Home > >

---

Push Traffic, 1 Offer, Multiple Geos, $122k Revenue (31)

---

Today I want to show you one of my campaigns that made about $122k revenue on Push traffic.

First some things to clarify.
Maybe the title is a bit misleading for some people because it wasn´t only one offer in sense of only 1 offer ID on the network.
But it was all the same product so I would consider it as one offer anyway.

And here some facts about the campaign.

Offer: OneSafe PC Cleaner
Conversion Flow: CPI and CPS
Traffic Type: Push on several trafficsources
Bid Type: CPC, CPA, CPA Goal
Geos: UK, FR, DE, ES, IT, SE, IE, AU, NZ, CH, AT, ZA, US, OM, SA, BE, CL, QA, BH, KW, NO, FI, CA, NL, JP, SG, DK, IN, BO, CR, CU, DO, EC, GT, HN, NI, PA, PE, PR, SV, VE
Main Campaign Lifetime: About 2 months

Creatives: I tested different creatives, aggressive and not aggressive ones, with big image and without.
Usually the rule for that vertical should be "The more aggressive - The better" but surprisingly enough some non-aggressive creatives worked very good.
I had to test anyway because I couldn´t run aggressive creatives on all trafficsources.

I could imagine that maybe some users get scared away when the creative is too aggressive so that it´s better to run a good compromise between soft and aggressive.

Landing pages: I tested 3 different landing page styles with 2 variations per style and I also tested direct linking.
Out of the 3 LP styles one of them worked best in one variation across all geos but in one specific geo a different style killed it that didn´t work anywhere else.

I ran the winning landers then with different levels of aggressivity, depending on what the trafficsources allow.
On some sources I was allowed to run very aggressive when I tag my campaigns accordingly.

What surprised my was that direct linking worked best for the CPS campaigns.
That being said, by far most of the revenue was from CPI campaigns in combination with landing pages.

About the campaign lifetime, the main campaign lifetime was about 2 months where about 80% or so of the revenue was generated.
The real lifetime is a bit longer but when I started to run the offer it was on very low scale and at the end it also slowed down a good bit so the most traffic/revenue/profit comes down to about 2 months.

Let´s go to the stats now, these are numbers from the OneSafe offers in different Geos and sorted by revenue.
I won´t show which Geos exactly made most revenue because I still run other campaigns there:



Cost Trafficsource 1: $54,342.29
Cost Trafficsource 2: $33,719.76
Cost Trafficsource 3: $599.10
Cost Trafficsource 4: $6,398.95
Cost Trafficsource 5: $1903.64
Cost Trafficsource 6: $453.73
Cost Trafficsource 7: $1037.06
Cost Trafficsource 8: $2,191.76
Cost Trafficsource 9: $175.46
Cost Trafficsource 10: $307.67
Cost Trafficsource 11: $240.90

Total Cost: $101,370.32
Revenue: $123,727.31
Profit: $22,356.99
ROI: 18.07%

About the cost, I always set a higher % trafficloss in my tracker than it really is so the real ROI is about 5%-10%% higher and real profit should be around $30k or so.
The reason I set a higher trafficloss is that I prefer to have better real stats that what my tracker shows.
That way I don´t have a rude awakening when I keep something running just because my tracker shows inaccurate stats that look better than it is.
For that campaign I set 20% trafficlos for all campaigns in my tracker because for some specific reasons I assumed that trafficloss could be pretty high despite running push traffic.

As you can see I tested several different sources and most traffic came from only 2 sources.
Other sources were also good in terms of quality but they had not that much traffic.
And then there were also some sources that didn´t work that good for these campaign.

Twinaxe, why the hell are you so stupid and tell the exact offer when everyone says that the offer is the most important element in the campaign?
Do you really give us a golden egg on a silverplate?

It´s pretty simple, the offer is not available anymore and I couldn´t find a new offer that shows similar good results for me.
Otherwise I wouldn´t post this thread and better continue running the stuff myself.

---

Nice one twinaxe! This is a great example to prove that even "lower" ROI can result in very nice profits when done at solid scale

Thanks for sharing it.

---

Maybe another little takeaway from the campaign is that a big part of the revenue and profit was from pretty low payouts.

And it was only from Windows desktop traffic, man when I would have a comparable offer that would take mobile as well then the stats would probably look a bit different

---

1 Desktop Offer.

In over 40 Geos. Hundreds of landers.

Tested on 11 different sources.

Spent over $100k in 2 months.

To earn 20k in profits.

Jesus.

My only dumb question would be how did you come across the offer? AM suggestion? Top offer? Randomly selected?

Just curious.

---

Oh yes, there is much more mobile traffic compared to desktop... I'm not sure when this happened, but some placements have like 90% mobile and only 10% desktop, and I'm talking about regular websites, not some mobile apps

---

From where do you take the hundreds of landers?

I tested 3 different landers with 2 variations each and then ran in all geos but 1 the same lander just with different levels of aggressivity and also tested direct linking.

