Affiliate marketing built iGaming's acquisition channel before streamers or paid social existed, and it still moves real volume today. The problem was never whether affiliates work. It's that most programs get structured around driving traffic instead of driving depositing players — and those are not the same thing, no matter how good the click numbers look on a pitch call.
Picking a payout model
- CPA — a fixed fee per qualified FTD. Predictable cost, easiest to budget, but affiliates chasing volume-based CPA have an incentive to send lower-quality traffic that converts once and never returns.
- Revenue share — a percentage of net gaming revenue from referred players, usually for the player's lifetime. This is what serious, established affiliates actually want, because it rewards them for sending players who stick around, not just players who sign up.
- Hybrid — a smaller CPA plus a lighter revshare, which is where most durable affiliate relationships land, for the same reason hybrid deals work with streamers: it balances near-term cash flow for the affiliate against long-term alignment with the brand.
None of these should be a fixed rate applied uniformly across every affiliate. An affiliate with a proven track record sending high-retention players has earned negotiating leverage a brand-new, unproven partner hasn't — treating both the same either overpays the unproven one or underpays the proven one, and the second mistake is the one that actually costs the program, because that's the partner most likely to walk to a competitor offering better terms.
Vetting affiliates before they go live
Traffic source matters more than traffic volume. An affiliate driving organic SEO traffic to a comparison page targeting your actual licensed markets is worth more than one running paid traffic at scale to a broad, unqualified audience — even if the second one's raw click numbers look better on a pitch call. Before onboarding, check what markets their existing traffic actually comes from, whether their site or channel matches your compliance requirements, and how they've historically reported (affiliates who are vague about their own numbers are a red flag before they've sent a single click).
Ask for a sample of their existing traffic reports from other operators before signing anything. An established affiliate who's run legitimate programs elsewhere will have this ready without hesitation — it's a normal ask in this industry, not an invasive one. Hesitation or evasiveness on this specific request is a stronger red flag than any single metric on a pitch deck, because it usually means the numbers they'd otherwise show off don't hold up to scrutiny.
Contract terms that actually protect the program
Beyond the payout model, the specific terms written into an affiliate agreement determine how much control you keep once traffic starts flowing:
- Cookie duration — how long an affiliate gets credit for a click before it expires. Shorter windows (7-30 days) reduce disputed attribution; longer windows (60-90+ days) are more affiliate-friendly and common with top-tier partners who negotiate for them.
- Minimum deposit thresholds for a qualified FTD — without one, an affiliate can technically hit CPA targets by sending players who deposit the smallest possible amount and never return, which pays out on volume without ever producing a real customer.
- Geo and market restrictions — spelled out explicitly, not assumed. An affiliate sending traffic from a market you're not licensed in isn't a bonus, it's a compliance liability with your name on it.
- Negative carryover clauses — for revshare deals, whether a losing month for the operator (a big payout to players) carries forward and reduces the affiliate's next payout, or resets to zero each month. This one detail changes the real economics of a revshare deal substantially.
Sub-affiliate and master-affiliate networks
As a program scales, some affiliates stop sending traffic directly and start recruiting their own network of smaller affiliates underneath them — a master affiliate structure. The master earns their own standard rate on any players they refer directly, plus an override commission on everything their sub-affiliates generate. This is a legitimate and common way to scale reach fast, because the master affiliate does the recruiting and quality-control work a brand would otherwise have to do itself.
It comes with a real tradeoff: less visibility into where traffic actually originates. A master affiliate's sub-network can include partners the brand never vetted directly, which is exactly where the vetting criteria above (traffic source, market fit, reporting transparency) matter most — applied to the master's aggregate numbers isn't enough; spot-checking sub-affiliate-level data is what catches a bad actor hiding inside an otherwise-performing network.
The practical guidance: don't build a program's tracking infrastructure assuming a flat, single-tier structure if there's any chance it scales into sub-affiliate territory later. Multi-tier commission logic is significantly harder to retrofit onto an existing flat-rate system than to build in from the start, even if the first version of the program launches without a single master affiliate in it.
