Restarting: the buildup follow along

[silverace]

Hi everybody!

Apart from this being somewhat of an introduction, this is going to be a follow along that will be somewhat a-typical in the beginning.

First off: I'm someone who is re-entering the affiliate space after being active 3 years ago. I used to do PPV.
I'm from the Netherlands (near Amsterdam) and affiliatefix seems like the right place to start this thread! Always seemed like a friendly forum.

The goal of this thread is to first determine a gameplan, get everything in place to start my restart(?)(lol) and then actually get started. For this I would love your input and that is the reason of this thread.

My situation:
I'm currently finishing my bachelors degree, I hope to be done at the end of januari and then immediatly start IM'ing. Before that I have to decide on a server, tracker, traffic source, cpa network, and a vertical.

What I already decided on:
VPS: digitalocean has always been great for me and is really cheap.
Tracker: I think prosper202, contingent on your input. I want to spend most of my money on traffic and if it does an OKish job filtering ISP's etc I'll go for it before switching to Voluum when I get the cash.
Traffic source: I think mobile pops. Still undecided on the specific traffic source to begin with. I think mobicow or zeropark?
Aff network: I'm thinking about applying @ furthermobi since their offers look great, not sure if I can get in though.

I'm documenting my idea's on all the research that I'm doing here (ideas on offer selecting, optimizing, and so on), so I hope that apart from getting valuable feedback on this thread, I might drop some good summaries of the information that is available on the internet right now!

Talk to you guys soon,

 

[T J Tutor]

Hi everybody!

Hello @silverace and welcome to the AffiliateFix!

My situation:
I'm currently finishing my bachelors degree, I hope to be done at the end of januari and then immediatly start IM'ing. Before that I have to decide on a server, tracker, traffic source, cpa network, and a vertical.

This is correct. Stick with the tracker, vertical, and traffic source you choose until you master them. A couple of months at most, maybe a month for you. This way you will know everything required to add more as well as move laterally across the platforms with ease. Having two, or three, networks may be advisable.

PS: digitalocean has always been great for me and is really cheap.

Always popular.

Tracker: I think prosper202, contingent on your input. I want to spend most of my money on traffic and if it does an OKish job filtering ISP's etc I'll go for it before switching to Voluum when I get the cash.

I think most of us started out with P202, it works, but upgrade to a more suitable tool when you are able.

Traffic source: I think mobile pops. Still undecided on the specific traffic source to begin with. I think mobicow or zeropark?

Both very popular here. You can find plenty of members talking about them in the mobile threads. See if there are reviews in our Resources area.

Aff network: I'm thinking about applying @ furthermobi since their offers look great, not sure if I can get in though.

If you apply and then post in their Resources discussion thread, you may receive additional considerations.

I'm documenting my idea's on all the research that I'm doing here (ideas on offer selecting, optimizing, and so on), so I hope that apart from getting valuable feedback on this thread, I might drop some good summaries of the information that is available on the internet right now!

Talk to you guys soon,

Build yyour plan, work your plan, and you'll soon be where you need to be!

 

[silverace]

Thanks for the welcome @tjturtor!

Looking at Odigger I saw some great offers at FurtherMobi. That resource thread doesn't look that great though, even being from europe (although their affpaying page looks fine).

I still do have some older aff accounts at Peerfly, Adsimilis, Grabads and Neverblue (Neverblue still a thing?) which I might use to get started with.

 

[silverace]

Already did a quick run to test whether or not my server etc are setup ok. Already got a conversion from mobile pops! nice!

CTR is kinda horrible though, and the click discrepancy even worse. My network shows $5.48 spent for something like 1300 pops, making my CTR atrocious. I do think it has to be something on my side, since this isn't the kind of click loss a network would cause right? Even LeadImpact didn't use to be this bad

 

[T J Tutor]

Well, networks do scrub. However, I have read on this forum as well as others reports of click loss with Furthermobi.

Another really important factor can be your server. It is very possible that load times are not fast enough. It's possible that some are getting a server timeout while trying to get to your landing page.

