iGaming Affiliate ROAS and CPA Benchmarks: What Good Actually Looks Like

Every few weeks someone asks the same question in a media-buying chat: "What's a good ROAS for iGaming affiliate traffic?" The answers that come back are confident, specific, and useless. Someone says 150%. Someone says 3x. Someone says they cut anything under 200% on day one. None of them are lying. They are just describing their own book, on their own geos and offers, and quietly presenting it as a law of nature.

Here is the uncomfortable truth this post is built on: there is no single iGaming affiliate ROAS benchmark that means anything. The number swings so hard across geo, format, offer, and player lifetime value that a headline average is closer to noise than a target. The same is true of "typical CPA per FTD." What actually exists is a method for setting your own targets from your own data, and that is what good looks like.

This guide walks through why blended benchmarks mislead, why Day-1 ROAS looks alarming and usually should not, how to anchor on cost per FTD, and how to derive a target ROAS or target CPA from your own LTV curve. If you want the definitions underneath these metrics first, our guide to iGaming ad performance metrics lays out cost per FTD, ROAS, LTV, CR, and EPC in full.

A quick note before we start. This is written for advertisers and media buyers, the people buying traffic to acquire depositing users. It is not player-facing, and nothing here is advice on how to gamble. Run your campaigns inside the rules: respect age-gating, geo-restrictions, and responsible gambling requirements in every market you touch.

Why There Is No Average ROAS Worth Chasing

Start with why the question itself is the wrong shape. ROAS is revenue divided by ad spend. In iGaming, both halves of that fraction move for reasons that have nothing to do with how well you are running the campaign.

The revenue half depends on player lifetime value, and LTV is wildly different across markets. A tier-1 European player on a premium casino offer and a tier-3 Southeast Asian player on a low-deposit sportsbook offer generate revenue on completely different curves. The spend half depends on format and geo competition, which set what you pay for a click or an impression. Push traffic in one market and native in another are not priced remotely alike.

So a single "average ROAS" figure is an average across all of that variance at once. It blends premium and budget offers, cheap and expensive geos, wide-funnel and narrow-funnel formats, and short and long measurement windows into one number that describes none of them. Optimizing toward it is like setting your thermostat to the average temperature of the country.

The discipline here is the same one we apply to every metric: treat any single quoted industry-average ROAS or average cost per FTD figure with suspicion. Your own historical performance, on the same geo-format-offer combination, read over a consistent window, is the only benchmark that pays rent. Everything else is someone else's book dressed up as a rule.

What Does Vary Predictably

Ranges do exist, they just do not collapse into a point. What you can say honestly is directional. Higher-LTV markets support higher acquisition costs and can tolerate a lower short-window ROAS because the revenue tail is longer. Cheaper, higher-volume formats tend to post lower cost per click but a wider, noisier funnel. Competitive tier-1 geos cost more to buy into than emerging markets. These are shapes, not numbers, and shapes are what you plan around. The moment someone hands you a precise cross-market average, they have thrown away the only information that mattered.

Why Day-1 ROAS Looks Low and Usually Should

The single most common way buyers misread iGaming ROAS is by measuring it too early. It is worth understanding exactly why the early number is structurally ugly, because once you see the mechanism you stop panicking at it.

When you acquire a depositing player, you pay the entire acquisition cost up front. The player, at that point, has made exactly one deposit. Their revenue accrues over the following weeks and months as they come back and deposit again. So on day one you are holding all of the cost and almost none of the revenue. A Day-1 ROAS of 0.4 is not a dead campaign. It is a campaign whose returns have not arrived yet, which in this vertical is the normal state of a healthy campaign on its first day.

This is why the measurement window is not a detail, it is the whole answer. The same campaign can read as a disaster at 24 hours, break even at day 30, and be clearly profitable at day 90. Nothing about the campaign changed. Only the slice of the LTV curve you were looking at changed.

