Make POP Traffic Great Again - The easy way! (21)

While running eCommerce on Facebook is becoming more and more attractive for affiliates, that is still hard to crack down due to high budget requirements. On the other side, even though Pop/Redirect traffic is known to be one of the traffic types with less or even no user intent at all it is still converting pretty well for affiliates.

With the rise of automation, campaigns optimization is becoming easier than ever, so why not dare to Make Pop Traffic Great Again? Let’s gradually see the progress and results in this follow along.

Here’s a brief overview of this campaign’s details:

GEO: United States
Vertical: Sweepstakes
Traffic Source: PopAds
Spy Tools: Adplexity Mobile
Tracking: Voluum
Hosting: VPS Server

Finding the offer: This time again my source of choice to find winning offers remains Adplexity. From the search results, I picked up 3 landers and 3 Win iPhone offers where I was able to quickly gain access to and continue with landers cleanup to start the campaign.

Campaign Setup: Simple Voluum path. Starting with two tested landers (google & facebook themed) and three offers.

Traffic targeting: The one thing I like PopAds the most is it’s granular targeting (and later on you’ll see why), but because I still have no data and the campaign will start with raw traffic the setup is pretty simple.
Freq. Cap: 1 view / 24 h
Quality: All websites
Country: United Stats
Devices: All smart phones and tablets (Android and IOS)

Because lately I’ve been running in an extremely tight schedule it was impossible for me to start this follow along earlier and post campaign’s results progressively since after day 1, but we’ll see first days progress when scaling to the next traffic source

Automation rules:
1. Pause all publishers that have more than 120 visits and 0 LP clicks during the Last 14 Days - Run every 20 min
2. Pause all publishers that have more than 20 visits and LP CTR is greater than 75% during the Last 14 Days - Run every 20 min
3. Pause all publishers that have more than 400 visits and LP clicks is less than 2 and Revenue is 0 Last 14 Days - Run every 20 min
4. Pause all publishers that have Spent more than $ 0.7 and have 0 Conversions during the Last 14 Days - Run every 20 min
5. Pause all publishers that have Spent more than $ 1 and LP CTR is less than 0.3% and Revenue is 0 during the Last 14 Days - Run every 20 min

Campaigns split test: Because I wanted to make sure that the auto-optimized campaign was not going to cut out any converting placement, I first created the manual one where no auto-optimization rules are assigned and monkey work is required

Day 1: Auto | Manual
Spent: $ 18.25 | $ 18.34
CTR : 0.68 % | 0.40 %
CR: 4.94 % | 2.08 %
ROI: -33 % | -83 %




Day 2: Auto | Manual
Spent: $ 9.99 | $ 9.99
CTR : 0.59 % | 0.25 %
CR: 0 % | 20.00 %
ROI: -100 % | -8 %




Day 3: Auto | Manual
Spent: $ 14.96 | $ 15.04
CTR : 1.19 % | 0.41 %
CR: 4.20 % | 7.32 %
ROI: -20 % | -53 %




Day 4: Auto | Manual
Spent: $ 40.04 | $ 19.81
CTR : 1.16 % | 0.39 %
CR: 3.97 % | 6.98 %
ROI: -38 % | -65 %




Even though the CR rate of the manual campaign was higher than the auto-optimized from day 2-4, the CTR was horribly low to pretend to break even so I decided to stop sending traffic to the manual campaign and continue with the auto-optimized one. Also, because the camps were running in US and I’m located in Europe, to keep a close eye on the manual camp (not to get out of the control) required me to stay up all night which was impossible considering my daily schedule.

Getting back to the split-test since the 3 offers were converting almost equally, the Facebook themed lander had 0 conversions in the first 3 days. Based on this I cut off this lander and moved to a new one. After applying a few small changes to it in order to avoid "eye-blindness" and be more visible to the visitors (at least take the effort to check out what it has to say) I added this one to the funnel.



The results were not what I expected to be and here is why...

CTA Button Setup might affect our campaigns performance and the way we optimize it!
If we get back to caurmen's thread on how to run a Bot Test, this landers CTA button had a similar setup like the java script bot trap which was different from the one I was using in my successful lander.

