Partner iGamingAsk a question
Menu

Three commission models cover almost every iGaming affiliate contract you will be offered: revenue share, CPA, and a hybrid of the two. Which one pays more is not a matter of opinion. It is arithmetic with four inputs — how much revenue your players generate, how long they stay, what the operator deducts before the split, and what the contract says happens in a losing month. Change any one of those and the answer flips.

This guide builds the maths from the bottom up. Every figure below is illustrative unless it links to a program page on this site, in which case it is what that program publishes. Before any of it applies, one frame: everything here assumes traffic sent to operators licensed in the market you are advertising into — the UK Gambling Commission's licence conditions, Spelinspektionen in Sweden, the GGL in Germany, the DGOJ in Spain — to adults aged 18 or over only, with responsible-gambling messaging on the page. Those obligations reach the affiliate through the operator's licence, and breaching them costs you the account before commission is ever a question.

The chain: stakes, GGR, deductions, NGR, your share

Almost every dispute about affiliate pay comes from one confusion: the percentage in your contract is a percentage of net gaming revenue, not of what the player spent. The chain runs like this.

  • Stakes (turnover) — everything the player wagers, including re-wagered winnings. A big number that means almost nothing to you.
  • GGR (gross gaming revenue) — stakes minus player winnings. What the operator actually won. Defined in the glossary entry for GGR.
  • Deductions — everything the operator subtracts before it will share anything.
  • NGR (net gaming revenue) — GGR minus those deductions. Your percentage applies to this, and the definition of NGR varies by contract, which is exactly why you read it.
  • Your share — the rate, applied to NGR, sometimes after one more fee.

What comes off before the split

An illustrative month for one cohort of players, with each deduction named:

LineAmount (EUR)Note
GGR10,000Stakes minus winnings for the cohort
Bonus cost-1,500Bonus money that converted to withdrawable cash
Gaming duty-1,800Tax on gaming revenue in a licensed market; the rate is set by national law and is commonly a double-digit share of GGR
Payment processing-300Deposit and withdrawal fees
Chargebacks and fraud-100Reversed deposits, flagged accounts
NGR6,300The base your percentage applies to
Admin fee at 24%-1,512Only if your contract has one
Commissionable base4,788
Your 35% share1,675.8016.8% of GGR, not 35%

Read the last two rows again. A headline 35% deal, once the standard deductions and a 24% administrative fee are applied, paid 16.8% of what the operator actually won. Nothing here is sharp practice — every line is in the terms. But the number on the landing page and the number in your bank are different numbers, and only one of them is negotiable.

Revenue share: what a 35% deal actually pays over twelve months

Take the cohort above: 100 players who deposit in month one, generating EUR 63 of NGR each per active month. The variable that decides everything is churn. Assume 30% of remaining players stop playing each month — a harsh but not unusual rate for bonus-acquired casino traffic.

Summing the decaying cohort over twelve months gives roughly 329 player-months, so total NGR is about 329 × 63 = EUR 20,700. At 35%, with no admin fee, the affiliate earns about EUR 7,245 across the year. Per player acquired, that is EUR 72. With the 24% admin fee applied, the same traffic pays about EUR 5,506, or EUR 55 per player.

Revenue share has one property nothing else has: it does not stop at month twelve. If a handful of those players are still active in year three, they are still paying you, which is why the phrase lifetime in a contract is worth reading closely — it can mean the player's lifetime, or the lifetime of your account, or until the operator migrates platforms.

CPA: what you are really selling when you take a fixed fee per depositor

CPA pays a fixed amount per qualifying first-time depositor and nothing afterwards. You are selling the player's entire future value for a certain payment today. That is a genuinely good trade when your players churn fast, and a bad one when they do not.

Same cohort, illustrative CPA of EUR 150 per qualifying FTD: 100 players, EUR 15,000, paid within a month or two of acquisition. That is more than double the twelve-month revshare figure at 30% churn — and it arrives immediately, which matters when you are funding content out of your own pocket.

