Disclaimer: Please note that this case was provided by one of our clients, and the views and opinions expressed in this article are those of the authors and do not necessarily reflect the official policy or position of PropellerAds.

Thanks to the author for permission to publish his case study. The original text can be found here.

This case study definitely claims to be one of the best guides for launching Popunder campaigns!

Very detailed instructions on settings and optimization, not obvious recommendations and conclusions — we could not pass such a useful text! Although the campaigns were launched a long time ago, they still bring profit, and the approaches described by the author can be successfully applied right now.

Ad Format: Popunder
Affiliate Network: ClickDealer
Offer: Sweepstake
Payout: $20
GEO: Multiple
Spent: $6 430
Income: $13 323
Revenue: $6 893 
ROI: 107%


I will explain in great detail how I managed to make a good profit in a campaign that is still generating amazing ROI today! This case study may be used as a FREE guide describing the “ins and outs” of running sweepstakes vertical on POP/PPV traffic.

The volume of traffic available from the GEOs for this campaign will ensure that there is a way for everyone to get a slice of this sweet and delicious ROI pie. 

Without further ado, let’s start.


I’ve always looked at offers that have /INTERNATIONAL in their title with quite a bit of skepticism. It wasn’t until late last year that I’ve decided to give one of them the good old college try while being bored one day.

That offer wasn’t suggested by my affiliate network (ClickDealer), nor did it appear on any of the top performer’s newsletters. It was literally just me randomly browsing all available offers on the network to see if anything stands out. This further emphasizes the point of always testing everything! You never know what kind of golden nugget you may come across.

The Offer

Over 150 countries accepted this Sweepstake offer, most of them being Tier 2 and 3 GEOs.

It was available in multiple landing page variations (Samsung, iPhone, and their variations), but to be honest, I’ve only grabbed the very first tracking link that I later combined with an iPhone landing page, without even checking if that’s what the claimed prize was for that particular link variation.

This is how confident I was that most likely, this campaign will tank and make no money. Boy, was I wrong!


A very interesting part about this campaign is that zones that have a huge volume of traffic at PropellerAds, but very low CTR/Conversion Rate and would normally do poorly in most GEOs, have generated the most profits!

Yes, zones that I’ve written off and considered duds were now responsible for a huge chunk of my success with that campaign.

Landing Page

As PropellerAds becomes more and more strict with creatives they allow, it took a bit of headache and testing to get the landing page their compliance team would be happy to approve.

After a number of tries, playing a tug of war with the compliance team had a “Prize Box Style” landing approved.

There was a zero chance I was going to take my time to translate the landing page into the campaign’s native languages, such as Swahili. (Imagine the ROI boost if I did? Later on that towards the end of the case study).

So, I just ran the English version of the same LP. Now that I’m writing this, it would have been probably smart to localize the landing page even more, or at the very least, include a country dynamic token.


Tracker settings

Here are my exact tracker settings for this campaign. Binom automatically tracks PropellerAds costs, but make sure to set your cost to CPC and check off “Auto.”


Traffic Source Settings

The campaign setup is rather intuitive. After choosing a campaign name, select OnClick (Popunder) for Advertising Format and choose SmartCPM as your pricing model.


With CPM, you will pay a set bid price, regardless of what the other bids are. With SmartCPM, you will only pay slightly higher the amount of a second highest bidder if you win the auction, even if you set your SmartCPM to something astonishingly high.

Note: Last year, PropellerAds introduced the CPA Goal 2.0 bidding model that some affiliates claim to have good results with. This is something that might be worthwhile testing as well.

So you may be wondering, how did I choose the GEOs for this campaign? Well, I didn’t really; I just ran them all, separating them randomly into three different batches of GEOs.

I took my time to copy and paste each allowed GEO in the offer description in the PropellerAds Campaign Creation window. Once the first campaign is created, you can easily duplicate it, creating additional ones to test more zones and remove under-performing countries.

The frequency was set to 1 per 24 hours. Automatic bidding disabled.

Due to running RON and knowing that I will most likely get flooded with traffic since nobody else normally runs these GEOs (At least until now ), I chose a rather conservative CPM of $1.8.


Testing & Optimization

Now, for the fun part. The one that separates the men from the boys. I’ve split my testing process into three distinct stages.

Stage 1: Run a RON campaign for a few days

Stage 2: Copy it over to the whitelist version with all collected zones.

Stage 3: Cut non-performing GEOs


First, I’ve created a RON campaign, each for mobile and desktop devices. Due to the amount of GEO’s that were allowed in these campaigns, I separated it into two batches, with countries selected randomly.

Keeping an eye out on it like a hawk after the launch, I aggressively cut every zone that received over 1500 impressions and less than 2% CTR.

Even with a low CPM bid, the high amount of traffic flooded campaigns almost immediately. It will be fairly obvious which zones are complete duds; they would get the most traffic in the least amount of time with the least amount of clicks. So be careful, you don’t want to waste your budget on those. The duds will have around 3000-5000 impressions and a CTR of 0.50 or less. Cut those immediately!

Here’s another trick. Every day or so, I would up the bid by 25% or so to ensure that I’m getting traffic from the higher quality zones that the original bid wasn’t high enough to receive traffic from. This way, I get a reasonably legitimate whitelist of zones without missing any of the golden ones.

I did not follow the standard rule of cutting zones that exceeded 1.5x of the offer payout without a single conversion. I cut zones based on their CTR.

Yes, it’s possible that some zones would have a huge CTR but receive absolutely no conversions. I opted to wait for Stage 2 of the process when running my whitelist campaigns to cut those out.


After 7 days of running the RON campaign, it was time to create duplicate whitelist versions of it. Normally, I would stop the RON campaign and concentrate on the Whitelist ones, but instead, for whatever reason (greed and lack of patience for the most part), I’ve decided to turn RON campaigns into a whitelist campaign, cutting the number of GEOs to only those that converted.

Hence, if you’re looking at the campaign screenshots below, the data might not be 100% accurate in terms of ROI for each campaign variation.

I ran both the former RON campaigns (that now became whitelist campaigns with top GEOs) and new whitelist campaigns (all available GEOs) simultaneously for about another week.


Stage 3 was zeroing in on only top countries from the whitelist campaigns. I’ve removed the countries that did not show signs of life from currently running campaigns and created new separate campaigns with only top GEOs and top zones. The results of this are shown here:



Not being lazy and willing to go just one step further, almost always pays off in the world of performance advertising (and life in general).

After running this campaign for a while, I’ve noticed that there was a significant amount of traffic and conversions coming from the French-speaking GEOs.

So I created separate campaigns for those, translating the landing page to the French language.

To no surprise, after a split test, the landing page adapted to the native language of the traffic performed significantly better.

I’ve tried doing the same for the Spanish demographic, but did not have as good of results as I did with the French.


Hopefully, you found this case study helpful. Depending on when you read this, the offer might not be longer available on the network, but one thing for sure, there is going to be other, similar offers that you can apply exactly the same techniques to and make some money.

Tier-3 countries are perfect for those with a smaller budget, not willing to fight for traffic against the big dogs.

Until next time.


You reached the end. Bravo! Now you can join our Telegram chat not to miss other cases like this 🙂