Home > > Newbie Follow-Alongs

A Guy From Lithuania Gets Rich or Dies Tryin' (41)


07-11-2020 08:29 PM #1 mantas (Member)
A Guy From Lithuania Gets Rich or Dies Tryin'

So, my journey continues..

That's how I feel about it:



I have launched my first sweepstakes campaign on PropellerAds. Testing 5 landing pages, 3 offers. Payouts: $5, $0.24 and $0.24 .

After spending $14.79 and getting 0 conversions I have stopped the campaign.
32% click-loss.

Landing Pages:


Offers:


After investigating the Adplexity, I have found that the offer is being run on another traffic source without any landing page. So I am preparing to launch a new campaign testing that new traffic source. Probably I will be direct linking also and I will test another 3 offers from a different vertical than sweeps because they are are the best offers in my CPA network right now and in the same geo and carrier.


07-11-2020 09:55 PM #2 jeremie (Moderator)

Hey,

Try to include a bit more info. such as the geo, the type of traffic and the traffic source, because without these info, it is hard to tell if you tested enough,
For example, as you got no conversion, have you tested that your post back link is properly configured?

Quote Originally Posted by mantas View Post
So I am preparing to launch a new campaign testing that new traffic source. Probably I will be direct linking also and I will test another 3 offers from a different vertical than sweeps because they are are the best offers in my CPA network right now and in the same geo and carrier.
I can tell @twinaxe is not going to be happy


07-11-2020 09:59 PM #3 jeremie (Moderator)

Here are 2 articles that could help you

https://stmforum.com/forum/showthrea...-Read-This-Now
https://stmforum.com/forum/showthrea...n-starting-out


07-13-2020 02:29 PM #4 twinaxe (Senior Moderator)

Hey there, thanks for giving us the chance to take part on your journey

Launched my first sweepstakes campaign. Testing 5 landing pages, 3 offers. Payouts: $5, $0.24 and $0.24 .
This looks strange for me, I guess the $0.24 offers have same conversion flow but what about the $5 offer?

Do you run SOI offers together with Pin submits or CC submits or so?

After investigating the Adplexity, I have found that the offer is being run on another traffic source without any landing page.
What do you mean with "the offer"?
Above you said you were testing 3 offers so is the offer one of hese or something completely different?

After spending $14.79 and getting 0 conversions I have stopped the campaign.
Why did you decide to stop at this point?

Did you set a test budget before?

For beginners I recommend to use a rule like this one

Number of landing pages x Number of offers x Average offer payout x 10 = Test budget

That way you use the same approach for your campaigns, this can help alot to develop a clean system compared to just run random amounts for each campaign.

In your example it would be

5 landing pages x 3 offers x $1.83 x 10 = $274.5

You see, it´s way too much for low payout offers like the $0.24 ones.

It´s better to keep different conversion flows separated and rather run similar offers together.

About the landing pages, what kind of landers did you test?

In the beginning it´s enough to test different landing page styles to see if something is working good.

For sweeps there are basically only quiz/questionnaire landers, spinwheel and gift box.

So I am preparing to launch a new campaign testing that new traffic source. Probably I will be direct linking also and I will test another 3 offers from a different vertical than sweeps because they are are the best offers in my CPA network right now and in the same geo and carrier.
It´s great when you want to test different things but take care to don´t create a big mess so that you end up testing just random stuff without any system.

This will only cost you money and chances to succeed that way are not very high.

It´s better to decide what traffic to run, what trafficsources to use, then use CPA networks who have offers for it.

Then try to find offers that match your traffic and don´t run offers just because they are the best offers in your network.


07-17-2020 05:50 AM #5 mantas (Member)

Quote Originally Posted by jeremie View Post
Hey,

Try to include a bit more info. such as the geo, the type of traffic and the traffic source, because without these info, it is hard to tell if you tested enough,
For example, as you got no conversion, have you tested that your post back link is properly configured?



I can tell @twinaxe is not going to be happy
I was worried about the competition but since I am not profitable there is nothing to hide
I am running ZA on PropellerAds - Vodacom carrier


Quote Originally Posted by twinaxe View Post
Hey there, thanks for giving us the chance to take part on your journey

This looks strange for me, I guess the $0.24 offers have same conversion flow but what about the $5 offer?

Do you run SOI offers together with Pin submits or CC submits or so?
I am not sure about the conversion flow of the $0.24 offers.. I have just wanted to test more offers in one campaign so I have added these two low payout offers. After spending about $2.90 on each offer and getting no conversions I have stopped those two lower payout ones earlier in the Voluum campaign. That's why I don't worry about the big difference in the payouts - it is easy to stop the offers in the campaign anyways.

Quote Originally Posted by twinaxe View Post
What do you mean with "the offer"?
Above you said you were testing 3 offers so is the offer one of hese or something completely different?
Sorry for my pure communication skills
I have had in mind the $5 offer as it is one of the top offers in Mobidea.

Quote Originally Posted by twinaxe View Post
Why did you decide to stop at this point?

Did you set a test budget before?

For beginners I recommend to use a rule like this one

Number of landing pages x Number of offers x Average offer payout x 10 = Test budget

That way you use the same approach for your campaigns, this can help alot to develop a clean system compared to just run random amounts for each campaign.