I first tested the offer about 1 1/2 years ago in PL but didn´t have success with it so I just moved on.

Then many months later I saw the offer few times in top offers for 1-2 geos.
I just tested again but didn´t have much success and only ran it on very small volume.

I still continued testing because I just wanted to get it running successful.
From that moment I went full in and started testing more creatives, many geos, sources and also tested CPI vs CPS.
In this time I also had quite a few days with $XXX in loss but I kept running because I knew it will be worth it.

Then i cracked it and started scaling, the result you can read above

Same here, I think that with some things I am too oldschool.

For example lots of pop traffic (and probably push as well) comes from streaming or filesharing websites.
I can´t understand why such sites have such high % of mobile traffic.

Really man, who the f*ck watches movies or series on the small phone screen?
A bit Youtube here and there ok, but full movies?

---

What a share twinaxe. It's awesome, thanks a lot. With posts like this I see how you structure and run your campaigns, it's very helpful.

And I was thinking that in most cases you test campaigns to just find something what work from beginning lol

---

It´s only how I ran this campaign, other campaigns I run completely different.

Unluckily not all campaigns can be treated equally.

And in most cases you are absolutely correct with that assumption

But here I just tasted blood and my killer instinct got activated.
I was really triggered by that offer and bite onto the stuff like a pitbull until I got it running successful.

Once I found my best setup to run it I didn´t need even need to spend much work on it anymore to run it on higher scale.

---

What was the change you made that got you front running it at a loss to profiting?

Sent from my YAL-L21 using STM Forums mobile app

---

Ahh, okey It's good to know. So probably for me it would be better to use strategy of launching a lot and kill faster than trying to grind it like you did above. As to do that I'd need more experience.

---

It wasn´t a change that got me from loss into profit.
It´s rather the opposite, I didn´t give up on it and kept it running until it worked

In the end it was the normal process a campaign goes through.
The high losses in the beginning just happened because I tested on high scale from the very beginning.

I also could have tested one geo after another and build it slowly that way.
Then I could use the stats from the first campaign for the next one, then the stats from the first 2 campaigns for the next one and so on and so forth.

That way I probably could have saved lots of money but it would have taken much longer to do so.

I am a rather impatient guy so I decided to just do some mass testing and tested about 20 geos or so at once.
The other geos I then started a bit later when I knew that every new geo I tested was a winner as well.
Also when I wanted to run it on scale I tested CPI and CPS together.

The above tactic gave much faster results compared to running everything one after another but it also resulted in few k $ loss before it took off.
That´s just normal so I was prepared for it but it´s definitely not recommended for everyone.

Then I stopped CPS and focused on CPI so that I could collect stats first to use with the CPS campaigns.
The reason I tested CPS that early in the campaign already was that it converted but in the end it didn´t convert good enough to run it basically on RON.
And with CPS payouts about $40-$50 it can become pretty expensive

When I then concentrated on CPI I could scale alot more and performance increased and at the same time it gave me more stats to use for the CPS versions of the offers.

Another thing that increased the performance a good bit was that I first built a global blacklist from all geos combined.
That worked very good already to get rid of the worst stuff but of course it´s not that accurate as running geo specific blacklists.

When I had few hundred conversions in a geo I created BL only for these geos and that helped a bit for better performance as well.

The approach that I did with this campaigns is probably rather for people with more experience and a higher budget to spend.

When you run such mass tests it´s normal that you can lose lots of money in the beginning so you should be dead certain that it´s worth it and for this you just need some experience.

I knew that the vertical is doing good and I was 100% confident that I can get it running.
When there would be just a touch of a doubt I wouldn´t run it that way.

---

I figured you were translating the geos for each language right? That counts as a separate lander for each language in my book

---

Oh, in that case you are almost right.

I had it translated to several different languages but for example all the LATAM geos speak Spanish, then for IN and the NORDICS for example I used English creatives and lander and didn´t even bother to get it translated to local languages because it worked pretty good already.

So in the end it weren´t that many different landers.

---

Fair enough. Eyeballing your geos, I'll let you slide with 12 languages x 3.7 landers = 44 landers total

---

It were 5 languages.

EN, DE, FR, ES, IT.

I basically tested only two landers for each language in two variations.

I tested the third lander a bit later and it only needed very low budget because I had enough stats already to compare the performance.

I can imagine that the whole stuff probably looks like it was a huge workload.

But after the initial phase where I really had to spend quite some time and money and energy it was very easy to keep that stuff running at higher volume.

---

This is great. Answers a lot of questions for me

---

Great share! I actually made a bank with this exact offer on ZeroPark (POP) about 3 years ago when it was available on GlobalWideMedia. CPI only, never got around testing CPS. I think the top geos were /DE and /AR. It comes and goes on different networks, so I wouldn't entirely write it off just yet.
Plus I find, at least for the CPI version the other "pc cleaning" offers converts almost as well. You know the kind that changes name every 6 months but has the same landing page.

