Tired of Lead Quality Problems? Let me do a test with PPS dating ... (10)
Tired of Lead Quality Problems? Let me do a test with PPS dating ...
THE PROBLEM:
IMPORTANT NOTICE : This is a LIVE campaign that I'm still running, so if my affiliate manager finds this thread, I would appreciate to keep it PRIVATE, thanks 
OK, so we all know what's the biggest problem with SOI or DOI adult dating offers .... LEAD QUALITY ... It's quite easy to get people to convert with a properly designed funnel, but in a competitive world such as adult dating, you won't get far without a TOP payout per lead. And that won't happen if your lead quality is low ... in other words, if the leads aren't becoming paying customers, you won't get a payout bump and you won't become profitable.
There is one problem here ... funnels that convert well into leads, are often a bit misleading, which results in poor lead-to-sale ratio... and that equals to low quality. We, the affiliates, have to find some kind of a balance here ... we need to keep high CVR to leads, but it still has to convert to sales too. On top of that, advertisers protect themselves against poor leads, by using daily conversion caps and it takes quite some time to get it lifted as the leads need time to mature.
POSSIBLE SOLUTION?
Both of the above mentioned problems can be solved by using the PPS (Pay Per Sale) model instead ... It's a way less risky model for the advertisers, so they are not using caps and they don't care that much about the quality anymore ... cause a sale is a sale, right? And we don't have to care all that much about how good the clicks convert to leads, because we will be focusing on the final sales more.
There are also some negatives of course. First of all, this is not for small budgets, we are talking about payouts in the range of $50-$150 +/-, so a rather large spend is required in order to make any decision. Another big negative is that the sales can happen even weeks after the initial click was sent … this is the “maturing” of the leads.
Some members decide to start paying later on, every dating site has automated processes in place, that try to convert the leads into paying customers, over and over … The “oldest” conversion I got this way was about 6 months after the initial click So in reality, you can run PPS campaigns even with a slight loss initially and they will become profitable within a few weeks or months. This is why you need to pick a reliable partner to work with in case of PPS or Revshare.
LET'S TRY IT
Enough of the theory, let me show you what I did.
NOTE: I run almost all of my campaigns on the PPL (leads) model, so this was more or less a test to see what would happen.
I picked a PPS dating offer in a large EU country, and due to my good relationship with the advertiser, I was able to start with a quite large payout of about 90 EUR per conversion and future bumps were possible based on how long the members will be renewing their memberships.
I focused on the 300x250 banner format, so mostly NTV spots. The traffic comes from the known large networks such as Trafficstars, Trafficjunky, Trafficfactory... all the standard ones. All banners were done by my designer, same as the LPs … inspired by what I found on Adplexity, but ALL modified. Due to the new regulations, I had to get rid of pretty much all misleading elements and tone down the angles a bit too. This turned out to be quite a problem, networks reject a lot of stuff they would let fly just a week or two ago.
As for the settings … it's all CPM bidding (smart wherever possible), low caps 1-3 per day, bids set close to the top (2-3 spot) since I need quality traffic for this. I didn't set a daily budget but picked just certain sites/spots in order to limit the initial volume.
----------------------------------------------------------------------------------------------------------------------------------------------------------------
NOTE: I made a mistake in the payout settings during the first 2-3 days, that's why the numbers on
One more NOTE: I ALWAYS set a bit higher rates per click in Voluum, because of sudden price increase for example, and I also like it this way … makes me feel better when I know the actual numbers are a bit higher than what I see in Voluum… might seem weird to you, but that's how I like it The real traffic costs are always at least 5% lower than you see in the screenshots.
----------------------------------------------------------------------------------------------------------------------------------------------------------------
Here we go with the first screenshot ... I started with one campaign, to see what happens :
OFFERS:
LANDERS:
As you can see I was pretty lucky with this 1st campaign. One LP and one Offer started to perform better straight off the bat. One thing to note here, one of the sales came in few days later, that's why you see higher payout on it – we made a mistake in the settings, so the payouts were not reported correctly during the first 2 days.
When working with SOI or DOI offers, I wouldn't consider this enough data at all, but since we are on PPS, each conversion has a higher statistical value so to speak, so I decided to cut the LPs and Offers early. I didn't touch banners yet.
I left all Offers run for one more day initially, but since I got another fast conversion on the already winning Offer, I decided to cut the other 2 and keep on testing with just 1 LP and 1 Offer. Now, this is not my favorite situation at all, I always want some variety.
But in this case, the numbers have spoken clearly and I wanted to quickly see whether this offer has potential at all or not.
The following screen shows the performance after 4 more days. As you can see, the campaign got another late lead, which changed the performance of one LP that I have already paused (This is the tough part with PPS or Revshare … when I paused the LP, it had 0 conversions, and I only found out about this now, when doing the screenshots).
SCREENSHOT after 6 days:
Looking at the overall numbers, I was still negative, but only by a little bit, so I knew this offers had some potential. Considering the couple % I always raise the CPC values with, it was VERY close to profit already.