In your example it would be

5 landing pages x 3 offers x $1.83 x 10 = $274.5
I can't wrap my head about spending so much. If I am looking for a funnel(one combination = traffic + landing page + offer) which would be between 100% and 200% ROI why do I have to spend 10 times on it if it does not convert after 1x 2x or 3x the payout? If something works it works. Isn't it it more effective to spend less on each test but to test broader and new stuff?


07-17-2020 12:40 PM #6 twinaxe (Senior Moderator)

I am not sure about the conversion flow of the $0.24 offers.. I have just wanted to test more offers in one campaign so I have added these two low payout offers. After spending about $2.90 on each offer and getting no conversions I have stopped those two lower payout ones earlier in the Voluum campaign. That's why I don't worry about the big difference in the payouts - it is easy to stop the offers in the campaign anyways.
It´s not good to run offers with so huge difference in payout together.

It´s better to run similar offers together, that way it´s easier to make decisions.

Sorry for my pure communication skills
I have had in mind the $5 offer as it is one of the top offers in Mobidea.
Alright, thanks for clarification.

But in the beginning I recommend to work with lower payouts.

There you can test much more and learn much more for the same money compared to running higher payouts.

Especially when you say yourself

I can't wrap my head about spending so much.
Learning and testing just takes some investment and generally we can say that the higher the payouts the more you have to invest.

If I am looking for a funnel(one combination = traffic + landing page + offer) which would be between 100% and 200% ROI
Good luck with it.

Running 100%-200% ROI on scale isn´t realistic.

Better get used to the idea of running 10%-20% when you want to run on high volume.

I know, it probably sounds a bit extreme but the reality is that you mostly have to run on lower ROI when you run on higher scale but the profit from the high volume will still be better than the profits from high ROI on low volume.

why do I have to spend 10 times on it if it does not convert after 1x 2x or 3x the payout?
You don´t have the funnel yet, you still need to find it with testing different landers and offers and targetings.

As long as you don´t have the winning combination of 1 LP + 1 offer you need to spend some money to find it.

And when you have more unknown variables in the calculation you need to spend more.

In the previous example you have the situation that you want to test 5 landers, 3 offers and you don´t have a good BL yet so these are all unknown variables that you have to consider.

When you only spend 1x or 2x payout before you stop you never know if it´s the lander, the offer or the traffic/targeting itself that won´t work.

When you start tests and then stop them too early you should better not run them at all because then you waste money on tests that don´t reach significant enough numbers to decide anything and the results are basically worth nothing then.

If something works it works. Isn't it it more effective to spend less on each test but to test broader and new stuff?
Of course the best situation that can happen is when you start a test with several landers, several offers on RON traffic without a BL and you find a combination that works right away.

But that doesn´t happen often and hoping for it is like gambling.

Again, when you just start from scratch without any good stats from previous campaigns you don´t even know what element in the campaign isn´t working.

The first tests often are the most expensive ones but they also help you to keep costst lower for future tests when you run them properly.

For example when you want to run sweeps in a specific geo and you identified a good converting lander there then you can often use the lander for very long time.

When we now check the example calculation from before and compare it it´s like this:

This is the original calculation - 5 landing pages x 3 offers x $1.83 x 10 = $274.5

Now let´s check how it looks when you already trimmed down to one landing page

1 landing page x 3 offers x $1.83 x 10 = $54.90

You see, it´s 5 times less already.

When you then run some campaigns and have a good blacklist you can also change the factor 10 to factor 5 because you don´t need to exclude that many placements anymore.

Then the calculation would be like this

1 landing page x 3 offers x $1.83 x 5 = $27.45

When you invest in proper tests in the beginning the future tests will be much cheaper and also faster.

But when you don´t run proper tests in the beginning and always stop them at random times and without spending enough then the test results are basically void and you have to start from scratch again and again.

This will cost you much more in the long run.


07-17-2020 04:35 PM #7 skolvikings (Member)

Quote Originally Posted by twinaxe View Post
When you invest in proper tests in the beginning the future tests will be much cheaper and also faster.

But when you don´t run proper tests in the beginning and always stop them at random times and without spending enough then the test results are basically void and you have to start from scratch again and again.

This will cost you much more in the long run.
Is it also fair to say then, that if you keep hitting the threshold on your tests (say 50% of planned spend with no conversions), and you don't know what is the problem, you just have to keep talking to your AM trying to get good offers and keep testing more stuff, until you finally do get at least one part of the funnel working, and then go from there?


07-18-2020 02:33 PM #8 twinaxe (Senior Moderator)

Quote Originally Posted by skolvikings View Post
Is it also fair to say then, that if you keep hitting the threshold on your tests (say 50% of planned spend with no conversions), and you don't know what is the problem, you just have to keep talking to your AM trying to get good offers and keep testing more stuff, until you finally do get at least one part of the funnel working, and then go from there?
More or less yes.

By far not everything we test is a winner and when we have no idea what element in the tunnel didn´t work it´s often best to test something else.

Most important would be to test new offers because offers have a higher impact on the campaign performance than landers and placements.


09-12-2020 03:42 PM #9 mantas (Member)

Hey, it's me again!

So I have started a new campaign in the different GEO. Why? Because I have started this follow along when I have not even went through half of the 40-day tutorial. Sorry @vortex. Don't get mad at me I am action taker and this time I have started too early only to find there are more golden nuggets later in the tutorial.

First of all, I will be following a testing strategy provided in the tutorial to the tee this time.