Another test would be to split test the same offer from two different networks.

You should also be split testing your creatives and your landers.

 

[silverace]

Thanks for that! I'm not yet with furthermobi btw, the clickloss is from the adnetwork to my LP's. Loading times could be the case, I'll see what happens if I splittest this LP with a lighter lander. I'll also run the pages through pingdom and see what happens


Edit:

Seems ok, I might test some intro/exit pops with my lander to see if people rapidly click the screen away. Besides that I'll just let it run for a little longer. If it keeps this way I can always upgrade my digitalocean server and see if that improves anything!

 

[silverace]

Contacted support of the traffic source on the large click loss issue, let's see what happens. In the mean time I filtered out some websites and devices that weren't clicking my LP. Got a couple more conversions.

The real data though, when factoring in the click loss is pretty bad:

Stoping the campaign when the old funds I had are depleted and pausing everything till I find the reason for the click loss. Maybe it's the traffic source , maybe my server (already upgraded to more RAM which didn't make a difference so I don't think so). Plan is to move to Mobicow anyways, but I first want to have everything working: a solid gameplan and a reason for the clickloss.

As for the solid gameplan, I already started.

Create/find/rip 4 landingpage angles: done. I have 4 landers in place ready to test. All the landers have a different style and approach to convince the user.

Gameplan:
- Run $5 to each lander and analyze the data. It is to early to start cutting devices, ISP's etc (unless 1/2 sources hog the traffic without getting clicks or conversions), but I would get an idea of what might work.

- Pick 3 landers to continue with, and run $5 to each lander again. The worst performing lander will be eliminated from step 1, the rest will have $10 a pop in data. Not only should I get somewhat of an idea what lander works at that point, I also get overall data by now having spent $35 testing a campaign.

- Pick the best lander (or maybe best 2 landers in case they are close), start cutting individual targets which show signals in the data. Carrier heavily underperforming? Your out. Device that just doesn't want to convert (not even click my landers)? Adios.

- split test offers

- profit

Does that sound somewhat OK for a gameplan? Should I spend more on landers before making decisions? All the feedback is welcome

 

[T J Tutor]

Maybe it's the traffic source , maybe my server (already upgraded to more RAM which didn't make a difference so I don't think so). Plan is to move to Mobicow anyways, but I first want to have everything working: a solid gameplan and a reason for the clickloss.

Yeah, split testing an offer between two networks is a good idea when possible.

As for the solid gameplan, I already started.

Create/find/rip 4 landingpage angles: done. I have 4 landers in place ready to test. All the landers have a different style and approach to convince the user.

Gameplan:
- Run $5 to each lander and analyze the data. It is to early to start cutting devices, ISP's etc (unless 1/2 sources hog the traffic without getting clicks or conversions), but I would get an idea of what might work.

- Pick 3 landers to continue with, and run $5 to each lander again. The worst performing lander will be eliminated from step 1, the rest will have $10 a pop in data. Not only should I get somewhat of an idea what lander works at that point, I also get overall data by now having spent $35 testing a campaign.

- Pick the best lander (or maybe best 2 landers in case they are close), start cutting individual targets which show signals in the data. Carrier heavily underperforming? Your out. Device that just doesn't want to convert (not even click my landers)? Adios.

- split test offers

- profit

Does that sound somewhat OK for a gameplan? Should I spend more on landers before making decisions? All the feedback is welcome

You are using good structure here. Put this in motion.

 

[silverace]

I've been talking with my traffic source (great support by popads.net!) and they think it probably is my server. Going to upgrade it and test more traffic there. They gave me the advise to run some traffic through bit.ly to my tracking link so I can see where the error is. Eager to find out more.

In case it is my server, what kind of digitalocean plan do you guys think I need & what kind of clickloss is to be expected doing mobile pops? My current server is the $20 plan(just upgraded, might upgrade tothe $40 plan to test the clickloss there).

 

[T J Tutor]

I wish I could tell you about Digital Ocean. I can only say I've heard good things about them. Hang tight, someone here must know more about DO.

