Pop-Under Affiliate Marketing Case Study: Pop-up Traffic with 43% ROI

Our friends from ExpertMobi shared a hot adult case study. They started with push traffic and eventually they switched to pop traffic. Here are some numbers:

• Traffic sources: Clickadilla, ExoClick
• Affiliate network: ExpertMobi
• Offer: [WEB] Bongacams / US, SE, GB
• Flow: User signs up with their email > Conversion (SOI)
• GEO: US
• Campaign period: 26.07.19 — 08.08.19
• Cost: $1744,4
• Total revenue: $2497,4
• Profit: $753
• ROI: 43%

Today we'd like to share a case study for the offer [WEB] Bongacams / US, SE, GB and a few thoughts on our experience with adult offers/traffic in general.

At first, we tried to make push traffic work with adult campaigns, but its quality was poor. As a result, even when we managed to find good sources, advertisers didn't like it anyway in the end. So we decided to get back to pop traffic, that we've been working with for a pretty long time.

As for traffic sources, we decided to start with Clickadilla, 'cause we already had up-to-date blacklists there. We also wanted to try something new. As a result, we got:

• PropellerAds – they approved text banners for us, but we faced difficulties trying to get adult banners approved. We've been buying SE traffic from them for some time but didn't get any good numbers. 
• PopCash – didn't approve any adult creatives of ours, so we didn't have a chance to test their traffic at that moment. But this is a nice traffic source. At the beginning of September, they had nice sources with live traffic for the UK.
• ExoClick – it was a funny story: due to our mistake, we got traffic worth $600 instead of 3400 CPM sent in just half of an hour. They helped us solve this issue really quickly and we continued working with them as with our primary traffic source for this offer.
• EroAdvertising – we tested their traffic as well and it had potential.

At first, we were barely breaking even. We tested different geos and offers. Soon we realized that it was all about offers. Some of them were already outdated, others were not interesting enough for users. We needed to find something that would perform to be able at least to send traffic for a longer time period.

At first, we tried to use the same creatives adapted for all geos we tested. It is not hard to use AdPlexity to find them. Even though the information that you get from spy tools is often outdated, you can still use it to outline tendencies and plan your further work. You can also find creatives manually and make your own ones using the data you got.

Step-by-step we found creatives that showed the best results and started our attempts to scaling our traffic up. We did not pay enough attention to other traffic-optimization options but creatives and saw that performance fell even with the top-performing ones.

That is why we began our work on targeting by age and by time. 

We used the following logic:

• Adult campaign will not convert at 1 pm, because most part of the audience was at work at that time.
• We needed to target older people that were actually financially reliable.

Our first attempt at time targeting failed, as we started to send traffic from 8 am to 8 pm (UTC -4). Here is the screenshot:

We checked the numbers, then we decided to set the time targeting to 7 PM till 8 AM. Traffic volume slightly decreased, because the time period became shorter and the bid went up. The number of leads increased, and so did ROI, you can see it at this screenshot: 

As for the audience, we targeted male users at the age of 18-23 at first, but we didn't see good results. Then we decided to target 35+ male users and it worked just fine.

We also chose suitable creatives for such auditory. Young models didn't perform that well, so we chose realistically-looking milf models for our creatives. Here are some examples of the models we chose:

We used google captcha to stop unwanted traffic, just as we usually do. We started to remove sources on the following conditions:

• СTR of the lander < 2% = ban
• less than 30% of traffic passed through captcha = ban

Then we optimized the performance of LPs and sources that were of fine quality according to the conditions I mentioned above.

We checked the conversion flow manually several times. We followed the registration form of the offer till the end three times and got three conversions.

Below you can see the stats of two weeks:

We realized that 20-30% ROI is totally fine for adult traffic. Larger volumes allow getting larger profit.

The results were similar for top geos. It took us a couple of weeks to find the best creatives, landers and offers. Offers start performing with pretty low ROI. But if we tried to scale up, advertisers asked us to stop campaigns. 

We sent traffic from one of the most famous traffic sources and used standard landers.

So it turns out that adult traffic + adult landers + adult dating campaigns ≠ stable profits.

We've been thinking about this problem and we can't say for sure but still, we believe that this is not the problem of traffic sources. Maybe it's all about the message. Users need to understand what they subscribe for and what will follow. Some of the operations are not free and they need to be specified. There is no need to show everything down to users, but they at least need to be prepared for the following actions.

As a result, we figured out that it is more profitable to use less aggressive creatives with time, so that offers could live longer. It is also important to find the offer that suits your traffic as nice as possible. Now we understand that even with different advertisers the same traffic will work in different ways. One advertiser will like your traffic and ask for more, the other one will ask to pause it.

There are also no reasons to panic is advertisers ask you to pause traffic. They can also ask you to return traffic in a month. Why? It's simple. Advertisers pay for SOI sign-ups, but they actually check traffic quality according to how many sales/deposits/etc. they get from users. For example, there can only be one sale from 50-100 SOI conversions. And advertisers will identify profitable sources taking the number of sales into consideration. Well, there is always a chance that the advertiser will see no sales and will decline all traffic you sent without even checking users' behavior. And when they analyze it later and see that traffic was actually live, they'll ask you to resume it.

In the end, I can say that we got a lot of new ideas to test, so we'll continue working on it. 

Wish you all good luck and huge profits!

Share the article