---

I actually have a white label from the maker of this product that I have let sit dormant because I lost my ass on it running it on Adwords. This has piqued my curiosity in perhaps running it again

You might want to reach out to them about doing a white label of your own, mainly because of the renewals. Mine recurs every six months, and I still get renewals from when I first launched it in the fall of 2015

---

I´m not sure if it comes back this time although I would be very happy about it.

I also tested Reimage, Outbyte, Tweakbit but Onesafe was the best.
It´s pretty funny to see such differences between the products when you see that their landers basically look all the same.

Nice, didn´t ever think about running a whitelabel.

And these are some nice LTV, I know such long times only from some adult programs that I promoted.

---

why we use more then one traffic sources

---

Thanks twinaxe. You are the man

---

One word: Scaling

When the campaign is doing very good you want to get as much traffic as possible.

First I try to scale in different ways on the original source but then it´s good to use other trafficsources as well for maximum volume.

---

This is so epic, I just have 1 question.

I know you tested fast and loose to get data quickly but how deep (in terms of $) was your loss in testing at it's lowest point before your camp clawed out of red overall.

---

So basically, if a landing page works on Push, it should usually also be good in other GEO's right? So generally you would recommend to just translate that lander and take it to another GEO?

---

Hard to tell, in the very beginning I started the offers on pops but wasn´t much volume.

I quickly switched to push then because the pop quality wasn´t good enough.

On pops it was on lower scale anyway but always break even or little profit like $20/day.

When I saw that the CPI offers converted that good on pops I immediatley tested it as CPI and CPS when I switched to push.

There I had days with $-300 or so, maybe more maybe less, can´t remember exactly anymore

That was mostly because I spent much money on CPS although I didn´t have good BL/WL for it yet.

When I stopped the CPS offers the performance/profit increased.

The CPI offers were doing very good from the beginning so there weren´t much losses on my main source.

When I then started running it on another big source I had more losses because it was one of my "scaling sources" with no user activity/freshness targeting.

So there I had to spend some money first to identify the good placements and sources for a WL, then the performance there was very good as well.

I wouldn´t generalize it for each and every campaign but here it worked.

For many geos two LPs were basically head-to-head so that I ran them together for some time until I decided which one is better.
In most cases it was always the same one.

I had a third LP that I also tested but it tanked completely in all geos but there was one geo where the third LP outperformed the others.

Anyway, back to your question.

In the end there are not that many different LPs for that vertical anyway.

It´s similar to sweeps where we have questionnaires, spinwheel, gift box and maybe gift envelopes.

It doesn´t matter in what geos you run your campaigns, mostly one of these landers will always work.

Here it´s similar, there are not that many different LPs for that vertical so one of them will mostly (or even always) work.

---

Great share dude! Even i tried promoting outbyte offer, but while i was doing direct linking to offer page i lost $50 on first day with 0 conversions, so, i gave up the offer.
On clickdealer i saw this offer in top 10 offer list many times and was often tempted to promote this offer but i always resisted.

---

Direct linking isn´t working good for these offers.

There are basically 2-3 LPs that work good.

---

Very interesting Twinaxe, thanks

The chalanges our affilates have today in such offers are:
1. In 2022, there is not much, or not at all, CPI offers for Desktop Software. Not even for Pc cleaners which used to be very CPI-Friendly

2. In 2022 the trending (and avaliable) offers for affialtes are anti-viruses, not pc cleaners.
Pushub team members that have been Runing pc cleaners all the way back from 2012 (Reimage, Systweak, Uniblue, Avanquest, etc) share that Some PC cleaners closed down or just became to small to be relevant, other moved to internal media buying to modrate ad quilty so they wont get burned by household name Anti viruses and main Browsers. All of them dropped CPI as it was not bringing enoguh sales and up-sales like before.

Thanks
Mark
Pushub support

---

Hey man could you elaborate on a few things for scaling?

Every time I fuck around with push, I frequently get green camps that do $5 to $10 profit but I have no idea what to do in regards to scaling them.

You mention the campaign timeline was 2 months, what was your budget on day 1? 2? 7? etc.
If you could roughly let me know what your process is for scaling, eg. 15% every 16 hours

How important is a WL/BL for this?

---

I’m not twinaxe. But. Hard to say without knowing what you’ve tried so far.

Did you start with high quality / freshness levels only first? Start building a BL on that?

Then expand to medium and keep blacklisting?

And then take all the converting feeds from those two and put them in a WL and start a new camp on low quality?

Or. Did you just select all and run RON?

---

There are also PC cleaners, even very good ones but not public

Yup, I started running Norton in beginning of 2020 and it was doing pretty good.

Payouts for the offers are also excellent.

Nowadays McAfee seems to be doing just as good as Norton.

When I have a good funnel there is no limit for budget, I just try to spend as much as possible.

I also created many campaigns each day to don´t miss any traffic.

For that offer I always ran BL on high activity/fresh users and WL on low activity/older users.

---

Home > >