The practical consequence is that you cannot read ROAS honestly without a view of lifetime value. You judge a campaign against the shape of your own LTV curve: how much of a cohort's eventual value typically lands by day 7, day 30, day 90. If your Day-30 ROAS is tracking the curve you expect from profitable historical cohorts, the campaign is fine even if the raw number is still under 1.0. If it is falling behind that curve, you have a real problem. The curve is the benchmark, not the calendar date. Our FTD playbook from registration to revenue walks the full path a player takes across that curve.

Blended Benchmarks Hide the Winners and the Losers

Even inside your own account, an averaged number lies to you. This is the trap that survives even after you stop chasing industry figures.

Suppose your account-wide cost per FTD is a comfortable-looking figure and your blended ROAS clears your floor. That blend can be one push zone delivering deposits cheaply, subsidizing a native zone bleeding money on expensive, low-LTV players. On the blend, everything looks healthy. Underneath, you are funding a loser with a winner's margin and calling the average a success.

The fix is to refuse the blend. Break ROAS and cost per FTD out by geo, format, and traffic zone before you trust any figure, because those are the levels you actually act on. A zone-level view tells you which sources to scale and which to cut. A blended view tells you nothing you can act on, while feeling reassuring, which is the worst combination a dashboard can offer.

This is also where two zones that look identical on cost per FTD turn out to be nothing alike. Both might deliver deposits at the same price, but if one sends players with double the 90-day LTV, they are not the same zone: one scales, one drains. Raw cost per FTD will never surface that difference. Only LTV layered on top, per source, will.

Anchor on Cost per FTD, Watch Reg-to-FTD as the Leading Indicator

If ROAS is the honest final verdict and it takes weeks to read, you need something you can steer on now. That anchor is cost per FTD, with the registration-to-FTD ratio as your early warning light.

Cost per FTD is your ad spend divided by the number of first deposits it produced. It is the cleanest near-term control because you can read it before the long-run revenue is in. You set a maximum you are willing to pay per FTD (more on where that number comes from below), and you manage sources against it day to day. It caps acquisition cost directly, which is exactly what you need while ROAS is still maturing.

The registration-to-FTD ratio is your leading indicator. Registrations happen fast; deposits trail them. If a source registers users well but converts them to FTDs poorly relative to your other sources, that gap is telling you something is wrong upstream, at the offer, the deposit flow, or the traffic quality, before your cost per FTD fully reflects it. Reading reg-to-FTD early lets you catch a souring source days before the deposit lag makes it obvious. Our post on lowering cost per FTD goes deeper on managing that number in practice.

The Range We Actually Work In

Since the whole point of this post is that no universal number is worth much, here is ours, with the caveats attached rather than filed off. On offers that convert well, the cost per FTD we optimize toward sits in roughly the $25 to $75 range. Below about $25 per FTD is not realistic in this vertical, and any network that promises it should worry you more than reassure you. Above the range usually means the funnel, not the traffic, is the problem: a weak landing page or a clumsy registration and deposit flow can push cost per FTD to $100 to $200 on the same traffic that would have hit the lower band with a tighter funnel. That is not a disclaimer, it is where the work is. The final number is set as much by your offer and your landing page as by the media, which is why we optimize placements and zones and adjust bids on real FTD data rather than promising a figure up front.

What we can point to concretely is delivery, because impressions, clicks, click-through rate, and cost per click are ratios that do not depend on the size of your budget. A recent push campaign in Turkey delivered about 5.4 million impressions and roughly 14,000 clicks in around two weeks, at close to $0.03 per click and a 0.39% click-through rate. That is one geo and one format, and it does not transfer to Indonesia, Brazil, or to native inventory, but it is a real example rather than a projection. Across our network projects since 2024 we have delivered over 1.2 billion ad impressions and managed more than $800,000 in advertiser spend, mostly on onclick and popunder alongside push and banner.