Baseline and qualification criteria — the conditions that void a CPA

A CPA only pays on a qualifying depositor, and the qualification clause is where the money leaks. The conditions that recur:

  • Minimum deposit — commonly EUR 10-20; a player who deposits EUR 5 pays you nothing.
  • Wagering or turnover threshold — the player must stake the deposit a set number of times before you qualify.
  • Baseline volume — you must deliver a minimum number of FTDs per month, or the whole tier drops.
  • Geo restriction — only depositors from named licensed markets count; the rest are simply not paid.
  • Bonus-abuse and fraud exclusions — flagged accounts are removed retroactively, sometimes months later.
  • Duplicate-account rules — one payment per person, per household, per device fingerprint.

Ask for the qualification clause in writing before you agree a rate. A EUR 200 CPA with a 20× wagering qualifier is often worth less than a EUR 120 CPA that pays on deposit.

Hybrid: the structure, and when it is genuinely better than either

A hybrid pays a reduced CPA plus a reduced revenue share — say EUR 75 up front and 15% ongoing. Run it against the same two churn scenarios:

Model30% monthly churn10% monthly churn
Revshare 35%EUR 7,245EUR 15,820
CPA EUR 150EUR 15,000EUR 15,000
Hybrid: EUR 75 + 15%EUR 10,605EUR 14,280

All three columns use the same 100 players and the same EUR 63 of NGR per active month; only retention changes. The pattern is the point: hybrid never wins either scenario. It sits in the middle by construction. You take a hybrid when you do not yet know your retention — which, on a new site with no data, is honest — and you renegotiate to whichever pure model your own numbers later justify. Treating hybrid as the clever default is how affiliates leave money on both sides of the table for years.

Negative carryover: the clause that decides whether revshare pays at all

Gambling revenue is volatile at small player counts. One player hitting a large win can push a whole cohort's monthly NGR below zero. What happens next is decided by a single clause.

Illustrative: your cohort produces EUR 1,800 of NGR in month two, then a EUR 4,000 negative month in month three, then EUR 1,500 a month for the rest of the year.

  • Without carryover (deficit resets monthly): month three pays zero, month four pays 35% of 1,500 = EUR 525, and every month after that pays normally.
  • With carryover: the EUR 4,000 deficit rolls forward. Month four clears 1,500 of it, leaving -2,500. Month five leaves -1,000. Month six clears it and pays 35% of EUR 500 = EUR 175. Three months of real revenue produced EUR 175 instead of EUR 1,575.

That is a EUR 1,400 difference on one bad month, from a clause most affiliates never ask about. Variants matter too: some contracts reset the deficit at the start of each calendar month, some at each quarter, some never; some carry it per brand and some pool it across every brand in the group. Ask which, get it in writing, and read our longer treatment in the post on how negative carryover decides a revshare deal. Some programs state plainly that they do not apply it — 2RBO and PlayAttack both publish exactly that — and the statement is worth more than two extra percentage points on the headline rate.

Admin fees and bundling: two more ways the split moves against you

An administrative fee is a flat percentage taken off NGR before your share is calculated. It is disclosed in the terms and rarely in the marketing. The spread across published terms is real: Chanz Affiliates publishes a 24% administrative fee plus VAT deducted before commission, while FDJ United Affiliates publishes an admin fee of 0%. On the numbers above, that single line is the difference between EUR 7,245 and EUR 5,506 a year on identical traffic — a bigger swing than moving from a 30% to a 35% headline rate.

Bundling is the second one. If a group runs several brands and bundles them into one account, a losing month at brand A is netted against a winning month at brand B before you are paid. Sometimes that helps you; when one brand is structurally unprofitable, it does not. Ask whether brands are reported and paid separately.

