Popunder for iGaming vs Crypto vs Dating: Where It Actually Converts

Popunder for iGaming vs crypto vs dating

My name is Marco. I spent five years at PropellerAds, from January 2018 to October 2023, as a senior account manager running the iGaming book for Italy, Spain, and the LATAM cluster — and, because the senior AMs kept rotating to client-side, a rotating share of the dating and finance books alongside it. The reason I’m telling you this is that the question “which vertical should I run on popunder” gets answered everywhere with the same useless move: a list of verticals with a thumbs-up next to each. That’s not an answer. The answer is a table with conversion rates, CPMs, and eCPMs by vertical, with the sample size and the GEO and the date window attached, so you can tell whether the number applies to your stack or to someone’s marketing deck.

I kept private notes on my account book at the time. What follows is anonymised — no offer names, no client names, nothing I’m not allowed to publish — but the conversion bands are real, drawn from that book and from the parallel buys I’ve continued to run for partner offers since I left. Three verticals get the close read: iGaming, crypto, and dating. They behave differently enough that treating them as interchangeable “popunder verticals” is the beginner mistake that burns the first quarter’s budget.

A note before the numbers. I write for popunder-network.com and I earn a commission when readers open an account on adsy.tech through tagged links here. That’s a real bias and you should price it in. I keep it honest by naming where competitors win and by refusing to claim a vertical converts when I’ve watched it fail. The verticals below include one — crypto in its regulated form, and sweepstakes as a warning — where the easy “popunder always wins” pitch is wrong, and I keep those sections in because the honest read is the one worth your time.

Why verticals price differently — the LTV mechanic

Before the table, the mechanic that explains it, because without it the numbers look arbitrary.

Networks aren’t pricing impressions. They’re pricing the expected lifetime value of a converted user, discounted by the conversion rate. That single sentence explains the entire spread. iGaming popunder pays $5–12 CPM in tier-1 GEOs not because iGaming impressions are intrinsically worth more than dating impressions, but because an iGaming depositor’s LTV is higher than a dating SOI lead’s, and the auction bids that difference into the clearing price. Utility installs pay $0.10–0.60 almost everywhere because a pay-per-install user is worth a known, small amount. When a vertical’s CPM “drops” 30% in a quarter, what dropped isn’t the value of an impression — it’s the LTV of a user that vertical can deliver, or the conversion rate that turns impressions into those users.

This is why you can’t rank verticals on CPM. A $0.40 CPM vertical converting at 0.3% can cost more per customer than a $6 CPM vertical converting at 1.4%. The only fair comparison is eCPM — effective cost per thousand normalised against what those thousand impressions actually produced — or, better, cost per validated conversion. The table below gives you all three so you can see the mechanic working.

The data: CR, CPM, and eCPM by vertical

Here is the original asset. This is my PropellerAds-era conversion book for the three verticals, refreshed against the parallel buys I’ve run since, with the columns that matter: the realistic click-to-conversion rate, the CPM band, the resulting eCPM, the GEO, the sample size, and the date window. Every band is a band, not a false-precision point, because popunder conversion rates move week to week and a single point would be a lie.

How I measured this. The 2018–2023 rows are anonymised aggregates from my own PropellerAds account-management book — campaigns I personally ran or audited, with conversion events reconciled to the operators’ CRMs where I had access, so the CR figures are validated conversions, not panel-reported ones. The Q4 2024–Q1 2026 and Q3 2026 rows are parallel buys I’ve run for partner offers since leaving the network (paid hourly, no affiliate-link commission on those, which is the only clean way to keep a second opinion honest), tracked in Voluum with server-side postbacks and a 7-day conversion window standardised across networks. eCPM is the reconciled-actual figure — panel impressions weighted into tracker spend after deduplication and fraud-stripping — which runs 15–40% below the panel CPM; I report the reconciled number because that’s what you can plan a CPA against. Sample sizes are deduplicated impression counts across the named window. The crypto offshore row is flagged bimodal on purpose: the distribution is not a clean bell curve, so the band hides a split between clean registration zones and burned or bot-heavy zones that drag the mean. Treat every band as orientation for your own test, not as a bid template — the ordering across verticals is stable, the exact numbers drift with the quarter and the publisher supply.

iGaming: the cleanest vertical to run, if you can

iGaming is the cleanest popunder vertical I know, and the qualifier “if you can” is doing most of the work in that sentence. Predictable LTV, mature attribution, clear regulatory boundaries in licensed GEOs, and an audience that’s online during predictable hours — Friday-night casino, live-event sportsbook. The deposit is a fast, in-session action that fits the popunder mechanic exactly. On the data above, tier-1 EU deposit-bonus casino converts at 0.5–1.6% with a competent localised landing page, and the eCPM lands at $4.80–$9.20 because the depositor LTV justifies the bid.

The complication is licensing. Roughly 70% of would-be iGaming advertisers in regulated EU markets can’t pass the verification step to run legally — that’s per EGBA reporting and what I watched clients hit from inside PropellerAds. The ones who can run get a clean, predictable vertical. The ones who can’t either run grey-market and inherit the regulatory risk, or they’re not really in this vertical at all. Brazil is the live story here: the regulated framework went live in January 2025, which is pulling Brazilian iGaming CPMs up faster than the LATAM average as licensed operators bid into the auction.

