The Most Active and Friendliest
Affiliate Marketing Community Online!

“AdsEmpire”/  Direct Affiliate

From Zero to Profit on Mobile CPA, Follow my Journey

Weekly Update

I can't believe its Friday again, this was a short week for me due to family/holidays, however I think I make some little advance on this journey...

Thanks to @Morticai for his amazing video lessons (if you have not check him out, you NEED to look at his website: ppcmode.com) I was able to understand what I was doing WRONG: RON.

Before this, I was running RON on DNTX, and tried to follow the optimize and "blacklist" process, only to find that I was lost... every change I was making on my ad campaigns was simply making me sink into the shadows...

Lesson Learned: Always limit your variables as much as you can... On Tuesday, I re-started my ad campaign, but this time I did the following:
1. I didn't touch my previous LP design.
2. I setup basically 4 ads, each targeting 2 different categories and allowed traffic from Android and IOS and Wifi and ATT. Pretty much I wanted to test each category.
3. I put my bid on the low avg, for all of them.

From this point, my idea was to test every category, and noticed which one will perform better (again, perform doesn't necessary mean profit). From there, I'll optimized deeper into Platform, Carrier, etc...

Yesterday night I jump when I saw the following:
Screen Shot 2015-09-10 at 7.57.26 PM.png

Again, I have not touched my LP, basically I'm narrowing down my variables to try to find a "category" winner, these are the totals from when I started this new approach:

Screen Shot 2015-09-11 at 9.40.20 AM.png



Now the big question remains: how many clicks will I need to consider a test statistically good???

I've heard people 100 clicks, others 300 clicks, to be honest I don't know... but, sticking to 300, I'll consider one of this tests as good.

The next step on this strategy is to COPY the ad campaign that's winning, and increment the bid according to what DNTX calls "delta". Every morning I receive an email from them "suggesting" to increase my bids a "delta" value to get more traffic. To be honest, I think this value is just a strategy to push you into increase your bids: they want you to spend the most possible.

Well, that's what I have for now, I'll continue the optimization process by narrowing down every variable one step at the time, and hopefully will have a winner...

Cheers!
 
Weekly Update

I can't believe its Friday again, this was a short week for me due to family/holidays, however I think I make some little advance on this journey...

Thanks to @Morticai for his amazing video lessons (if you have not check him out, you NEED to look at his website: ppcmode.com) I was able to understand what I was doing WRONG: RON.

Before this, I was running RON on DNTX, and tried to follow the optimize and "blacklist" process, only to find that I was lost... every change I was making on my ad campaigns was simply making me sink into the shadows...

Lesson Learned: Always limit your variables as much as you can... On Tuesday, I re-started my ad campaign, but this time I did the following:
1. I didn't touch my previous LP design.
2. I setup basically 4 ads, each targeting 2 different categories and allowed traffic from Android and IOS and Wifi and ATT. Pretty much I wanted to test each category.
3. I put my bid on the low avg, for all of them.

From this point, my idea was to test every category, and noticed which one will perform better (again, perform doesn't necessary mean profit). From there, I'll optimized deeper into Platform, Carrier, etc...

Yesterday night I jump when I saw the following:
View attachment 4866
Again, I have not touched my LP, basically I'm narrowing down my variables to try to find a "category" winner, these are the totals from when I started this new approach:

View attachment 4867


Now the big question remains: how many clicks will I need to consider a test statistically good???

I've heard people 100 clicks, others 300 clicks, to be honest I don't know... but, sticking to 300, I'll consider one of this tests as good.

The next step on this strategy is to COPY the ad campaign that's winning, and increment the bid according to what DNTX calls "delta". Every morning I receive an email from them "suggesting" to increase my bids a "delta" value to get more traffic. To be honest, I think this value is just a strategy to push you into increase your bids: they want you to spend the most possible.

Well, that's what I have for now, I'll continue the optimization process by narrowing down every variable one step at the time, and hopefully will have a winner...

Cheers!