Structure of the campaign: 7 landers, 9 offers(from two different networks), offer payouts range from $0.83 to the $2.59

Also, I have cancelled the Adplexity because I think it is not worth the monthly payment for the value it gives me. I have downloaded a few sweepstakes landers in the ZA a month or two ago. I really doubt that the angles changed dramatically during this time. So, I have simply edited my ZA landers for the other GEO I am currently running in.

Here are the results after a few days:



Some hours later two offers have reached 2 conversions(both from the same network).




Chosen the offer with the higher payout and proceeded to test the landing pages.

Unfortunately, what seemed a quit well converting offer in the beginning, it has failed completely later for whatever reason. It had started converting worse than 2 in 10x adpsend.

So I had stopped the offer and started the landing page testing again with the other offer which had reached 2 conversions in the beginning.

And for the second time in the row, this offer is also a complete failure.



Only 3 conversions in the 20x adpsend. Maybe it could be a problem with an affiliate network doing some shady stuff(why the offers started so well and failed later so badly?) So I am going back to the initial stage of offer testing again. This time I will only test the offers from the another network.

Overall stats for the campaign:


09-15-2020 11:44 AM #10 twinaxe (Senior Moderator)

Structure of the campaign: 7 landers, 9 offers(from two different networks), offer payouts range from $0.83 to the $2.59
Why is there such a big difference ($0.83 vs $2.59) in the payout?

Do you run offers with different conversion flows together in one campaign?

Also, I have cancelled the Adplexity because I think it is not worth the monthly payment for the value it gives me. I have downloaded a few sweepstakes landers in the ZA a month or two ago. I really doubt that the angles changed dramatically during this time
Sure, you don´t have to keep it running continuous, just grab a good collection of landers there and you often can use them for quite some time.

You are absolutely right, lander trends often don´t change that fast and landers can sometimes work for years.

Keeping an ongoing Adplexity subscription however is helpful when you want to use Adplexity for more than "just" downloading landers.

Then it can be really good as a real research tool to also spot good opportunities pretty fast

Maybe it could be a problem with an affiliate network doing some shady stuff(why the offers started so well and failed later so badly?)
Mostly the reasons are way simpler than that

Well, you only posted very basic stats.

Did you also check the placements for high volume placements that only eat budget but don´t convert?


09-16-2020 06:16 PM #11 mantas (Member)

Quote Originally Posted by twinaxe View Post
Why is there such a big difference ($0.83 vs $2.59) in the payout?

Do you run offers with different conversion flows together in one campaign?
No, all of them are SOI offers.


Quote Originally Posted by twinaxe View Post

Did you also check the placements for high volume placements that only eat budget but don´t convert?
Yes, I am continuously blacklisting the placements that are in loss by the 2x payout.


the grey ones are the blacklisted placements.

the grey ones are blacklisted



09-16-2020 07:59 PM #12 twinaxe (Senior Moderator)

No, all of them are SOI offers.
Strange that they have such different payouts.

Yes, I am continuously blacklisting the placements that are in loss by the 2x payout.
At what step are you to do so?

Testing the "winning" offer with 7 landers?

Because in this step you can't really pause placements for bad quality when you don't know for sure which element in the campaign isn't converting.

When you test 7 landers and stop placements at 2x payout without conversions then each lander receives only traffic for 1/7 of the adspend.

This would be only 2/7x offer payout per lander.

When you blacklist too aggressive you risk to cut good placements.

Also, how did you calculate budget for the campaign?

And do you run the campaign with a systematic approach or do you decide by gut feeling when to pause landers, placements and stuff?

What's your general campaign setup like targeting and stuff?

Do you set a test budget and let the campaign just run to collect stats or do you also stop or change campaigns during running tests?

And are the 7 landers you test different landing page styles or are there also variations of same LP styles?

Really, with the info provided it's hard to tell what's going wrong, maybe it's just the offers that aren't good enough.


09-16-2020 09:03 PM #13 mantas (Member)

Quote Originally Posted by twinaxe View Post
Strange that they have such different payouts.

At what step are you to do so?

Testing the "winning" offer with 7 landers?

Because in this step you can't really pause placements for bad quality when you don't know for sure which element in the campaign isn't converting.

When you test 7 landers and stop placements at 2x payout without conversions then each lander receives only traffic for 1/7 of the adspend.

This would be only 2/7x offer payout per lander.

When you blacklist too aggressive you risk to cut good placements.

Also, how did you calculate budget for the campaign?

And do you run the campaign with a systematic approach or do you decide by gut feeling when to pause landers, placements and stuff?

What's your general campaign setup like targeting and stuff?

Do you set a test budget and let the campaign just run to collect stats or do you also stop or change campaigns during running tests?

And are the 7 landers you test different landing page styles or are there also variations of same LP styles?

Really, with the info provided it's hard to tell what's going wrong, maybe it's just the offers that aren't good enough.
I blacklist placements at any step. But I look at the stats of the placements of the overall campaign from the day 1. I get it what you mean now... It is too aggressive, indeed, especially when the offers are not converting.

But... Isn't it how it should be done according to the Amy's 40-day tutorial? That's what I am following. https://stmforum.com/forum/showthrea...l=1#post363258


09-18-2020 11:49 AM #14 twinaxe (Senior Moderator)

I blacklist placements at any step. But I look at the stats of the placements of the overall campaign from the day 1.
You can blacklist placements at any step but you can blacklist for low quality only when you have a good converting funnel.

