Pop-unders vs Banners: What Adult Ad Formats Really Pay (Our Numbers)

Ad network dashboards make every format look promising. We pulled a full week of per-zone statistics across our tubes to see what each format actually earns on a small to mid-sized adult site. The gap between formats is much bigger than most guides suggest.

The numbers

One week, same ad network, same sites, revenue per thousand impressions:

• Pop-unders: about $1.00 to $1.30
• Banners, native widgets and sticky bars: $0.01 to $0.12
• In-video banners: about $0.01
• Video pre-rolls: under a dollar per week in total, because only a minority of visitors ever watched one to the end.

On our busiest tube, the pop-under brought in about two thirds of the site’s entire ad revenue. Every banner zone on that site together earned less than the single pop zone.

Why pops dominate on small sites

Advertisers pay for attention. A pop-under opens a full window, so it is priced like a visit. A banner on an adult tube competes with a playing video and is mostly ignored, so it is priced accordingly. On top of that, banner fill on smaller sites is often low, and a lot of what does fill is house ads or remnant campaigns.

That doesn’t mean you should remove banners. They cost little and add up with traffic. But if you’re deciding where to spend effort, pops are where the money is.

Frequency matters more than the network

The biggest single change we made was not switching networks but fixing frequency. One of our sites was showing at most one pop every 12 hours per visitor, a default we had never changed. That gave about 0.26 pops per session. Our best site showed up to two per 10 minutes and got about 1.1 per session.

After moving the first site to two per 10 minutes, pops per session roughly doubled within days. Before you test a new network, check what frequency cap you are actually running.

The other side: too many pops drive visitors away for good. We keep a cap and a minimum gap between pops, and we never fire them on the first click on the play button. A visitor who leaves after one video earns you one pop. A visitor who comes back every week earns you hundreds.

Comparing pop networks fairly

When we compared two pop networks on the same sites, the results depended on the site. One network paid more per thousand on our straight tubes with mostly Turkish traffic, the other did better on a hentai site. Rules for a fair test:

• Same site, same frequency, at least a week each.
• Compare revenue per thousand real visitors, not per thousand pops, because networks count impressions differently.
• Look at the country mix. Rates vary many times over between countries, so a traffic shift can look like a network change.

Bots make every number look better

One of our sites showed about 18,000 users in one week in analytics. Almost 17,900 of them were one-second desktop visits from a single data-centre region, a headless browser checking pages. The real audience was around 300 people. Ad networks filter some of this, analytics doesn’t. Before you calculate earnings per visitor, take out obvious bot traffic or the number will be off by a factor of fifty.

A simple sanity check: count how often your ad script is actually loaded (from your server logs or the network’s impression count) and compare that with analytics users. Real visitors load the ad script. Many bots don’t.

What we’d do on a new site

• One pop-under zone with a sensible cap (we use a couple per 10 minutes with a gap), from day one.
• One banner above the fold and one native widget below the video. No more until traffic grows.
• No pre-roll until you have enough video views to measure completion.
• Check revenue per zone weekly, not the dashboard total.
• Remember the real fix for low revenue is traffic. No format turns 300 real visitors a week into an income.