The break-even calculation: how many months a player must last

One formula settles the revshare-versus-CPA argument for your traffic:

Break-even months = CPA ÷ (monthly NGR per active player × your revshare rate)

With the figures above: 150 ÷ (63 × 0.35) = 150 ÷ 22.05 = 6.8 months. Your average player must stay active for nearly seven months before revenue share overtakes that CPA. At 30% monthly churn the average active lifetime is about 3.3 months, so CPA wins comfortably. At 10% churn the average lifetime is about 10 months, and revshare wins — and keeps winning into year two, when CPA has paid nothing for a year.

You will not know your churn on day one. You can know it by month six if you are tracking properly, which is the practical argument for taking a hybrid or a short CPA trial first and renegotiating with data. Put your own figures into the commission calculator rather than reusing ours; the answer is sensitive to inputs, which is the whole point of this guide.

Worked comparison: 100 players, three models, twelve months

Collecting everything into one view. Same illustrative cohort throughout: 100 first-time depositors, EUR 63 NGR per active player per month, 30% monthly churn, no admin fee unless stated.

ScenarioYear-1 payoutPer playerStill earning in year 2?
Revshare 35%EUR 7,245EUR 72Yes, small
Revshare 35% with 24% admin feeEUR 5,506EUR 55Yes, smaller
Revshare 35% with negative carryover, one bad monthabout EUR 5,850EUR 59Yes, if the deficit cleared
CPA EUR 150EUR 15,000EUR 150No
Hybrid EUR 75 + 15%EUR 10,605EUR 106Yes, small
Revshare 35%, 10% churnEUR 15,820EUR 158Yes, substantial

One more thing this table cannot show, and it is the most important. All six rows assume 100 real depositing players arrived. It is entirely possible for a year of tracked affiliate clicks to produce almost no depositing players — the traffic is there, the conversions are not, and inspection shows much of the click volume arriving from countries the site does not even serve. Every row above was therefore worth exactly the same to us that year: nothing. Negotiate your rate, but understand that the commission model is a multiplier on a number that starts at zero, and the work that moves that number is audience, not contract terms. If you want to see how the funnel arithmetic behaves before the commission stage, run it through the traffic and revenue estimator.

What to ask a program manager before you sign

Send these as a numbered list and ask for written answers. A manager who answers all ten is a manager worth working with.

  1. What is the exact NGR definition, line by line, including bonus costs and gaming duty?
  2. Is there an administrative fee, and is it applied before or after my percentage?
  3. Does negative carryover apply, and if so does it reset monthly, quarterly, or never?
  4. Are brands reported and paid separately, or bundled?
  5. What are the exact CPA qualification criteria, and when is a payment considered final?
  6. Can qualified conversions be reversed retroactively, and within what window?
  7. What is the cookie duration and the attribution rule — first click or last click?
  8. Is there a sub-affiliate share, at what rate, and for how long?
  9. What is the minimum payout, the payment schedule, and the currency?
  10. Under what conditions does the account go dormant or the rate get reduced?

For calibration, the terms published across our affiliate program directory sit in narrow bands where they are published at all: headline revenue shares mostly land between 20% and 45% with tiered top rates above that; the sub-affiliate shares that are published at all sit at 4-5%; minimum payouts are most often EUR 100, with EUR 50 at the low end and wire thresholds reaching EUR 500 or more; and most programs publish no cookie window, the few that do stating 30 to 60 days or "lifetime". A rate far outside those bands is a question to ask, not a reason to sign.

What we would do this week

  1. Pull the terms of every program you are already in and fill one row per program with six columns: NGR definition, admin fee, carryover, bundling, cookie window, minimum payout. Most affiliates have never seen their own portfolio side by side, and the gaps are usually obvious within an hour.
  2. Calculate your own break-even months using the formula above with your real numbers, or the commission calculator if you do not have them yet. Write the number down; it decides every future negotiation.
  3. Send the ten questions to one manager — ideally at a program you are already sending traffic to — and use the written answers as the template for every future application. If you cannot get an account to ask from yet, start with the guide to getting approved by affiliate programs.

Next in this trackSub-affiliate programs explained