Here's a snippet of the button's setup:

<a href="javascript:void(0)" rel="noreferrer" onclick="exit_a1();PreventExitPop = false" class="claim">

 
<script type="text/javascript">
     function exit_a1() {
            window.onbeforeunload=null;
            window.location = 'http://myclickurl.here';
        }
  </script>

Day 5: campaign's performance dropped drastically as I was keeping almost the same bid and for that bid range the quality of traffic coming in was not so good. Also the number of poor performing pubs was still rising.

Spent: $ 37.70
CTR : 2.95 %
CR: 0.38 %
ROI: -92 %

Day 6: Bid update and better ROI

Spent: $ 51.39
Revenue: $ 39
CTR : 2.99 %
CR: 2.69 %
ROI: -24 %

Because I hate to waste money and that is one of my main concerns when running a new POP campaign I started a parallel white list campaign who's main propose is to cover the costs of finding new profitable publishers from the original campaign.

Day 7: Auto | White list

Spent: $ 89.29 | $ 44.60
Revenue: $ 63.20 | $ 46.40
CTR : 2.16 % | 1.18 %
CR: 4.07 % | 16.07 %
ROI: -29 % | 4 %

One important thing I would like to emphasize here is that by getting use of auto-optimization rules I managed to exclude poor performing publishers that might have cost me even more than I paid purchasing traffic to identify them which by the way

Total publishers blocked: 3214
Identification cost: $ 100.32


Currently the targeted placements are at an average of $11 CPM while the whitelist campaign is running on $6 CPM and with no auto-optimization rules.

In the next days I will post performance results of this campaign on PopAds along with scaling it on other traffic sources as well by split testing Auto vs. Manual.

Cheers!

This looks interesting indeed ... Looking forward to the next part!

Awesome, great to see another automation follow-along! Also looking forward to seeing progress.

Update: Like every other Monday and Tuesday the total number of conversions has fallen



September 18/19: Auto | White list

Spent: $ 41.67 | $ 139.01
Revenue: $ 18.00 | $ 143.40
CTR : 0.81 % | 1.34 %
CR: 5.13 % | 12.77 %
ROI: -57 % | 3 %

Because the white listed campaign was getting most of the traffic from 2-3 publishers, I divided them into two campaigns by allocating separate budget to the other low volume ones, pointing to the same white list camp link in Voluum.
While it looks like there are not many publishers left to discover for this campaign and already having a few ones identified, today I'll open this same campaign in PopCash. Same landers, same offers and no initial black lists.

Really interesting case study so far!

Have you tried split-testing a few landers to improve the EPC?

Originally Posted by bbrock32

Really interesting case study so far!

Have you tried split-testing a few landers to improve the EPC?

Hi, I am a newbie here, could you please tell me where did you setup those automation rules. I didn't see those roles on Voluum.

Originally Posted by boycemrwx

Hi, I am a newbie here, could you please tell me where did you setup those automation rules. I didn't see those roles on Voluum.

Originally Posted by platinum;325307

[B

Finding the offer:[/B] This time again my source of choice to find winning offers remains Adplexity. From the search results, I picked up 3 landers and 3 Win iPhone offers where I was able to quickly gain access to and continue with landers cleanup to start the campaign.

Very cool follow-along. Looking forward to your updates!

Originally Posted by 6packplan

Could you please give more details in how you pick the offer from Adplexity? I have tried few times but never knew how it is done .
Thanks

September 20 results:

Man vs machine! I really like this and am interested to see how it turns out.

Do you always find such a big drop in ROI when you take a winning funnel to a new TS and how long do you give it until you decide if the new TS will be able to turn a profit?

So how did the weekend go? Do we have any progress?

Hey guys, sorry for the delay in posting the results but I took the time on formatting the stats in tabular view for easier readability.

Here are the updates of this weekend.

Amazing, so you reached the land of profits Thanks for this threads, I'm sure it serves as great source of motivation for the new guys!

BTW: don't waste time with the tables, just post screens as before, it's easy to understand anyways and it also gives the whole case study way more credibility.

Looking forward to future updates.

Originally Posted by matuloo

BTW: don't waste time with the tables, just post screens as before, it's easy to understand anyways and it also gives the whole case study way more credibility.

Excellent, keep it coming

A VERY nice FA! Or should I say case study? Thanks for this and other contributions you've made to the community! You've been on fire!

Originally Posted by vortex

A VERY nice FA! Or should I say case study? Thanks for this and other contributions you've made to the community! You've been on fire!