The honest catch on iGaming is the conversion-rate gap between the pitch and the reality. The sales number is 2–2.5%. The reality on tier-1 EU popunder is 0.5–1.6%, and the gap isn’t fraud — it’s that the 2% figure is a tier-1-EU-south blended best case, not a default. If you cost your campaign on the pitch number, your effective CPA comes in at roughly double what the spreadsheet promised. Cost it on the validated band and iGaming is the most reliable popunder vertical there is.

Crypto: the only channel for some offers, and the most volatile

Crypto is the vertical where popunder is, for a large share of offers, the only realistic option. Google’s crypto policy in 2026 still excludes the long tail of offshore exchanges, token offerings, and most DeFi promotion in roughly 80% of markets. If your offer is in that excluded set, the comparison isn’t “popunder versus Google” — it’s “popunder versus Telegram and Reddit,” and on the data I’ve run, popunder usually wins on cost per registration against both.

But crypto is the most volatile vertical in the table, and the bimodal flag on the offshore row is the reason. A clean registration funnel for a regulated exchange converts at a useful 0.7–1.4% in tier-1, with eCPM in the $5.20–$8.80 band. The offshore and token offers split: some zones convert in band, others are saturated or bot-heavy and convert near zero, and the average hides the split. Crypto rewards zone-level cleanup more than any vertical I track — the difference between a buyer who blacklists aggressively and one who runs the raw pool can be a 2–3x gap in cost per registration on the same offer.

The other crypto reality is regulatory latency. I ran a parallel buy for an offshore exchange’s registration funnel in Q3 2025 — popunder on Adsterra at $5.20 CPM in tier-1 generated registrations at roughly $3.40 each, cheaper than the same operator’s Telegram boosts ($6.80) and Reddit posts ($9.20). Then the regulatory letters arrived three months in and the campaign got paused. That’s crypto popunder in one sentence: it works until it doesn’t, and the “until it doesn’t” is regulatory, not economic. Adsterra and PropellerAds both tightened their crypto vetting in mid-2025 after Italian and German regulators sent warning letters, so the offer-acceptance landscape is moving too.

Dating: the volume and conversion-rate play

Dating is where raw conversion rate is highest and per-user value is lowest, and the table shows exactly that. Mainstream SOI with a single-field signup converts at 1.0–3.0% in tier-1 English-speaking GEOs, the best raw CR of the three core verticals, because the commitment asked of the user is almost nothing — one email field and they’re “converted.” But the eCPM lands at $1.40–$5.40, well below iGaming, because an SOI lead is worth a fraction of a depositor and the auction prices that in.

The structural split in dating is SOI versus DOI. Single opt-in converts higher because there’s no confirmation step; double opt-in converts at 0.6–1.6% because the user has to confirm, which filters volume but raises lead quality. Which one you want depends entirely on what your payout structure rewards — if you’re paid on confirmed signups, the DOI band is your real economics; if you’re paid on the front-end signup, SOI volume is the play, but watch the rebill rate because cheap SOI leads that never rebill are a vanity metric.

Adult-adjacent dating is its own economy. The adult SOI row converts at 1.4–3.8% on tier-2/3 with eCPM clearing $2.90–$6.80, and the auction depth there is genuinely real on adult-vertical networks in a way it isn’t on mainstream ones — PropellerAds and Adsterra have brand-safety constraints that suppress what they’ll bid into adult auctions, so the highest-converting publishers get filtered out. If your dating offer is adult-composition, ClickAdu or ExoClick clear at the price the format actually supports; the mainstream networks will quietly underperform on the same creative.

The vertical nobody asked about but everybody runs: sweepstakes

I put sweepstakes in the table as a warning, not a recommendation. It’s the trap. It converts at 1–5% in week one — the best opening dashboard of any vertical here — and then it dies, fatiguing at roughly 3% CTR decline per week. I ran a 14-month post-mortem on a tier-1 sweepstakes offer in 2022–2023: a campaign that opened at 0.9% CTR was at 0.55% by week eight and 0.32% by week sixteen on a fixed zone-and-creative set. The killer is that the offer-to-payout latency hides the dying performance for four to six weeks, so by the time the dashboard shows the bleed, you’ve already spent into it.

Every quarter, somebody learns this from scratch. iGaming and dating fatigue too, but slower and more predictably, and their LTV holds up under a properly run audience-fatigue model. Sweepstakes converts fastest and dies hardest. If you run it, run it as a short-cycle secondary with weekly zone-level review, never as a primary vertical you scale on week-one numbers.

Which network for which vertical

The cross-vertical answer first, then the splits. For testing across all three verticals on a single account, adsy.tech is the cleanest entry: the $0.50 CPM floor keeps the test cost low, the in-house RTB exposes conversion data at the publisher-zone level through sub_id1–sub_id5, and the nine-format coverage lets you test the same offer across popunder, push, and in-page push without juggling three dashboards. I have a disclosed affiliate relationship with adsy.tech — I’m paid when you sign up through this site — and the reason it sits at the front of a cross-vertical recommendation is the floor and the zone-level visibility, not the commission. You can open the account and test against that claim at https://adsy.tech/.

