Conquering Mobile Pops With Automation (9)
There have been a lot of rumors lately saying that Pop/Redirect traffic is dying out, it’s not worth loosing time and money on it unless you cloak etc etc etc.
If you are using that as an excuse, sorry to burst your bubble, this is ABSOLUTELY NOT TRUE and here I’ll show you the PROOF.
Pop/Redirect traffic is still a very solid traffic source that can generate high $x,xxx a day profit if done correctly. Obviously, that requires intensive work, accurate data analysis and a sound optimization process.
To better show you what I mean, let’s get into a real-world scenario where every new and experienced affiliate/media buyer here on STM is familiar with.
Disclaimer: The following campaign has been running on auto-pilot using TheOptimizer Mobile while conducting internal platform tests where the main focus has been on the platform’s performance itself.
To better test the stability of our platform we decided to run a real-world test in a high-volume geo using one of our supported traffic sources.
How did we find the offer?
This was a 15 minutes job! Went on Adplexity, checked the results of all Pop/Redirect campaigns in UK that was running for at least 5 days (date range last 2 weeks).
After picking one of the most promising campaigns from the search results, we “borrowed” and cleaned one lander and got access to the offer on the affiliate network detected in Adplexity.
Disclaimer: This was just to test TheOptimizer. I would never suggest starting with one lander and offer only. As you may have heard a million times already, split-test, split-test and then split-test some more.
Funnel Setup
The laziest funnel setup I’ve ever created! Just 1 Lander and 1 Offer, no split-test at all.
Auto-Optimization Rules
Once we created our campaign in
Here we go:
Rule 1 – Block Bot Traffic 1 (0% LP CTR)
Condition: Block all targets with more than 100 impressions on the traffic source and 0 Clicks on the Landing Page.
Action: Alert & Execute
Rule Level: On Publishers
Rotation: Every 10mins
Stats Interval: Last 7 Days
Rule 2 – Block Bot Traffic 2 (Unrealistically High LP CTR)
Condition: Block all targets with more than 20 Impressions on the traffic source and LP CTR More than 80% with 0 Revenue.
Action: Alert & Execute
Rule Level: On Publishers
Rotation: Every 10mins
Stats Interval: Last 7 Days
Rule 3 – Block Low CTR Targets
Condition: Block all targets with more than 70 Impressions on the traffic source and Less than 1% LP CTR on the Landing Page with 0 Revenue.
Action: Alert & Execute
Rule Level: On Publishers
Rotation: Every 10mins
Stats Interval: Last 7 Days
Rule 4 – Stop Loss (Avoid losing more than $100)
Condition: Pause Campaign if Net Profit is less than -$100. This was to make sure that we were not going to waste more than $100 in this test.
Action: Alert & Execute
Rule Level: On Campaigns
Rotation: Every 1 hour
Stats Interval: Last 30 Days
With everything in place, all we had to do was to start the campaign on ZeroPark and monitor our rule’s execution performance.
During the first week this campaign was bouncing from -70% to -85% ROI without showing anything promising at all. But, this was an expected result since we were bidding extremely low in order to block publishers progressively as we raised the bid.
Even though this campaign was in loss, we were extremely satisfied with the results. Auto-optimization rules managed to block approximately 150+ bad publishers per day, until there was no more traffic available on that specific bid.
And this is where this test became more interesting…
On week two after rising the bid by 60%, we started receiving more and more traffic which was a typical mix of good a bad publishers, complex enough to manually optimize it.
On day one of week two we logged around 200+ blocked publishers within the first 6 hours by Rule 1. Furthermore, during this week, the auto-optimization rules continued to block unwanted publishers resulting in higher traffic quality and ROI.
Considering our conservative bidding strategy that intentionally throttled the traffic flow, here are the results of the following weeks in numbers. 
Week 3:
Week 4:
Week 5:
Week 6:
Week 7:
Week 8:
Week 9:
Week 10:
Week 11:
Week 12:
As you may notice there is a fluctuating ROI trend that changes with bid and optimization performance.
Once reached a bid of $7.5 CPM with an overall spent of $1,366.07 and convinced of the rule’s performance, we decided to pause this test campaign since keeping it profitable was not our goal.
Along with the official beta opening of TheOptimizer Mobile plaftorm for STM Members we will continue with another Man vs. Machine case study focused on mobile Pop/Redirect traffic.
Are you ready to challenge yourself against Automation? Reserve Your Seat for a Free Account during Beta and start your own follow-along.
nice demo. very curious what happened after week 7 though since it seems it's going downhill from there.
Originally Posted by danielt
But, because this camp's objective was just to test auto-optimization rules performance our conservative bidding strategy wasn't going to keep it profitable any longer until we paused it.
got it. it does make sense. it would be nice to see some bidding war strategies on this front 
Originally Posted by danielt
I think it's pretty clear that these days, with the amount of competition in pops, automation is a must.
3-4 years ago the bids were much lower and this covered up for the bad widgets and bots losing money.
Nowadays with prices shooting up, unless you have a good offer and be aggressive with the blocking it will be really hard.
Let's assume the average campaign gets traffic from 10,000 publishers and you have 20 running campaigns.
It's pretty much impossible to catch all bots and bad performers on time before they burn your budget even if you work 16 hours a day.
Originally Posted by bbrock32
I think that by aggressively cutting poor performing placements based on their LP CTR or Spent without Revenue in the beginning of a new campaign will help quite a lot reaching out good ones. This way one can make room for additional testing budget for those placements once the camp is in profit without having to spend a fortune since the beginning.
Really great tool, Platinum! I'm giving it a try and excited about the ability to write rules and automate some of the decision making.
i want to try it later
Home > Mobile >