Dating CPA offers: payouts, GEOs and funnels that work

When an affiliate campaign on dating CPA offers loses money, the leak is usually in one of three places. The affiliate picks a payout model that does not fit their traffic, they send every visitor to one offer no matter the country or device, and they write ads that promise something the dating site cannot deliver. Fix those three things and the vertical is one of the steadier ones in adult affiliate work.

The short version: match the payout to how warm your traffic is (cold paid clicks suit lead payouts, search and review traffic suits pay-per-sale and revshare), split campaigns by country and device from day one, use a pre-lander when the traffic is cold, and judge everything on earnings per click after the network has scrubbed bad leads, not on raw conversions. And never imply a real woman nearby is waiting for the user. That is the claim that has drawn regulators.

Below is how each of those pieces works, with the math, the sources, and the parts where we tell you plainly that the numbers vary. If terms like EPC and postback are new, start with what affiliate marketing is and how payouts work, and set up tracking with our guide to affiliate trackers and postbacks before you spend on traffic.

Mainstream, casual and adult dating are different businesses

"Dating offer" covers products that barely resemble each other. Mainstream apps sell relationships and long-term subscriptions. Casual and adult dating sites sell quick contact, usually through credit packs or short memberships. CrakRevenue's own blog defines its dating category as adult dating sites and apps², and describes freemium sites where users buy credits to chat. That is a very different purchase from a Hinge subscription.

The mainstream audience is large and well measured. In a Pew Research Center survey from July 2022, 30% of US adults said they had ever used a dating site or app, rising to 53% of those under 30. Among people who had used one, 46% had used Tinder, and 35% had paid for a dating platform at some point. Men were more likely to have paid than women, 41% against 29%. If you are building a review site or comparison pages, that paying audience is who you are writing for. Our sister site MeetGirls.now (ours) is built on exactly that kind of comparison, and our guide to dating apps for meeting women covers the apps themselves.

Casual and adult offers convert differently. The user wants something fast, the funnel is short, and the money is often in a first purchase. Traffic rules differ too: most mainstream ad platforms will run ads for a relationship app with permission, and almost none will run ads for a hookup site in the same way. Decide which side you are on before you buy a single click, because it decides your traffic sources, your creative, and your payout model.

SOI vs DOI vs PPS vs revshare: which fits your traffic?

Lead payouts (SOI and DOI) suit cold paid clicks; pay-per-sale and revshare suit warm search and review traffic. Compare them on earnings per click after the network scrubs bad leads.

CrakRevenue's help center lays out the payout types, and the definitions are worth knowing exactly, because the small print is where the money goes.

• SOI (single opt-in): you are paid when a user registers with an email address. No confirmation step.
• DOI (double opt-in): you are paid only after the user also clicks the confirmation link sent to their email.
• PPS (pay per sale): you are paid when the user buys a paid registration, such as a trial or membership. The help center notes that token or credit purchases do not count toward PPS.
• RevShare: a percentage of what the user spends, with no expiry while they keep spending.
• Multi-CPA: one offer that can pay on several of these actions, depending on what the user does.

The network's 2023 dating guide put PPS payouts at $20 to $100 or more depending on country and site, and lead payouts in the low single dollars for US DOI at that time. Those figures are old. Treat any payout you read in a blog, including ours, as a hint and check the offer page in your dashboard, at the time of writing (November 2026) and every time you launch.

A worked EPC example (illustrative numbers)

EPC, earnings per click, is the number that lets you compare payout models on the same traffic. The numbers below are invented to show the math. They are not benchmarks and you should not plan a budget on them.

Say you buy 1,000 clicks at $0.20 each, so $200 of traffic, and send them to the same dating offer under four payout models.

At $0.20 a click your break-even EPC is $0.20. SOI loses $40. DOI makes $40, PPS makes $70, and revshare is down $80 in month one but keeps paying if those six buyers keep spending. Two things change the picture fast. If the network rejects 10% of your DOI leads as junk, DOI revenue drops to $216 and the margin nearly halves. And PPS on six sales is a coin toss: one sale more or less moves EPC by $0.045. You need far more than 1,000 clicks before a PPS result means anything.

