Home > POP / PPV / Redirect > Follow-along Campaigns

---

1st Ever Follow Along - 1st Ever Mobile Pop Under Campaign On Propeller Ads (5)

---

Hi Everyone,

After spending few thousand on native and adult direct managed media buys with $20-60 payouts in USA (I know huge beginner mistake for a beginner!), decided to tackle mobile pop unders since they are great for newbies to learn the ropes.

Here is my first ever follow along, so definitely open to advice and its all learning for me.

Offer: AVG App install - spied from Adplexity

Geo - Ghana

Payout $0.30

Budget $30

Landers = 2
- 1 Direct Link
- 1 Created - See below - it makes sound when you get pop under and its targeted to Android only





Campaign started on 24th of Nov. for $10 / day - bid was set to $1.15 (Middle of range)

Here are my stats for Day 1 (Nov. 24th)
Impressions = 9,561
- Voluum shows only 4,212 visits = is there an issue here?
Clicks = 187
Conversions = 1 (Direct link)
ROI = -99%

Day 2 (Nov. 25th)
Impressions = 9,051
- Voluum shows only 6,051 visits = is there an issue here?
Clicks = 372
Conversions = 1 (Direct link)
ROI = -99%

For both days, here are zone reports from first 2 pages from voluum report.









Campaign is stopped for now after $20.

Questions I have going forward:

- Continue running campign but start cutting zone id's like crazy?
- If yes, is there a standard rule of thumb? like cut placements that spend 1-2x payout without 1 conversion?

- Have not tried ripping a lander or 2 off adplexity to use as benchmark - would this be a wise strategy?

- If there are any major mistakes I'm making - would love any advice here.

Thank you

---

Hey there,

Yes defintely test more landers and angles. Split test them and pick a winner based on statistical significance.

---

Congrats on taking action!

I think you need to reconsider your whole approach here. Firstly, unless you wanna restart the whole thing fresh, I would not consider continuing on with this campaign. You've spent over 60x the payout of this offer for the campaign, with two conversions: -99% ROI as you've mentioned. This is not good.

A commonly used 'rule of thumb' budget for this is: #-of-offers * #-of-landers * payout * 3 to 5 . So with this campaign, for example, you could do this formula as: 1 * 2 * $0.3 * 3 -> 1 * 2 * $0.3 * 5 = $1.8 -> $3 . In other words, you could have spent about $3 total to test this campaign before continuing based on good performance, or pausing and moving on based on poor performance.

So very important to take away here, yes you overspent here, but it's not the end of the world. In the future, don't just launch and guess your way around, but be thoughtful and precise, especially with your budget, and especially with the basic math (as much as you CAN -- I know there is a lot to learn).

A $10/day budget for this size funnel & such low offer payout is overkill. As PropellerAds will not let you set a budget LESS than that, you should flesh out your campaign with more offers and landers such that the funnel size is appropriate for this minimum budget, at the least.

You could either choose to flesh out this campaign a little more and basically "start over" with it, with more offers rotating and more landers as well, or you could choose to test a new vertical/GEO and do the same thing. In general, I suggest testing 3-5 offers and 3-5 landers at a time for an initial test, minimum. When the payouts are so low < $0.5/$1, I personally have a tendency to test even more offers at a time (like 10+), though you need to spend more to do this. I did the math for just one vertical I was running on pops/display, and it was on average 1 out of every 25-30 offers I tested that I managed to make green in some capacity -- and that was using landers that I knew pretty well to be decent! You need to test a LOT of offers to find a good one. This is why you should start with offers that you've spied doing high volume, or that you KNOW someone is making work because of what your affiliate manager tells you. Some networks, like Mobidea or BillyMob will even show you the data very transparently, so start with decent offers.

I may even split test bids at the same time, briefly even at first to see what gets traction, though this could be saved till after you make some cuts, if you don't wanna redundantly waste money on testing your funnel. If your bid isn't right for the TS, you might never get any decent data to work with, so you should do this early on.