Hey Artie great to see you in the green! Since your overall campaign is profitable I wouldn't touch anything, just keep collecting data.

Far as statistically significant data it all depends on your conversion rate.
I use this tool: Sample Size Calculator (Evan's Awesome A/B Tools)

So if you're currently getting a 10% CVR but you need a 12% CVR to be profitable you it would look like this:
Screen Shot 2015-09-12 at 9.54.56 AM.png


So you would need 3.5k clicks just to know if the change was actually significant or not.

Most people that say "100 clicks" so on, don't know shit. The reason for this is, it all depends on what your out of the gate conversion rate it at.
 
Weekly Update

First I want to thank @Morticai again for his comments, I've learn a lot from them!!...

This week have just flown by, and I wish I could have make more significant advance. I've spent a good time learning about statistics, it is really important to understand your tracker information: we are buying data! and we MUST understand every parameter our tracker has to offer.

Every day I spent a good amount of time on my tracker, I went through all the reports, all metrics, filters, and each time I went I learned something new, or at least I got a clearer understanding of it. I think this is really important.

On my last post, I updated my running campaign report, I found one category on the domain-redirect traffic I'm running which could lead to profits... then I switch to that category all my traffic, and I didn't touch my other variables (same LP, same offer). I left the campaign run with the same CPC bidding on the traffic source, and I found the results not on the greens... However, I felt I was on good track... so far I had narrowed one category, and one LP. Then I thought, lets try another LP variant... what I did: I redesign my LP to only show a funny gif animation, then the user had to press a button and then I presented the original LP, I made all of this within the same LP, only using Javascript to present the content I needed. This was also a good experiment for me, I learned a good amount of HTML/CSS/Javascript. Once I had the design, I redirected all my traffic to the new design, so far these are the results:

Screen Shot 2015-09-18 at 12.53.16 PM.png


As you may note, my CTR dropped significantly. I've run this LP design for the last 3 days, and I'm working on a new LP design from scratch, which I'll be splitting-testing against my original LP. My thought is to have at least two designs, and then try different angles within those LPs... each time I'll have at least two options to try (enough for a rookie :)).

Parallel to all this testing, I wanted to try another experiment with another traffic source on my LPs and offer. On my first days, I made a deposit into ExoClick without too much research, only to find out they specialize on adult (which I'm not ready to)... Then I thought about running some popunders there... well, this were the results:

Screen Shot 2015-09-18 at 1.05.07 PM.png


From the 11K impressions on the tracker (these show one LP per row statistics):

Screen Shot 2015-09-18 at 1.07.25 PM.png


Not a single conversion!!

Well, I understand the learning curving here: we must understand the kind of traffic on each source. First I was running on domain-redirect, then I tried popunders. I noticed that using the {site_id} dynamic token from exoclick, I was receiving 99% of the traffic from here:

Screen Shot 2015-09-18 at 1.12.13 PM.png


From this info, it was easy to deduct cpx24.com is reselling traffic to ExoClick.

Lesson learned: Don't attempt to run non-adult traffic on exoclick.

I read somewhere that in order to succeed on AM, you need to understand the relationship between A) traffic source (your audience), and B) Targeted Marketing (Lps, angles, etc). I think this was the perfect example.

Well, I wish I'll have some more good news to share but unfortunately no. This is a honest journey, where I will update things as happening (bad or good), and not only the earnings screenshots that everybody loves...

Also, I remember I read about how to stay motivated while "loosing" money... I must admit when I read it I thought: yea!, I can survive that... well, there are times where you will definitely need extra help to keep motivated: updating this thread, talked to my wife about my advances, talking to some of the few guys who have reached to me has helped!!...

Cheers!!
 
Here again!

This morning I was watching my stats, as everyone claims this game requires time/knowledge and experiment everyday to be mastered... I feel it every day!...