When you blacklist in test stage you don´t know if the placement is low quality or if the LPs or offers are weak.

Would be great when you could answer the questions from my previous post as well, otherwise it´s next to impossible to give good help


09-20-2020 04:43 PM #15 mantas (Member)

Quote Originally Posted by twinaxe View Post
You can blacklist placements at any step but you can blacklist for low quality only when you have a good converting funnel.
After the realization that I have been blacklisting when I did not have a good-enough converting funnel, I have decided to launch a campaign from scratch again.

The first campaign with one offer has completely failed. Then, the same day I have launched a similar campaign. The only difference was the offer. And guess what... it is profitable now!



So, I have just learned that the most important thing is a good offer. There is no point of blacklisting and optimizing if the offer sucks... Offer is the KING!

Quote Originally Posted by twinaxe View Post
Also, how did you calculate budget for the campaign?
I just stop the campaign when after the most recent 10x payout in adspend the offer is converting worse then 2 conversions. Actually, after the realization of the offer importance, I am coming to a conclusion that offer sucks if it does not convert once after 5x the payout adspend.


Quote Originally Posted by twinaxe View Post
What's your general campaign setup like targeting and stuff?
I target only the wifi traffic and exlude Android Webview.

Quote Originally Posted by twinaxe View Post
Do you set a test budget and let the campaign just run to collect stats or do you also stop or change campaigns during running tests?
I tweak the campaign during the test. First phase is to get an offer which gets 2 conversions first - then stop all other offers. Proceed to the landing page testing phase. Cutting the lander when it is 90% confidence that the other lander is better.

Quote Originally Posted by twinaxe View Post
And are the 7 landers you test different landing page styles or are there also variations of same LP styles?
All of them are as different as possible - they have different angles so to say.


09-21-2020 12:00 PM #16 twinaxe (Senior Moderator)

The first campaign with one offer has completely failed. Then, the same day I have launched a similar campaign. The only difference was the offer. And guess what... it is profitable now!
That´s great man, it can´t be said often enough how important the offer is.

There is no point of blacklisting and optimizing if the offer sucks
Exactly, a great offer will even convert somewhat on crap traffic but a crap offer won´t convert on the best traffic.

I just stop the campaign when after the most recent 10x payout in adspend the offer is converting worse then 2 conversions. Actually, after the realization of the offer importance, I am coming to a conclusion that offer sucks if it does not convert once after 5x the payout adspend.
Especially in the beginning it can help to use rules like the one I already mentioned some time ago

Number of landing pages x Number of offers x Average offer payout x 10 = Test budget


With that rule the test budget can look pretty high sometimes but the more you test the higher the cost.

Some tests can also be stopped before the budget runs out when a specific condition is met and when you then calculate the budget for the next test it´s already much lower.

You can also use some kind of stop loss like "5x average offer payout and 0 conversions = stop offer or landing page".

That way you will never lose the whole test budget.

I tweak the campaign during the test.
What kind of stuff do you tweak on a running test?

Stopping offers or landers when they just don´t work is ok when your testing these elements but changing targeting or so in running tests isn´t good because it can lead to completely different results then so that the original stats could be worthless then.


09-21-2020 05:47 PM #17 mantas (Member)

Quote Originally Posted by twinaxe View Post
What kind of stuff do you tweak on a running test?

Stopping offers or landers when they just don´t work is ok when your testing these elements but changing targeting or so in running tests isn´t good because it can lead to completely different results then so that the original stats could be worthless then.
I blacklist the placements if it is in loss by 2x the payout.
I increase the Smart CPM to get traffic from more expensive placements and I also change the traffic delivery from distributed to standard if the daily adspend is too low.


09-22-2020 07:01 PM #18 mantas (Member)

And.... I am facing the same main issue I had been facing a few years ago. The reason why I had given up on AM. The campaign making above $10 a day, breaks even the second day, and starts losing money the third day and only getting worse...

Sunday:


Monday:


09-23-2020 04:50 AM #19 larsometer (Senior Member)

Seems like you have to solve a big road block.

Maybe it makes sense to provide some more details so that the pros can chime in and help.

Do you have data from tracker that shows the placements (you can cover some digits) in a day to day comparison?


09-23-2020 01:45 PM #20 twinaxe (Senior Moderator)

I increase the Smart CPM to get traffic from more expensive placements
I don´t recommend to make such changes in running campaigns.

On one hand it can mess up the algorithm and kill even good campaigns, on the other hand you receive different traffic on different bids so the previous test results would be invalid.

It´s better to run a new test for it.

The campaign making above $10 a day, breaks even the second day, and starts losing money the third day and only getting worse...
Impossible to say anthing with just these stats.

Did you only check the total performance or did you also check the placements?

Quote Originally Posted by larsometer
Maybe it makes sense to provide some more details so that the pros can chime in and help.
Exactly, otherwise it´s just guesswork


09-23-2020 05:56 PM #21 mantas (Member)

Quote Originally Posted by larsometer View Post
Seems like you have to solve a big road block.

Maybe it makes sense to provide some more details so that the pros can chime in and help.