Notice what those numbers do not include: a ROAS or a conversion rate. That is deliberate, and it is the honest part. Older campaigns in this space frequently ran without server-side conversion tracking wired in, so a quoted historical ROAS or CR is often a guess dressed up as a result. We would rather tell you the delivery is real and the cost side is honest, and then get you a ROAS you can actually trust the only way that works: a small test on your own offer, with S2S postback firing from the first click, measured on your own numbers.

Setting a Target CPA and a Target ROAS From Your Own LTV

Here is the part that turns all of this from diagnosis into a plan. Your targets are not looked up, they are derived, and they are derived from your own lifetime value.

Target CPA: the most you will pay per FTD

Your target CPA is a ceiling on cost per FTD, and it comes straight out of LTV and margin. Estimate the net revenue a player generates over your measurement window, apply your margin, and decide how much of that you are willing to spend to acquire the player while still hitting your profitability goal. That figure is your maximum cost per FTD. Anything acquired above it is unprofitable by your own definition; anything below it has room to scale.

The honesty check is the LTV input. Early LTV projections lean on retention you cannot yet observe, and bonus-hunting users can inflate apparent early activity that never becomes real value. Keep the estimate conservative until you have genuine 60 to 90 day cohort data. A target CPA built on optimistic LTV is a license to overspend that only reveals itself months later.

Target ROAS: the revenue-to-spend goal over your window

Your target ROAS is the revenue-to-spend ratio you need over a defined measurement window. It is the more complete truth, because it ties directly to money in versus money out. But it depends on LTV data to be meaningful, and it reads late, because the revenue accrues over weeks. You set a target ROAS at, say, day 30 or day 90, informed by what your profitable historical cohorts have done by that point.

The two targets are the same goal expressed at two different distances. Your target ROAS over your LTV window defines what profitable looks like; your target CPA is the near-term proxy that keeps you inside it while the revenue is still arriving. You steer daily on target CPA and reconcile against target ROAS as the cohort matures.

The tradeoff, stated plainly

Neither target is strictly better, they control different things:

Target CPA gives you a lever you can pull today at the cost of assuming your LTV estimate holds. Target ROAS gives you the real answer at the cost of waiting for the revenue to land. Run both: cap acquisition cost with target CPA now, confirm profitability with target ROAS later, and revise the CPA ceiling whenever the cohort data moves your LTV estimate. The choice between CPA-style and revenue-share-style deals feeds directly into this, which our breakdown of CPA vs revshare vs hybrid for gambling covers in depth.

None of This Works Without Real Data

Every target above assumes the numbers feeding it are real. In iGaming, that assumption is not free. Deposits lag clicks by days, client-side pixels get blocked and drop conversions, and a deposit that fires late against the wrong source poisons every benchmark you set.

This is what S2S postback tracking solves. Instead of relying on a browser to report the conversion, your platform sends the FTD and revenue events server-to-server, directly to the ad network, the instant they happen, with the click ID attached. Your cost per FTD and ROAS then reflect actual deposits rather than whatever a browser managed to fire before it was blocked or the user closed the tab. For a vertical where the money arrives days after the click, that server-side timing is the difference between a benchmark you can trust and a number you are guessing at. Taroviser supports S2S postback so your FTD and revenue events land where the optimization actually happens.

The other half of real data is keeping fraud out of it. A zone posting a beautiful cost per FTD on invalid traffic is not a bargain, it is a benchmark built on deposits that will claw back weeks later. Taroviser runs multi-layer anti-fraud, from automated invalid-traffic filtering and bot detection through zone-level scoring to human review, precisely because a corrupted input makes every downstream target meaningless. And Taroviser's continuous AI optimization works toward cost per FTD off the data you wire in, which is only as good as the postback and the fraud filtering underneath it.