Tracking and fraud prevention: what to actually check
Affiliate fraud is common enough that tracking has to assume it will be attempted, not treat it as an edge case. The specific checks that catch it before a payout goes out:
- Server-to-server (S2S) postback validation — confirming a conversion event fired from your own backend, not just a client-side pixel an affiliate could spoof or replay.
- Click-to-deposit timing analysis — a real player takes minutes to hours between clicking an affiliate link and depositing. Conversions landing in seconds, in tight clusters, are a strong automation signal.
- Device and IP fingerprinting — the same device or IP range showing up across dozens of "unique" signups is either a shared household (rare at scale) or a bot farm (common).
- FTD-to-signup ratios that look too clean — real traffic has a messy, inconsistent conversion rate. A suspiciously stable, high ratio across every reporting period suggests manufactured conversions rather than organic ones.
- Multiple affiliates claiming credit for the same player — a direct sign of cookie-stuffing or link hijacking, where one affiliate's link silently overwrites another's attribution.
Reconciling this weekly, the same way streamer attribution should be reconciled, catches problems before a full payout cycle goes out on bad data. A monthly cycle means an entire month of bad data compounds before anyone catches it — by which point the money's already gone out.
Scaling the network without diluting quality
The instinct once a few affiliates perform well is to onboard as many new ones as possible. The better move is qualifying new affiliates against the traffic-source and market criteria above before they go live, not after — a smaller network of affiliates who actually convert outperforms a large one padded with low-quality traffic, and it's far easier to manage. This holds even more once sub-affiliate networks enter the picture, since a single bad master affiliate can bring an entire hidden layer of low-quality traffic in at once.
Where this fits in the bigger system
Affiliates are one acquisition channel among several. Streamers, paid ads, and affiliates all feed the same top of funnel, and the brands getting the most out of affiliate marketing are the ones tracking it with the same rigor as every other channel. Not treating it as a separate, looser system just because it's the oldest one.
The channels also overlap in practice more than the org chart suggests. A streamer with an audience and a personal following is functionally running an affiliate-style relationship the moment they're paid on performance instead of a flat sponsorship — the attribution and fraud-prevention discipline covered here applies just as directly to a streamer deal structured on CPA or revshare as it does to a traditional affiliate site. Building one tracking and reconciliation process that covers both, rather than two separate systems that don't talk to each other, is what actually scales as the acquisition mix grows.
FAQ
CPA or revshare — which should a new affiliate program start with?
CPA first, for predictable cost and easier budgeting while you're still vetting who actually sends quality traffic. Move your best-performing affiliates to hybrid or revshare once their retention numbers prove they're worth the longer-term payout.
How do you catch affiliate fraud before it costs you?
Treat fraud as the default assumption, not the exception. Watch for FTD-to-signup ratios that look too clean, deposit patterns that look automated, click-to-deposit timing that's too fast to be human, and multiple affiliates claiming the same player. Reconcile weekly — a monthly cycle means a full month of bad data before you catch it.
Is a bigger affiliate network always better?
No. A smaller network of affiliates who actually convert outperforms a large one padded with low-quality traffic, and it's far easier to manage. Qualify before onboarding, not after — undoing a bad affiliate relationship costs more than never starting it.
What's the biggest risk of a master-affiliate/sub-affiliate structure?
Reduced visibility. A master affiliate's own numbers can look strong in aggregate while a low-quality sub-affiliate hides inside their network, sending traffic that never gets vetted directly. Spot-check sub-affiliate-level data, not just the master's rolled-up totals.
What contract term do most new operators forget to include?
A minimum deposit threshold for a qualified FTD. Without one, an affiliate can hit CPA targets on paper by sending players who deposit the smallest possible amount once and never come back — technically compliant with the contract, worthless to the brand.