So far this is my initial experiment with redirect traffic on DNTX, and I don't know, but based on others posts I guess is a good time for me to explore new traffic:

Screen Shot 2015-10-02 at 11.08.31 AM.png


My thoughts so far on DNTX:

1. You will get a lot of non-targeted GEO traffic!... I usually get around 10-25% traffic coming from different GEO, so be prepared to have some Country Rules.
2. I've tried at least 20 different LPs versions, and 4 or 5 different Offers through the time I was running there... Every time I tried to "optimize" based on OS, Carrier, etc... the conversions went away...
3. Its quite expensive... to get good amount of traffic (not quality), you should target at least .02 to .03 CPC... that's around $25 avg for 1000 clicks... (not actual clicks, but people seeing your ad). From there you have to make your best efforts on the LP funnel to get to the offer...

Overall, I guess this was good time for me to learn... but I don't think I'll be able to make a profit here... so, I'll be trying new sources, new ideas, etc...

Another point I have been pushing forward, is that I've spent a good amount of time watching competition on pop traffic sources, it seems everyone is either doing CPI or Sweeps... I guess this is the time to ask as a new guy if its a good idea to keep going where everybody goes, or trying something different... not a easy question though...
 
@artie7 thanks for writing your journey, you have encouraged me to star my own in few days.

I am completely new to the AM industry, and I am also from Latin America. I have decided to make this happen because of my family... so far I am in the read/learn fase of the process. I keep reading on Charles Ngo start here blog and I am getting to the point where I what to star taking action.

Your journey has helped me to avoid some mistakes and form learn form yours like

- Be very careful to pick the right traffic source
- Split test 1 variable at a time ( I also loved @Morticai video)
- and a few more things

The best things so far about all this is that you also have inspired my by showing your green results. this is defiantly worth it

You have teach me so far that :

- testing makes the difference
- you will not loose money on the process, you will purchase data
- It is not easy but it is doable

@artie7 I would love to ask you:

- How useful have you find the Mobile Immersion access, is ti really worth it?

See you along the way

Thanks for all!

Ganoaffiliate
 
@artie7 thanks for writing your journey, you have encouraged me to star my own in few days.

I am completely new to the AM industry, and I am also from Latin America. I have decided to make this happen because of my family... so far I am in the read/learn fase of the process. I keep reading on Charles Ngo start here blog and I am getting to the point where I what to star taking action.

Your journey has helped me to avoid some mistakes and form learn form yours like

- Be very careful to pick the right traffic source
- Split test 1 variable at a time ( I also loved @Morticai video)
- and a few more things

The best things so far about all this is that you also have inspired my by showing your green results. this is defiantly worth it

You have teach me so far that :

- testing makes the difference
- you will not loose money on the process, you will purchase data
- It is not easy but it is doable

@artie7 I would love to ask you:

- How useful have you find the Mobile Immersion access, is ti really worth it?

See you along the way

Thanks for all!

Ganoaffiliate


I'm very glad my follow-along comments and lessons had been a good resource for you, thanks for the kind words!... As for your question:

As I stated before, I'm coming from a IT background and had the opportunity to attended different professional courses/certifications paths on the field (RedHat, VMware, HP, etc) hence I'd say the MIP will look like a BETA training path to me, yes there are good resources on the very basics (tracking, campaign setup examples, traffic sources recommendations, etc.) but you can also look for the same information online.

If you want guides to put yourself ready to run campaigns in a short period of time, MIP is a good invest for you. However don't expect to become profitable using the resources there (landers, banners, etc).

On the other hand if you have the time and willingness to read through different sources, and experiment a little bit by yourself, then you can be save without it.

This if just my honest and humble opinion, you are of course free to experiment by yourself...

Cheers!
 
Hey Artie I've been working on a pretty lengthy guide that I think will help you quite a bit. You probly know most of the stuff in Part 1 but it might be worth th e read regardless: How Dominate Pop & Redirect Traffic in 2015 - Part 1

The problem is there are a lot of "coaches" teaching complete garbage "tactics" instead of teaching you how to think for yourself and build a strategy.

How have your campaigns been as of late? Any questions you have that I can shed some light on?
 
MI
Back