 

[silverace]

Guess it's going to be splittesting server configurations before I can run campaigns. ran a couple more clicks and splittested with bit.ly so I could get another clicktracker's results.

Popads: 363 views
bit.ly: 268 views
My server: 86 views.

Deff the fault of my own server. Going to upgrade once again and see what happens.

Edit: some stresstest results with 8000 requests in 20 seconds.

$20 digitalocean plan

$40 digitalocean plan

Doesn't seem to make a big difference. Maybe I'm going to completely reinstall my server as there are 33 apache modules enabled as well, which is pretty bad. First I'm running a little more traffic to this one to see what happens.

Edit2;

Even the $80 plan fails. solution was simple. I installed apache on my server instead of Nginx. Nginx is event driven and way better for tracking. Work to do!
Edit3:

This follow along is becoming way too technical

 

[T J Tutor]

This follow along is becoming way too technical

You are learning, no such thing as too technical really. There will always be some of this in your business. You simply have taken the bull by the horns and are getting past a technical difficulty. You are doing well, you aren't giving up. That's the number one most important element in your posture and demeanor. This is why you are going to master all of the elements in your business, because you never give up!

 

[silverace]

Thanks for that T J!

Fixed everything (I think). Server is now running on Nginx. Results of Nginx on the $5 digitalocean plan (spoiler: better than the $80 plan running apache)

Time to get my LP's up and running again and submitting a campaign. Hopefully I can do it before tonight so I have some data when I wake up tomorrow

 

[Neo0707]

What traffic source did you use? Did you do landing page on PHP or Html?Thanks

 

[silverace]

@Neo0707 I was using popads. Also going to do so for my next test, with this server setup my click loss should be a normal %! Let's see what happens

Landers are HTML with some PHP elements (also JS & Jquery btw).

 

[Neo0707]

Let's see GOOD LUCK.

 

[silverace]

Still dealing with clickloss which really shouldn't happen by now. It's deff not my server which should be able to handle the pressure. Not sure exactly when the clicks are counted but could entry pops have something to do with it? It's the first JS that is loaded and the page keeps loading till it's clicked? Time for a splittest

Also reduced the size of my Javascript files: Javascript Compressor - compress code online for free and my images: Kraken.io Image Optimizer which are awesome and helped my page go from 144 KB to 90.5KB!

Edit:

Quick question, would you guys run 2x$5 plans (1 exclusive for tracking and 1 for landers ) or 1 $10/$20 plan to do both on 1 server? The first option might be something I could try and see if it changes anything?

 

[silverace]

Yet another testrun in which I will splittest my tracking.

1x $20 digitalocean plan to host my landing pages
1x $10 digitalocean plan for prosper202
1x Adsbridge free plan to compare results with
1x bit.ly -> prosper202 link to see how much clicks are sent by the traffic source.

Page loading speeds are in favor of my prosper202 setup (1.1 sec loading time through prosper tracking link vs 1.8 sec through adsbridge), but let's see what happens!

 

[Prince.AdSenceMedia]

@silverace you can try with AdSenceMedia.com. Scrub free, real time detailed tracking report, source IDs & sub IDs upto 5. all top offers in all the verticals with very good payout range with great CR% - EPC . Adult,Dating, Mobile,iOS,Android, email-zip submit all the offers are waiting for top good publishers All of you are warmly welcomed.

 

[silverace]

@Prince.AdSenceMedia I think I actually already have an account, buy you guys used to only do incentive offers right?

Well, adding Adsbridge didn't fix anything. It did 10% better than prosper but that's by far not good enough. Bit.ly got 100% of the traffic, while prosper got a little over 20% and adsbridge a couple % extra. I do really like the @AdsBridge interface btw!

It's getting tiring not to get the traffic I'm paying for, impossible to get a campaign profitable untill I fix this issue. If anyone has any idea how to improve on this I'm happy to hear any suggestions

The thing that I did succeed in: Open a Mobicow account and get $50 free bonus on my deposit. Free campaign test for when my servers work again