What Good Actually Looks Like

Pulling it together: good is not a number you can be handed. Good is a method. You refuse blended and industry-average benchmarks, because they hide the only variance that matters. You read ROAS against your own LTV curve over a real window, not on day one. You anchor day-to-day on cost per FTD, watch reg-to-FTD as your leading indicator, and derive both a target CPA and a target ROAS from your own conservative LTV data. And you make sure the numbers are real with S2S postback and clean, fraud-filtered traffic. A buyer running that loop will beat a buyer chasing someone else's quoted average every time, because they are optimizing toward their own economics instead of an average that describes nobody.

If you want to see where these trends are heading and what to do about them, our companion post on iGaming affiliate marketing trends for 2026 maps the direction of travel for media buyers.

FAQ

What is a good ROAS benchmark for iGaming affiliate campaigns?

There is no single number that means anything across the industry. ROAS varies so widely by geo, format, offer, and player LTV that a headline average is closer to noise than a benchmark. The only benchmark that pays rent is your own historical ROAS on the same geo-format-offer combination, read over a window long enough to reflect how revenue actually accrues. Treat any single quoted average ROAS figure with suspicion and build your target from your own LTV curve instead.

Why is my Day-1 ROAS so low in iGaming?

Because you pay the full acquisition cost up front, but a player's revenue arrives over weeks of repeat deposits. Day-1 or short-window ROAS is structurally low in this vertical and does not mean the campaign is losing money. Judge it against your LTV curve and a realistic measurement window rather than the first day's numbers.

Should I optimize toward target ROAS or target CPA?

Both, at different stages. Target CPA, your maximum cost per FTD, is the cleaner control day to day because it caps acquisition cost directly and you can act on it fast, before long-run revenue is in. Target ROAS ties spend to revenue but only becomes trustworthy once you have enough LTV data to know what a cohort is actually worth. Most buyers run a target CPA in the near term that is derived from a target ROAS goal over their LTV window.

How do I set a target CPA per FTD?

Work backward from lifetime value and margin. Estimate the net revenue a player generates over your measurement window, apply your margin, and the result is the most you can afford to pay per FTD while still hitting your ROAS goal. Keep the LTV input conservative until you have genuine 60 to 90 day cohort data, because early projections lean on retention you cannot yet observe.

Why do blended ROAS and CPA numbers mislead?

A blended average hides the spread. One push zone delivering FTDs cheaply and one native zone delivering them expensively can average out to a healthy-looking number while you keep funding the loser. Segment ROAS and cost per FTD by geo, format, and traffic zone before you trust any figure, because the blend is exactly where profitable and unprofitable sources disappear into each other.

Do I need S2S postback to measure ROAS and CPA accurately?

For iGaming, strongly yes. Client-side pixels get blocked and drop conversions, and they struggle with deposits that lag the click by days. Server-to-server postback fires the real FTD and revenue events from your platform with the click ID attached, so your ROAS and cost per FTD reflect what actually happened rather than what a browser managed to report. Without it, any benchmark you set is built on partial data.

What CPA per FTD is realistic for iGaming campaigns?

On offers that convert well, a workable cost per FTD tends to sit in roughly the $25 to $75 range. Below about $25 per FTD is not realistic in this vertical, and any network promising it should worry you more than reassure you. A weak landing page or a clumsy registration and deposit flow can push it to $100 to $200 on otherwise fine traffic, so the exact figure is driven as much by your offer and funnel as by the media. Treat that range as a starting expectation and confirm the real number with a small test on your own offer using S2S postback.

Set Targets From Your Own Numbers, Not Someone Else's Average

If you are done chasing a benchmark that describes nobody's account, this is the fix: clean S2S postback so your FTD and revenue data is real, reporting granular enough to segment by geo, format, and zone, and an optimization layer that reads actual cost per FTD instead of estimates. Taroviser runs CPM, CPC, and CPA Goal 2.0, plus SmartCPM and SmartCPC, across 200+ geos and five ad formats, with no platform fee and no minimum, fast compliance-serious approvals, and a multi-layer anti-fraud stack behind the data. Talk to Taroviser about wiring your campaigns to measure the numbers that actually set your targets.