Do you have data from tracker that shows the placements (you can cover some digits) in a day to day comparison?
Stats for the placements. Note that the payout is $0.70

Sunday:


Monday:


Quote Originally Posted by twinaxe View Post
on the other hand you receive different traffic on different bids so the previous test results would be invalid.
Why would it be invalid? Each time the Smart CPM gets changed, the traffic to the campaign changes, timewise. But that traffic is divided into the equal parts (when testing 5 landers for example - it is split into the 5 parts).

Unless you theorize that for a certain CPM - one lander might be the best but for the another CPM - actually a different lander might be the best. But how likely it is to happen in practice?

Quote Originally Posted by twinaxe View Post
Impossible to say anthing with just these stats.
I have made some changes to the campaign on Monday. At the midnight, I have added a few other landers to start a split test round. After 5-6 hours the lander from day before have proved to be the winner anyways. So the rest of the day the campaign had only the winning lander(same lander from the day before).

Maybe this could have influenced the drop of ROI? And maybe the fact that Sunday is the day of the weekend?


09-24-2020 07:22 AM #22 jaybot (Veteran Member)

Quote Originally Posted by mantas View Post
Why would it be invalid?
Because anytime you touch the bid it fucks up the Propeller algorithm and your camp will never be the same again.

This can also be true with creatives, zones, languages, devices, hair color, music preference, and simply looking at the screen funny while using propeller.

Best to just start a new camp on propeller each time you want to change something.


09-25-2020 07:23 PM #23 mantas (Member)

Quote Originally Posted by jaybot View Post
Because anytime you touch the bid it fucks up the Propeller algorithm and your camp will never be the same again.

This can also be true with creatives, zones, languages, devices, hair color, music preference, and simply looking at the screen funny while using propeller.

Best to just start a new camp on propeller each time you want to change something.
But if you constantly launch new campaigns after each change, you will never get significant data on the placements. That's the main reason I try to work with the same campaign.

Or maybe you are speaking about a campaign which is not in the testing stage and is already profitable. Maybe then yes, I would not like to touch something in campaign which is already green.


09-26-2020 05:00 AM #24 jaybot (Veteran Member)

Quote Originally Posted by mantas View Post
But if you constantly launch new campaigns after each change, you will never get significant data on the placements. That's the main reason I try to work with the same campaign.

Or maybe you are speaking about a campaign which is not in the testing stage and is already profitable. Maybe then yes, I would not like to touch something in campaign which is already green.
Nah. Just combined the numbers in excel or notepad or whatever. Your placements are still valid.

I know it sounds silly, but we’re being serious about Propeller’s ‘Smart’ algorithms being a finicky bitch. At a minimum, if you change the bid or targeting, you absolutely should start a new campaign. This goes for pops and push CPC or CPA and smart CPM.

The only exception would be straight CPM on pops.


09-26-2020 06:56 PM #25 plutus (Member)

Quote Originally Posted by jaybot View Post
Because anytime you touch the bid it fucks up the Propeller algorithm and your camp will never be the same again.

This can also be true with creatives, zones, languages, devices, hair color, music preference, and simply looking at the screen funny while using propeller.

Best to just start a new camp on propeller each time you want to change something.
But wouldn't creating completely new campaign with the same tracker url do the job? Or is it required to setup everything from the scratch, even on the tracker side?


09-27-2020 05:19 PM #26 mantas (Member)

What up money makers!!! So I have started to gain some traction during this week. As the money is moving around fast I have decided to make weekly roundups on every Sunday. It will be kind of like a weekly accounting.

September 20-26


Revenue from the Mobidea: $494.83 Revenue from the another network: $11.33
Total Revenue: $506.16

Costs from the PropellerAds: $449.42 Costs from the another traffic source: $50.85
Total Costs: $500.27

Profit: $5.89 ROI=1.18%

Not a lot profit wise but I am happy with the increased revenue

P.S. Thank you for all your help super affiliates! And sorry that I have not provided a more detailed information for you in order to help me even more. I just have not had time to go into too much depth and this week has been REALY busy for me. My head is already spinning and I have experienced that it is easy to get into analysis paralysis in this affiliate game. So I am trying to balance out action taking with the thinking.


09-27-2020 06:14 PM #27 jaybot (Veteran Member)

Quote Originally Posted by plutus View Post
But wouldn't creating completely new campaign with the same tracker url do the job? Or is it required to setup everything from the scratch, even on the tracker side?
Hitting duplicate in the tracker isn't that much effort. Both work fine. One campaign link is definitely easier to keep track of in the tracker.

But. Some believe the traffic source will mess it up if it finds a duplicate campaign link.


09-30-2020 04:38 PM #28 twinaxe (Senior Moderator)

But wouldn't creating completely new campaign with the same tracker url do the job? Or is it required to setup everything from the scratch, even on the tracker side?
Of course you can also send everything to the same tracker campaign.

But different settings and bids can perform pretty different so I prefer to keep it more separated.

Not a lot profit wise but I am happy with the increased revenue
Doesn´t matter if it´s much yet or not.

It´s much more important to see that you are on a good way

Once you find your flow the numbers will increase from alone.


10-05-2020 05:41 AM #29 mantas (Member)

September 27 - October 3

Total revenue from 3 different networks: $201.60 + $4.16 + $121.62 = $327.38
Total ad-spend: $282.24 + $4.84 = $287.08
Profit: $40.30

It is $34 profit or 584% increase compared to the last week.

Feels good...