That is why the rough rule holds. Cheap, cold, high-volume traffic (push, pop) usually suits SOI or DOI, where conversions come quickly enough to optimize on. Warm traffic from search, reviews or your own list suits PPS and revshare, where fewer users buy but each one is worth more. CrakRevenue's own tips post makes the same split, and adds that a lower upfront payout with recurring commission can beat a higher CPA over months. One more caution: the EPC shown on a CrakRevenue offer page is an average over a period, typically three months, based on paid conversions across affiliates. Its help center calls it a benchmark, not a promise. Your traffic is not their average.

Which GEOs and devices pay best for dating offers?

There is no fixed answer: networks group countries into tiers by user value, the cheapest click is rarely the most profitable, and the only reliable number is your own test, one country and one device type per campaign.

Country tiers

Networks and ad platforms group countries into rough tiers by how much their users are worth. There is no official list, and every network draws the lines a little differently, so check how yours defines them. CrakRevenue suggests that Tier 1 users respond to more polished, relationship-style funnels, while Tier 2 and 3 countries often have cheaper traffic and less competition, and may convert better on casual, mobile-first flows. The network is selling you offers in all of them, so test it yourself, but the logic is sound: the cheapest click is not the most profitable one, and an expensive market needs a bigger test budget before its numbers settle.

In practice, run one country per campaign, at least while testing. A campaign that mixes Canada and Brazil will show a blended EPC that is wrong for both.

Mobile vs desktop

StatCounter put mobile at 59.9% of worldwide web traffic in October 2026, with desktop at 38.51% and tablets at 1.59%. For dating, where the product itself is usually an app or a mobile site, the share of your traffic on phones may be higher still. Split mobile and desktop into separate campaigns. They cost different amounts, convert at different rates, and want different pre-landers: short vertical pages on phones, longer story-style pages on desktop, as CrakRevenue's tips post puts it. Android and iOS also pay differently by country, so split those too once you have volume.

Pre-landers, quizzes and smartlinks

Pre-landers and quizzes

A pre-lander is your page between the ad and the offer. It warms a cold visitor up so the offer's sign-up form is not the first thing they see. Common types are a short quiz (age range, what you are looking for, how far you would travel), an advice article, or a plain comparison. Quizzes work because each answer is a small yes, and the sign-up at the end feels like the next step rather than a cold ask. They tend to earn their keep on cold traffic. Visitors who arrive already looking for a dating site may do better with a short, direct page, so test both.

Two rules keep a pre-lander legal and keep your account alive. First, the pre-lander has to describe the offer honestly: CrakRevenue bans "blind leading," which means sending users to an offer without them knowing what they are getting. Second, you cannot skip the offer's own page and post users straight into its form ("direct-to-form"), or edit the offer's landing page without permission. Both are on the network's list of prohibited traffic, and it says violations can end your affiliate status.

Are smartlinks better than direct offers?

For mixed traffic, usually; for one country and device you already know, a direct offer gives you more control.

A smartlink is one link that routes each visitor to whichever offer the network thinks will convert best for their country, device and other signals. Its strength is mixed traffic: a forum post read from 40 countries, or a push campaign you have not split yet. CrakRevenue's cam funnel post describes a useful pattern where a dating smartlink takes the easy conversions first and then rotates non-converting users to cam offers. The same post says dating tends to do better in daytime and cam offers at night.

Direct offers win once you know your numbers. If one offer in one country beats the smartlink's EPC for two weeks, run it directly, write a pre-lander that matches it, and keep the smartlink as the fallback for everything else. Start broad, then narrow down.

Seasonality: January matters

The first Sunday in January is widely reported as the busiest day of the year for dating apps, a date the industry calls "Dating Sunday." The surge is put down to New Year resolutions. For 2026 the date was January 4, and Hinge told Global Dating Insights⁹ that the previous Dating Sunday brought 31.2% more likes and 24.5% more messages than an average Sunday.