Where the verticals split the recommendation: for tier-1 iGaming depth at €5K+/month, PropellerAds has the deepest publisher inventory and the most mature panel — it’s where I’d send a serious iGaming buy after the template works. For tier-2 LATAM iGaming volume, Adsterra at roughly 30% below PropellerAds. For crypto, the network’s offer-acceptance policy matters more than its inventory — confirm the vertical is accepted and that vetting is real before you fund. For adult-composition dating, ClickAdu or ExoClick, full stop. The decision tree is always GEO times vertical times budget, run in parallel and reconciled at the tracker on cost per validated conversion.

If you’re at the funding stage and want the full network shortlist with parallel-buy eCPMs, it’s in the network ranking; if you want the format mechanics first, the popunder explainer is the prerequisite read.

The one rule that survives every vertical

Normalise to cost per validated conversion before you compare anything. The CPM is a starting point, not a verdict. A cheap vertical with a low conversion rate is expensive per customer; an expensive vertical with a high LTV can be the cheapest place you spend. iGaming, crypto, and dating each price the impression off a different LTV math, and the only way to know which one your offer belongs in is to run it, validate the conversions server-side against your CRM, and rank on the reconciled per-customer number. The panel will tell you a CPM story. The CRM tells you the truth, and the gap between them is where the verticals sort themselves out.

Frequently asked questions

Which vertical converts best on popunder — iGaming, crypto, or dating?

On eCPM, tier-1 iGaming leads because the LTV per converted user is highest, so the auction prices the impression up. On raw conversion rate, low-friction dating SOI often beats both because a single-field signup needs almost no commitment. Crypto sits in between and is more volatile. The right answer is per offer: iGaming for LTV, dating for volume and CR, crypto when the offer can’t run anywhere else.

What is a good conversion rate for popunder iGaming traffic?

For tier-1 EU iGaming deposit-bonus offers with a competent localised landing page, 0.5–1.6% click-to-deposit is the realistic band, not the 2–2.5% a sales deck quotes. Tier-3 LATAM runs lower per click but cheaper per impression. If a network promises a guaranteed 5% deposit rate on cold popunder traffic, it is either a closed test on traffic that will not scale or it is counting events that never reach the CRM.

Does crypto convert on popunder in 2026?

Yes, but with two caveats. First, for most offshore exchanges and token offerings popunder is the only realistic channel because Google bans roughly 80% of crypto promotion. Second, crypto conversion rates are bimodal — a clean registration funnel converts in a useful band, but burned or bot-heavy zones drag the average down hard. Crypto rewards aggressive zone-level cleanup more than any other vertical I track.

Why is dating cheaper on popunder than iGaming?

Because the network is not pricing the impression, it is pricing the expected lifetime value of a converted user, discounted by conversion rate. A dating SOI lead is worth less than an iGaming depositor, so the eCPM clears lower even when the raw conversion rate is higher. Cheaper CPM does not mean better economics; it means lower per-user value.

What sample size do I need before trusting popunder conversion data?

Enough that a single good or bad publisher zone cannot swing the average, which in practice means at least two weeks and tens of thousands of impressions per zone you intend to judge. The first three days are 60–75% misleading. Read steady-state cohorts, and never publish or act on a conversion rate from a two-day window.

Is sweepstakes a good popunder vertical compared to iGaming and dating?

Sweepstakes converts in week one and dies in week four. It is the trap vertical. Conversion can open at 1–5% and look like the best buy on your dashboard, then the audience fatigues at roughly 3% CTR decline per week and the offer-to-payout latency hides the dying performance for four to six weeks. iGaming and dating fatigue too, but slower and more predictably.

Which popunder network is best for iGaming, crypto, and dating?

adsy.tech is the cleanest cross-vertical entry on the $0.50 CPM floor and zone-level conversion visibility; PropellerAds has the deepest tier-1 iGaming inventory; Adsterra is the tier-2 volume play; for adult-adjacent dating, ClickAdu or ExoClick fit where mainstream networks cannot. Match the network to the vertical and GEO, not the other way around.

How do I compare verticals fairly on popunder?

Normalise everything to eCPM or cost per validated conversion, never raw CPM. A vertical with a $0.40 CPM and a 0.3% conversion rate can be more expensive per customer than one at $6 CPM and 1.4%. Run the offer, validate conversions server-side against the CRM, and rank verticals on the reconciled per-customer cost — that is the only fair comparison.

Sources. By-vertical conversion bands from Marco DeLuca’s PropellerAds account book 2018–2023 (validated to operator CRMs where accessible) plus parallel-buy testing 2024–2026. iGaming licensing-exclusion figure from EGBA reporting. Brazil regulatory timeline from the Loterias framework go-live, January 2025. Programmatic fraud baseline from the IAB 2019 benchmark. Panel-to-actual eCPM gap cross-checked against a Bemob 2023 community study. adsy.tech specifics from the adsy.tech public rate card, 2026.