The next thing to do, was to start playing with banners … I paused some, added new ones. When working with PPS, testing banners takes way longer, that's why I didn't test all that many yet. The following screenshot shows the overall performance of banners for this campaign from the start until 23rd Feb. The campaign is still running and as you can see, it is already profitable a bit.
1 banner totally kills the others for now, so I'm pushing that one the most … again, this is something I don't like, so there are more banners running as tests at the moment. Hopefully some of them show good performance too.
BANNER PERFORMANCE from start until the 23rd feb:
Of course I didn't just stop here, it was time to scale … always try to scale when you find a bunch of offers that show some performance. The good thing was that the same offer was available in multiple GEOs, so I started to launch them one by one.
Some worked well, some bombed… you can see the overall performance on the following screen. Bigger part of the campaigns are still running, it's still quite early in the optimization process (some are only running for a few days) and I'm still collecting data on most of them. The good thing is, I'm already in nice profit with some and I'm sure it will get better with time.
Please note that since it's several GEOs, the payouts are also different per GEO. Same goes for the prices of traffic etc …
OVERALL CAMPAIGN peformance ... the list is long, the following two screens show about 80% of the campaigns :
And more Campaigns ….
There are some more but really small ones As you can see, I'm spending quite a big part of the profits on testing more, this lowers the initial returns, but should pay out in the long term.
Couple things to note here :
1. PPS dating can work, I was able to make it profitable within about two weeks. The profits are still rather low, but that's mostly because I'm testing many campaigns at the same time. To be honest, I was surprised it went that easy ... I think it's because I used LPs and banners that I knew were good, based on my previous campaigns.
2. It takes way longer to test on PPS compared to PPL. It also holds me back from buying way more traffic, as I need to wait longer for the initial results. I don't feel like throwing a couple $1000s at right of the bat, before seeing some actual data, because I do not usually work with PPS in the first place.
3. If it wasn't for the high payouts I got straight from the beginning, I wouldn't be able to make it work without heavy loses at all. This is where the relationships come to play… BUILD them. This particular advertiser I know from Affiliate Conferences – AWE 2016.
BTW: Sorry for hiding the GEOs and Offers names … I'm still running these campaigns
Questions? Fire them at me
Great post Matuloo
I guess I have a theoretical question.
When testing PPS offers, obviously we have to run a banner for much longer than normal.
So does banner burnout theoretically not matter as much when it comes to PPS?
Theoretically instead of running a single banner for 200 clicks, we may have to run it for 2000 clicks (and beyond if it's a winner)
So do you think in theory the higher EPCs from a PPS offer will offset banner burnout?
Originally Posted by jabong82
I guess I have a theoretical question.
When testing PPS offers, obviously we have to run a banner for much longer than normal.
So does banner burnout theoretically not matter as much when it comes to PPS?
Theoretically instead of running a single banner for 200 clicks, we may have to run it for 2000 clicks (and beyond if it's a winner)
So do you think in theory the higher EPCs from a PPS offer will offset banner burnout?
I'm also looking at the conversions from clicks to SOI and then to DOI ... if the ratios are bad, I replace the banner just as in case of a PPL campaign. I still need to convert well to the DOI leads, otherwise the chance of actually making a sale goes down.There is one thing to keep in mind here ... just because a banner doesn't convert all that special to DOI, it doesn't matter it can't actually sell well.
I will do way more banner test with these offers soon, once I finally learn to speak the same language with traffic networks and their new regulations
Thumbs up for this great case study Matej!
Originally Posted by platinum
Originally Posted by jabong82
I guess I have a theoretical question.
When testing PPS offers, obviously we have to run a banner for much longer than normal.
So does banner burnout theoretically not matter as much when it comes to PPS?
Theoretically instead of running a single banner for 200 clicks, we may have to run it for 2000 clicks (and beyond if it's a winner)
So do you think in theory the higher EPCs from a PPS offer will offset banner burnout?
Mm you used actual conversion event from sponsor as a conversion metric during optimization ?
Thats a bit suicidal with adult and 90 eur payout.
Why didnt you use pre-conversion events such as free joins etc.. and optimize around those events ?
There shouldn't be much more than 2x discrepance in free join / paid for different landers / banners,
so its safe to assume it s a metric. Especially since you need to be compliant with both sponsor and broker.
Originally Posted by johner911
I hear ya with the free join/paid analogy, however the variation in free to paid joins is rather high for this campaign per banner, LP and source ... really hard to set any average to be honest, so I'm giving more weight to the actual paid joins. The traffic is quite a mix from several sources/placements using multiple angles ... I'm afraid that trying to set an average to go by, would be a mistake in this case.
Great post Matuloo!
It shows that adult dating can still be profitable if you do it right.
I think in some cases it might be easier compared to pops since competition is lower these days.
PPL might be good to start and learn the basics but PPS is where things get stable and won't have to worry about advertisers complaining about quality.
Originally Posted by bbrock32
Right now, with the new banner rules, it's a bit harder to get good ones approved, but that should settle within a few weeks I think.
It's certainly a good vertical to start with, not for everyone, but a list one
Home > Paid Traffic Sources > Adult Traffic (NSFW)