In regards to this campaign specifically, AV offers very rarely work well without landers so direct linking is usually waste of time (though it can be part of a split test, don't expect much there, rather focus on your landers). You should spy on this GEO on Adplexity and rip the top few landers and run those; don't make your own if you don't know what works, or at least mix it in with PROVEN landers. If Ghana is not on adplex ( I don't recall, then rip something from nearby country and make sure the language is appropriate). Generally, the "Google" logo style ones (the most aggressive as you can imagine) will work the best, but this will not be accepted by PropellerAds compliance. You'll either have to try to work that one to the most-aggressive, but still approved by them, or cloak (cloaking not realllly being the best option for you at this point). So all that being said, perhaps sweeps or mobile content would be a little easier to get running with.

The difference between PropellerAds & Voluum is normal, it's called "click-loss" and is inevitable with pop traffic.

Also, put your costs into Voluum (you can use manual cost update) and screenshot your stats from the web-app, so people can help you better here. We don't need pages of zone IDs for two conversions, but better yet, screenshots of your lander split test data with costs, or offer split tests, etc. Zone IDs if there's some real data to work with there, sure. If you upload your images to imgur.com, then use the BBcode tags to add them to your post, they will show up as full size rather than little thumbnails too -- this helps as well .

If you want inspiration, there are some very useful guides in the "newbies first steps" and the "newbie-friendly guides and tutorials section" of the forum. You can also learn a lot by from case-studies and other's follow alongs if you want a little higher-level view of things. So back to the drawing board -- good luck!

---

Hi Chris,

Thank you for the brutal, but honest feedback on my first campaign.

Figured there would be lots of red flags with my first campaign.

As to action next steps - I'll do the following:

- Sign up for Mobidea
- Read the newbie 40 day pop course from front to back
- For next campaign, do following
- test 3 offers with 3 landing pages each offer.
- Go with sweeps and mobile content offers since propeller is more strict.

---

Thanks for starting a follow-along!

First of all, I would definitely agree with wes888 - you need to test a LOT more than just 1 lander (direct-linking doesn't count as a lander ).

And Chris has given many valuable tips - he has launched a ton of campaigns so knows what he's talking about.

Clickloss of 20-30% is normal. Beyond that, there may be an issue.

Are you using a custom voluum domain? If not you should. Sometimes domain will get banned by google, which can potentially cause clickloss. I have a step-by-step on how to do that here:

https://stmforum.com/forum/showthrea...eparation-Work

Are you using premium DNS? If not, that's something you can test. Faster DNS redirection speed can also potentially limit clickloss. I talked about how to set up NS1 here:

https://stmforum.com/forum/showthrea...osting-and-CDN

Or if you want a cheaper service (which is slower though), check out Route 53:

https://stmforum.com/forum/showthrea...iliates-bother



Definitely do NOT try to get green all the way from a very negative ROI by cutting placements!

That will be really expensive, plus you won't have traffic left over to profit from when you're done!

ALWAYS do what you can to find the best offer and lander first, so you would make as much of the traffic profitable as you could, BEFORE resorting to cutting the unprofitable parts.



Most definitely! And please don't just rip 1-2. Try 5-10. Using unproven landers in the beginning is asking for failure. Use tried and tested landers first, test and find a good offer, THEN you can test lander variations and even completely original landers. Without a proven offer, you just can't test anything original (if your original/custom lander doesn't convert, you wouldn't know whether it's the lander or the offer that sucks).

As far as lander and offer testing strategy is concerned, I'm about to write a lesson for the newbie tutorial on exactly that. In the meantime, here's a synopsis you can choose to follow as you please:

https://stmforum.com/forum/showthrea...l=1#post331567


Hope that helps! Nice start and I look forward to seeing more updates!




Amy

---

Home > POP / PPV / Redirect > Follow-along Campaigns