thread.
Which model works for agencies and brands — and when
There's no universally best model. The choice comes down to three things: how confident the agency is in its traffic quality, how confident the brand is in its retention, and how long both sides are prepared to wait for the money.
| Model | Works for the agency when | Works for the brand when | Main risk |
|---|---|---|---|
| CPA | It needs fast cash turnover to fund media buying; its traffic reliably clears the baseline | It wants a predictable cost per player; it's entering a new GEO with no LTV data of its own | Agency: a strict baseline and a long hold. Brand: overpaying for one-and-done players |
| RevShare | It's confident in player quality and can afford to wait for payback | It wants to pay for actual revenue, not just for a deposit | Agency: opaque deductions and carryover. Brand: long-tail payouts on players the product itself retains |
| Hybrid | It needs to cover media costs and still keep a share of long-term revenue | It's willing to share risk but wants to incentivize quality | Complex calculations and disputes over details if the contract is loosely worded |
An illustrative calculation
Let's run the same cohort through all three models. All figures are hypothetical: they're chosen to keep the math simple and are not a benchmark for market rates.
| Model | Month 1 | Month 2 | Month 3 | 3-month total |
|---|---|---|---|---|
| CPA $150 | $15,000 (after hold) | 0 | 0 | $15,000 |
| RevShare 35%, with carryover | $4,200 | 0; −$4,000 carried over | ($9,000 − $4,000) × 35% = $1,750 | $5,950 |
| RevShare 35%, no carryover | $4,200 | 0 | $3,150 | $7,350 |
| Hybrid $70 + 20%, with carryover | $7,000 + $2,400 = $9,400 | 0; −$4,000 carried over | ($9,000 − $4,000) × 20% = $1,000 | $10,400 |
What the example shows: over a short horizon, CPA pays the agency more, and faster. RevShare only catches up if players stick around for months, and a single bad month with carryover noticeably shifts the total. The hybrid smooths out both extremes.
For the brand, the logic is mirrored: under CPA it pays upfront, regardless of what the players bring in later. Under RevShare it pays a share of actual revenue, but over a longer period.
What to put in the contract
Most payout disputes aren't about the rate — they're about definitions. Here's what to agree in writing before launch.
- Qualified player. Minimum deposit, wagering, number of deposits, the window in which they have to happen, and whether KYC is required.
- Hold. How long it lasts, when the clock starts and what gets checked in that time.
- Fraud. What counts as a violation (multi-accounting, bonus abuse, incentivized traffic without approval), what evidence the brand provides and what happens to the payout.
- NGR deductions. The full list: bonuses, taxes, payment fees, chargebacks and everything else.
- Carryover. Whether it applies, whether there's a cap or a reset period, and whether it touches the CPA component of a hybrid.
- RevShare duration and what happens to payouts after the partnership ends.
- Cap. The player limit per period and whether players above the cap get paid.
- Payments. Frequency, currency, exchange rate and conversion date, payment method, minimum payout threshold.
- Reporting. Dashboard access, postbacks, sub-ID-level breakdowns, how often data is reconciled.
- Changes to terms. How many days' notice each side gives before changing the rate or baseline, and whether a new rate applies to players already referred.
- Compliance. The brand's license in the target GEOs, 18+ audiences only, compliance with local laws and ad platform policies.
For how to define traffic quality and the dispute process up front, see Traffic quality and fraud. If any of these terms are new to you, keep our iGaming marketing glossary handy.
How to raise the payout model in negotiations
A good negotiation over the payout model is a trade-off of risks, not haggling over a percentage. A few lines that help get the conversation moving:
For the agency, when a brand offers pure RevShare in a GEO that's new to you:
“We're open to RevShare, but at launch the media spend comes out of our pocket. Let's run a hybrid for the first few months with a fixed fee per qualified player, then revisit the terms based on actual cohort data.”
For the brand, when an agency asks for a high CPA:
“We're happy to discuss the rate if the baseline includes a redeposit and the hold gives us time to check the players. If that doesn't work for you, we'd suggest a hybrid.”
The general rule for both sides: start with a limited test with a clear cap and a fixed date to review terms. That turns an argument about models into a conversation about data.
Avero matches agencies with brands that fit their traffic and GEOs, shortlists the right agency for a brand's task, and supports the negotiations on terms. More on our pages for agencies and for brands.
FAQ
Which model should we choose for a first test with a new partner?
When neither side has shared data yet, it makes sense to start with CPA or a hybrid with a small cap and a pre-agreed review date. That way the agency isn't funding media blind, and the brand isn't taking on a large volume of unproven traffic.
Can we change the payout model after launch?
Yes, if the contract allows it. Fix the review date, the notice period and which players the new model applies to. The cleanest option: the new terms apply only to players referred after the date of the change.
What if the brand won't disclose how NGR is calculated?
Without a list of deductions, RevShare can't be modeled in advance or verified afterwards. Ask for at least the list of deduction categories, without the amounts. If even that isn't available, CPA is the safer option.
Does negative carryover apply to the CPA component of a hybrid deal?
It depends on the contract — both setups exist. So spell it out explicitly, in one sentence: does negative NGR reduce only the revenue-share component, or all payouts to the agency?