10-05-2020 09:53 AM #30 twinaxe (Senior Moderator)

Quote Originally Posted by mantas View Post
September 27 - October 3

Total revenue from 3 different networks: $201.60 + $4.16 + $121.62 = $327.38
Total ad-spend: $282.24 + $4.84 = $287.08
Profit: $40.30

It is $34 profit or 584% increase compared to the last week.

Feels good...
Sounds good man

Now let´s work on keeping that numbers stable and then scale to more.


05-15-2021 06:20 AM #31 jaybot (Veteran Member)

I broke STM forum's database somehow


05-15-2021 06:22 AM #32 jaybot (Veteran Member)

Quote Originally Posted by vortex View Post
I just want to say - I completely understand your frustration.

Everyone has their own experience. We each have our favorite geos and verticals, so our opinions may be biased in the sense that they may only represent a certain group of geos/verticals.

Also: The law of attraction is VERY real! For example I may have read that webview traffic had a tendency to convert poorly, and then I "happened" to pick offers and landers that converted very poorly for webview traffic. What started as a suspicion would attract similar realities which would then turn the suspicion into a belief. It happens all the time.

The only common denominator across all successful affiliates (at least for pop and push) is that we all test extensively. No matter how many cooks are in the kitchen, this is something we'll all agree on.

The trick is to do all this testing efficiently so that you're not wasting money doing it. For the same test budget you always want to test as many things as possible, yet still giving all the variables a chance to perform. Don't spend so much money on testing any one thing that you're striving for 100% accuracy - it's just not worth it.

i.e. Don't test an offer until you're 100% certain it's a dud. If it doesn't start converting well from the very beginning, and there's not a lot of room for optimization in your campaign, such that you can make big improvements in ROI by cutting some major traffic segments, then forget it. There are so many other offers you can test!

Spending $46 on an offer with $0.20 payout is indeed way too much. I wouldn't have spent more than $2 on testing it unless I was getting -50% ROI or better, preferably better. (Assuming I was either direct-linking, or was using a proven winner lander.) For that budget you could have tested 20+ offers with similar payout. I say this with a ton of love - not meaning to judge at all. Every single affiliate needs to go through this.

And let's not forget that campaign performance can and often does decline over time, which means that even if you had gotten -50% ROI on your first 10x payout in spend, you may not be able to get the same performance hours or days from the initial test. All the more reason to set the bar high in the beginning in terms of how well your funnel needs to perform, in order to qualify for further optimization.

I talked about the importance of spending more money on testing than on optimization here: https://stmforum.com/forum/showthrea...l=1#post416427

Also: If you don't yet have a good blacklist for a certain geo, try to run offers with low payouts. Under $2, preferably under $1 even. The larger the payout, the more money you'll need to invest in order to generate the blacklist. Use a low-payout offer for this, and THEN test higher payout offers on the good remaining placements.

Hope that will provide some insight on how to spend your budget more wisely!



Amy
This is good advice


05-15-2021 03:51 PM #33 jaybot (Veteran Member)

I see my name getting thrown around a lot, so I think I should chime in.


First of all, as I constantly say: no one should ever use me as an example of how to make money. I'm as retarded as anyone and only good at losing money most of the time.


Quote Originally Posted by mantas View Post
@twinaxe, so a beginner should focus on mobile content offers with a sweepstake angle then? Isn't it a good idea to test SOI anyways like in my case just because the AM recommended so?

A beginner should focus on testing offers. SOI/PinSubmit/CCSubs/Sweeps/Adult/Dating/Casino/Crypto/Whatever you want.


Just be sure you have a plan and a budget.


SOI are usually cheap. That's why they are suggested to beginners. To save money on testing.


But you have to either a) do the research on the geo on payouts and traffic costs + volumes. or b) ignore that shit, and fuck up and lose money to learn your lesson. I somehow manage to do both most of the time.


Most AMs are absolutely dumb as rocks when it comes to what offers work with what traffic. Just assume anything an AM recommends you is converting for an expert verteran affiliate using FB traffic and doesn't apply to you. All you can do is try to guess what might work by looking at the top offers suggested and figuring out what kind of offers/geos are a waste of your time.


This even applies to direct networks/offer owners such as Haka. AMs see what geos/carriers are doing best. They have very little idea how or where they are being run. So they just suggest whatever the biggest numbers are telling them.


Quote Originally Posted by mantas View Post
Yes, it was too much for sure. But I read a lot of posts how Jaybot runs CPA Goal campaigns immediately so I wanted to test the same. And that's the worst thing about CPA Goal campaigns when the offer is not converting. Asked for the advice on when to test the CPA Goal in this post https://stmforum.com/forum/showthrea...l=1#post416243 but it has been missed somehow

Sorry. Was getting rejected in all sorts of airports this week trying to get 6 people to JP in the middle of a pandemic. Unfortunately failed, but didn't have time for STM and AM stuff. I'll answer over there in a bit.


Again. It's good to test what others do (except me, I'm an idiot). But as others have said, set an emergency break on that shit. If an offer is 0.20 payout you shouldn't lose $40 on the offer if you set the daily budget to $10.




Quote Originally Posted by mantas View Post
But how does it happen that I always choose the non beginner friendly geos? I tried ZA, got told that it is a not good for the beginner, then I tried PH, got told the same. Then tested TH, campaign s**ked, got told it is not good either. And so it goes.... Here is the list I have collected so far of bad tier 3/4 geos: ZA, KE, IN, PH, TH, BR.