Keep the source in mind: these are company figures, not audited ones. But the pattern is consistent enough to plan around. Have your pre-landers tested and your budget loaded before the holidays end, expect traffic costs to rise as everyone else does the same, and use the quiet weeks of December to build creatives. Within the day, Hinge says activity peaks in the evening after work, which is worth using for day-parting once your data confirms it.

Dating ad angles that convert without lying

Dating creatives have a reputation, and it is earned. "She is 2 miles away and wants to meet tonight" is the classic. It is also the kind of claim the FTC has gone after. The agency's advertising guide¹⁰ says an ad is deceptive if it is likely to mislead a reasonable consumer about something that matters, that implied claims count as well as stated ones, and that fine print cannot fix a misleading headline. It also says ad agencies can be held liable alongside the advertiser when they knew or should have known a claim was false.

The dating cases are specific. In 2014 the FTC settled with JDI Dating, which ran 18 dating sites and used computer-generated "Virtual Cupid" profiles to message users who then had to pay to reply. The order required $616,165 in redress and clear disclosure of virtual profiles. In 2019 the FTC sued Match Group, alleging that Match sent "you caught his eye" notices triggered by accounts it had already flagged as likely fraudulent, and that by its own analysis 499,691 subscriptions were bought within 24 hours of such an ad. Those were allegations, not findings. The pattern for affiliates is still plain: inventing a person, or a contact that never happened, to sell a subscription is the line.

Ad platforms say the same thing in their own policies, and most go further and refuse casual dating ads outright. Our guide to CPA traffic sources for adult offers quotes the Google and Meta rules.

What works instead is boring and true. Speak to the user's situation ("new in town," "over 40 and back on the apps"), describe what the site actually does, and use real product features: verification, video chat, a free trial if one exists. Use stock photos that do not pose as a specific person messaging the viewer. Say what it costs if you know. This converts a bit worse on the first click and much better on the metrics that pay you: confirmed leads, first purchases and users who stay.

Testing, scrubbing and getting paid

Test one variable at a time, in this order: offer, then country and device split, then pre-lander, then creative. Launch with several creatives rather than one, because ads on push and pop traffic tire quickly, and start broad (run-of-network) before building a whitelist of the placements that convert. Write down your kill rule before you launch, for example "pause any placement that spends three times the payout without a conversion." It stops you from rescuing losers with hope.

Then expect scrubbing. Advertisers check leads, and the network passes the result on. CrakRevenue's dating guide notes that advertisers can set daily lead caps or lead-to-conversion ratios, so a pile of SOI leads that never buy can cost you the offer. Incentivized clicks, bots and email spam get leads rejected and accounts closed. This is why we compare models on EPC after rejections: a source that looks great on Monday and loses 30% of its leads by Friday was never great.

Track everything server-side with postbacks, so every conversion maps to a click ID, a placement and a creative. The same CPA traffic sources guide covers trackers, bot filtering and where to buy the clicks in the first place.

If you want a network to start with, CrakRevenue runs the dating offers and smartlinks described above. That is our referral link: CrakRevenue pays referrers 5% of referred affiliates' earnings for life, so if you join through it, we earn money from your work too. Better you hear it here, next to the link, than find out later.

What to do this week

1. Decide mainstream or casual. Write it down, and pick traffic sources that allow it.
2. Pick the payout model for your traffic: lead payouts for cold paid clicks, PPS or revshare for search, reviews and email.
3. Build one campaign per country and per device. Do not blend.
4. Write one quiz pre-lander and one short article pre-lander, both honest about the offer.
5. Set a kill rule and a test budget you can lose without flinching.
6. Judge results on EPC after rejections, not on raw sign-ups.
7. Put tested creatives and budget in place before the first Sunday of January.
8. Reread every ad and ask: does it invent a person? If yes, rewrite it.