Dating Offers: How to Buy Traffic That Actually Converts

Dating offers have outlived nearly every trend in paid traffic. The reason is unglamorous: the demand is permanent, the action is small, and the decision takes seconds. Nobody researches whether to click "see who's nearby" the way they research a mattress.

That same shallowness is the trap. Dating is the easiest vertical to get clicks in and one of the easiest to lose money in, because the number your dashboard shows you and the number you get paid on are not the same number. Everything below is about that gap.

You are not paid for what you report

Start here, because it reframes every other decision.

In most dating programs the advertiser reserves the right to scrub leads — to reject the ones they consider worthless. Duplicates. Obvious fakes. Accounts that sign up and never open the app again. This is normal, it is usually written into the terms, and it is not a scam.

But it means your campaign has two conversion counts: the one that fires on the signup form, and the one that survives review. If you optimize toward the first, you are optimizing toward a number nobody intends to pay. Traffic that produces 500 leads at a 40% scrub rate loses to traffic that produces 300 leads at a 5% scrub rate — same spend, worse outcome, and the dashboard says the opposite.

So before you scale anything: find out your scrub rate per source. It is the single most useful figure in this vertical, and it is the one nobody volunteers.

SOI and DOI are different products

Two payout models dominate, and confusing them wrecks your maths.

• SOI (single opt-in) — the lead counts when the form is submitted. Cheap payout, easy conversion, high volume.
• DOI (double opt-in) — the lead counts only after the user confirms by email. Payout is often two or three times SOI.

DOI looks obviously better until you notice that a large share of SOI signups never open the confirmation mail. The honest comparison is payout × confirmation rate, and the confirmation rate belongs to your traffic, not to the offer. Cold pop traffic confirms far worse than a warm native reader — which means the same DOI offer can be excellent on one source and unrunnable on another. Both models are the CPA structure underneath; only the definition of the action moved.

Geo decides more than creative

Dating economics are geographic before they are anything else. A tier-1 lead can be worth many times a tier-3 lead, and the traffic price does not scale the same way — which is exactly the arbitrage, and exactly the trap.

Cheap tier-3 volume tempts everyone. It also has a language problem (your English creative is noise to most of the audience), a device problem (older Android, slow connections, so a heavy landing page never loads), and a scrub problem (fake signups concentrate where leads are cheapest). None of that makes tier 3 unprofitable — it makes it a different business with different creatives, run at a different price. The framing is in traffic tiers.

The practical rule: one geo, one campaign. Do not let five countries share a budget and a bid — they have different payouts, different competition, and different scrub rates. Splitting them is how you find out which one was carrying the others, and it is what targeting is for.

Format fit

Dating converts on impulse, which points at the cheap, high-volume formats:

• Push — the workhorse. An opt-in audience, a notification-shaped message, and an offer that needs no explanation. Cheap enough to test twenty angles in a week.
• Popunder — enormous volume at the lowest cost per visit, and your landing page does all the work. Cold, so lead quality needs watching hardest here.
• Native and teaser — worth their higher click price on content sites, because a reader who arrives from a story is warmer than one who arrives from a background tab. Often the better DOI play.

Device targeting matters more here than in most verticals. Dating is overwhelmingly mobile, and a desktop-shaped funnel will underperform for reasons that have nothing to do with your ad.

The creative honesty problem

Every dating buyer eventually meets the fake message notification: an ad dressed as an unread chat, complete with a badge. It gets an outstanding click-through rate. It is also a good way to lose money.

Think about who clicks it. Someone who believes they have a message. What they meet is a signup form — an entirely different thing from what was promised. They bounce, or they sign up carelessly to see the message, and that is precisely the lead that gets scrubbed. You bought clicks with a promise the page could not keep, and the payout stage is where the bill arrives.

Curiosity is legitimate; a false claim is not. "Three people near you are online tonight" opens a loop the landing page can actually close. "You have 1 unread message" does not. The general principle is in creative testing: judge angles by cost per accepted action, and the misleading ones lose on their own, without any need for a moral argument.

There is a compliance layer too. Creatives are reviewed before they serve, mainstream and adult inventory are not interchangeable, and publishers can block categories they do not want — see brand safety. Trying to sneak adult creative onto mainstream inventory wastes your time and the reviewer's.

Measure the thing you get paid for

All of the above collapses without tracking, and dating makes tracking harder than usual because the money moves after the click.

Put {clickid} in your landing URL, store it with the signup, and fire the server-to-server postback when the action happens — server-side, because a confirmation click in an email client never runs your browser pixel. Then label your traffic so the postback can tell you something: sub-IDs for angle, creative and page, with {zoneid} and {geo} already captured for you.

Now the questions you can answer are the ones that matter: which zone produces leads that survive scrubbing, which angle confirms, which geo pays for itself. That is a blacklist built from evidence rather than suspicion.

And watch the lag. A DOI confirmation can land hours after the click, so judging a campaign at the end of its first day systematically undercounts your best traffic. Give the window time to close before you kill anything.

A sane way to start

1. One geo, one device, one offer. Not a portfolio — a test.
2. Ask for the scrub rate and the confirmation rate up front. If nobody will tell you, that is information too.
3. Fund the test in conversions, not dollars — roughly 10–20× your target cost per action per variant.
4. Run even pacing so you sample real hours instead of whatever was cheapest at 3am.
5. Test angles, not colours. Three honest angles beat thirty button shades.
6. Judge on accepted leads. Scale in 30–50% steps and re-check scrub at each one — it tends to worsen as you reach further down the quality curve.

The honest summary

Dating is not a hard vertical to buy. It is a hard vertical to measure, because the gap between a click and an accepted lead is where all the profit and all the losses live. The buyers who last are not the ones with the cleverest creative — they are the ones who know their scrub rate per zone and act on it.

Create an Adgora account to run dating traffic with your own bids, caps and postbacks, funded in TON or USDT. Terms are on the pricing page, the formats are here, and anything unfamiliar above is defined in the glossary.

Key takeaways

• You are paid on accepted leads, not reported ones — find your scrub rate per source before you scale.
• SOI vs DOI is payout × confirmation rate, and the confirmation rate belongs to your traffic.
• One geo, one campaign — mixed geos hide which country was carrying the budget.
• Push and popunder for volume; native and teaser for lead quality and DOI.
• Misleading creatives fail at the payout stage, not the click stage — the scrub bill arrives later.
• Postback server-side and label with sub-IDs; allow for confirmation lag before judging.