Because you are a beginner.


Anyone who has run traffic has already made the mistake of ZA, KE, IN, PH, TH, and maybe sometimes BR. I've done it and I have seen the pattern of those exact geos with tons of newbie affiliates doing the exact retarded mistake so many times it's actually predictable.


ZA/KE/IN/PH speak English. Somehow this translates to 'I can run in all of those with the same lander' without thinking about traffic costs or volume.


ZA is fucking expensive 3G traffic, which is where all the high payout offers come from (mostly Vodacom). KE is roughly the same (with Safaricom) but on a slightly less expensive scale. I'l throw SG in here as well, as they speak English and have expensive traffic that will never convert on pops, but you'll probably try anyway.


IN, PH, ID, MY, and TH will crush you with volume. You will never find an offer even close to profitable off-the-bat on pops. TH is possible on push, but even that is rare. Payouts in most of these geos is absolute shit too. Like 0.07 or something. Again, TH has decent payouts now (I've seen $0.85 offers recently) but the traffic price has also gone up substantially. This is true for MY as well. Tons of volume and you will probably get raped, even tho payouts have been going up, so has traffic costs.


BR is unpredictable like the rest of LATAM. Usually shitty payouts but yuuuge volume. However, reiterating what @twinaxe said, the payout on that offer is so low that even if you got 100 leads a day, you get a whopping $20 revenue, and those offers usually have really low caps that you'll end up hitting early on. They can work. But they can also not work.


You keep saying Tier 3/4 but I haven't heard any of the normal EU geos like PL being tested. Or SK, HU, PT, IT, ES, HR, CZ, and probably a few others with ok pin submits/SOI.


Also surprised you haven't tried LT, since you're from there. I know of at least one pin submit offer that has converted well for almost a year there on Clickdealer.


Even TW can convert pretty well on Pops.


Why haven't you tried all the sub-saharan geos on Haka, Traffic Company, ikangoo etc? Those all convert really easily and traffic is pretty cheap in those geos. Plus, direct linking is no problem.


Quote Originally Posted by mantas View Post
I have been trying to find a specific system for launching and finding profitable campaigns for 10 months now. Finally, I have come to a conclusion that such a precise system does not exist.


Somebody might say target only the wifi, while another targets both wifi and 3G.
Somebody might say start with SmartCPM, while another guy says go hard with the CPA Goal broooo!!
Somebody might say stick with the PropellerAds, while others test 5 traffic sources at the same time.
Somebody might say exclude android webview, when I personally tested it and it converts better then other browsers...
Somebody might say go to Clickdealer, they have good offers, but the one I test does not convert.

10 whole months does sound excruciating. Especially without any small wins. I could ping a few others on here with similar stories who have been on STM even longer and still in a similar boat.


Hell, even I'm not that successful. Most of my money is simply float every month hoping for the next random win to cover all my current losses.


But to be honest, you've tested a total of 6 geos and a handful of offers in 10 months doesn't sound like you're at it 100%. I know beginners who are testing 10 offers a week or more. Whether direct linking, split testing in one funnel, doesn't matter. They're testing many, many, many offers and geos to find what works and putting the effort into the initial AM grind that everyone has to go through before something clicks.


Yes. Everyone has different advice. Because not everything works the same for everyone. That's a good thing.


Quote Originally Posted by mantas View Post
It just gets confusing.

Imagine how much more confusing it would be without STM.


But again, where else have you checked? Are you scouring other forums like affiliatefix, afflift, madsociety, warriorforum?


Information to run affiliate offers is out there, and you can spend all day on YT watching how others are doing it. But even with someone else's blueprint, you have to figure out your way of doing it... which I think you're about to...


Quote Originally Posted by mantas View Post
My next plan. Test how the f**ck I want, throw a bunch of sh*t into the wall and see what sticks.

See?


Quote Originally Posted by mantas View Post
I will test as much as possible in no systematic way. The only constants are going to be Pops as a traffic type and Sweepstakes as a vertical(or any offer that can be direct linked), tier 3/4 geos(excluded ZA, KE, IN, PH, TH, BR) and the payout must be under $10.

Sounds good. Break shit. Figure it out. Break it again. Repeat until it starts to make sense.


Quote Originally Posted by mantas View Post
Sorry for my language, I got emotional

Fuck that. Have you seen how I write?


05-15-2021 08:29 PM #34 iwanttofly (Veteran Member)

Quote Originally Posted by jaybot View Post
Apparently, this reply is too big for the forum's memory and I broke the STM database last night. So I'm trying again:
Feels like a Forrest Gump at the Reflecting Pool moment. I'm sure it was very deep and inspirational.


05-15-2021 10:03 PM #35 twinaxe (Senior Moderator)

Quote Originally Posted by iwanttofly View Post
I'm sure it was very deep and inspirational.
It was from @jaybot so I´m sure it wasn´t


05-16-2021 08:58 AM #36 mantas (Member)

I might sound stupid for this thinking but please correct me if I am developing a wrong mindset.

I am not feeling like a newbie at all. I think, I get the theory quite well. I even come up with my own ideas and strategies. I just have a wrong mindset for this game. Focusing on not important things.

Psychoanalyzed myself lol and realized that I am too much persistent, following everything to the t, trying to be a perfectionist, trying to become the first affiliate media buyer who never has a single losing campaign. Sticking too much to one network, one offer and trying to make it work instead of letting go and testing another one. Just like Tonny Robbins says, beating my head against the wall. Advice like focus and master one thing is helpful for some people but it does not work for my personality type. I need to find balance by focusing on the opposite side. Jumping around would actually be a better advice for me. Looking at the ground and collecting pennies while there is a pile of cash nearby if I just lift my head up.

I have been trying to imagine what super affiliates are doing on the daily basis regarding the campaigns. I guess, it is one word - testing. They wake up and ask themselves, what am I going to test today? (or maybe what type of beer am I going to drink today )

So I guess that's what I have been missing.

So the goal for the next week: launch 2 campaigns a day(or 14 campaigns a week) no matter what. And I will have to remind myself to stop wasting time on overanalyzing the stats of one campaign doing -70% ROI.

and thanks for you help, all these realizations would have not come to me without you all!


05-16-2021 10:45 AM #37 diplomat (Member)

1. Don't treat it as a game. It's business.
2. As you said, testing, testing, testing.. that's how it works.
3. Try to relax a bit, don't be so stiff.
4. Super affiliates are people like you. Only difference is that they test and test and test a bit more and have great understanding of what might work. Also, they may have access to more unsaturated offers or better payouts because they are just doing so large volume. They have bad days too
5. Did I mention testing?
6. Invest into yourself, especially into knowledge and networking. You have absolutely no idea how important networking is. It's crazy!


05-16-2021 10:14 PM #38 jaybot (Veteran Member)

Quote Originally Posted by mantas View Post
I might sound stupid for this thinking but please correct me if I am developing a wrong mindset.

I am not feeling like a newbie at all. I think, I get the theory quite well.
But you are a newbie .

Otherwise we wouldn't be having this conversation.

Yes, you've beed trudging along for 10 months. But you haven't been doing what many 'newbies' who have been at it for 10 days do at full-throttle.

I still consider myself a newbie. I have no clue what the fuck I'm doing most of the time (<---- some may call this testing).

Regardless of any successes so far, until I can put myself in the same fucking universe as any of the incredible affiliates on here (matuloo, twinaxe, diplomat, vortex, etc.) I am still very much in the newbie camp.

Again, there are many different routes to doing IM/AM I'm sure you'll figure out your own path soon, just don't forget that no matter how you do it, you gotta put in the actual work.


05-17-2021 02:39 AM #39 jeremie (Moderator)

Quote Originally Posted by jaybot View Post
But you are a newbie .
I would me more nuanced here.

@mantas, you should read the following paper from Dreyfus brothers. It is only 4 pages, but it describes the learning process of a skill, and it will help you understand where you are. Il also gives tips on how to progress to the next level.

Dreyfus, Stuart E.; Dreyfus, Hubert L. (1986). Mind over Machine. New York, NY: Free Press

Copy paste the line above in Google, you should see the PDF on wsid.info. I can't pull out the link for some reason.


I would say that you are an Advanced Beginner, struggling to pass to Competent level.

The big difference between both levels is 1) gaining perspective on what you are doing and 2) deciding for a goal/plan/outcome. If focus is not your thing, that's fine, but you may need to reflect on the last months, and put on paper where you want to be next, and start to draw some expectations / planning for the next year(s). Read "Stage 3: Competence", i think that part will "speak" to you.

Below a summary of the 5 levels + an interesting table, but I really recommend anyone to read the full paper.







05-17-2021 07:56 AM #40 jaybot (Veteran Member)

Quote Originally Posted by jeremie View Post
I would me more nuanced here.
Jaybot do nuance?

But that's what you're here for

Seriously though, the variety and depth of answers we're getting in this thread are where STM fucking shines.

Good read. And their chart (including level 6) reinforces the idea that as you progress, no matter how you get there, you should end up at the same competence levels.


05-17-2021 08:25 AM #41 larsometer (Senior Member)

I have been trying to imagine what super affiliates are doing on the daily basis regarding the campaigns.
Like with every master it looks so damn easy what they do. Look at David Gilmour (Pink Floyd) when he plays his solos.

It sounds incredible, tho his fingers hardly seem to move. What you don't see when he plays is all his years of practising, his failings, his anger and his frustrations.

You seem to be angry. That is actually very good, because anger can be such an incredible force if you don't use it against you (meaning beating yourself up).

Most likely you have learned more in the last 10 months than you are aware of. Don't just look on your problems but try taking a broader perspective.

Take a few steps back and look at the whole picture. You probably made a some beginner's mistakes. Now you can learn from them and regard them as valuable lessons (actually you paid for them).

Of course everybody wants to avoid beginner's mistakes. But psychology is a tricky thing. Sometimes you just get what you put your focus on. That is by no means esoteric stuff. That phenomen is actually scientifically proven and is called central fixation problem.

Every formular one driver knows it. Don't look at the obstacle but look where you want to go otherwise you just hit the obstacle. Funny thing is that you can read about it a thousand times... and nevertheless you only begin to understand it after your first few crashes.

To cheer you up a little... there is an excellent classic book which I guess can be brilliantly applied to AM --> Who Moved My Cheese?

It explains why the approach of mice can be much better than what many humans do just because they are on constant search for opportunities. It is a 1 hour fun read.


Home > > Newbie Follow-Alongs