Home > > Newbie Follow-Alongs

---

40 Day Challenge before quitting my job (20)

---

Hey everyone, this is my 40-day challenge follow-along, where I will learn through thick and thin the ins and outs of basic AM skills.

And it just so happens that if I do this 40 day challenge, I will meet the deadline I have set for myself to quit my current job and apply the skills I've gained full time on AM (You can read my bold intro here https://stmforum.com/forum/showthrea...-hey-everyone! )

I'll be using this guide as my framework: https://stmforum.com/forum/showthrea...ntro-and-Index

Subscribe to this thread, as I've been known to ask perplexing questions

---

Day 1

Following Amy's advice, I've applied myself to the 3 networks: Mobidea, ClickDealer and Gotzha. I thought what the hell, pointed my finger randomly at the aff network registry and picked out Mundo.

Mundo was a little different, as they required me to point to a website I own, so I dusted off my old personal static website generator, picked a good looking theme and just added huge landing page with text "Independent media buyer services and consultation. Website is currently under maintenance." Oh, did I ever tell you like to bullshit around and cut corners?

Afterwards, applied myself and got auto-approved by PropellerAds and PopAds as suggested by the guide.

Read a ton of stuff on the forum for onboarding beginners, most likely will forget at the end of the week, my style is to apply the knowledge and then remember it.

End of Day 1.

---

Day 2 - 8

Now here's where stuff gets interesting and my bank balance drops from all the spending I am doing.

While I was waiting for the aff networks to get back to me, I picked the gear I will be running:

Redtrack.io - Cheap-ass tracker. I found a coupon for 50% off, but then they told me its only monthly. Nonetheless, got an email from them offering 40% off annual plan, which will run me $576 for 3mil events per month. This is like barely 3 months needed to cover Voluum tracker, however, I am fully aware of the trade-offs between cheap and "innovative" vs expensive, but more experienced. If I get screwed over by this tracker more than a few times, I am definitely gonna look for another one.
Adplexity mobile - at this point I guess its a no-brainer to pick this for spying software. Might take some time later and investigate other options.

Bonus! AW Conference ticket in Barcelona - What can I say, I wanna meet with my peers and see what they're up to!

The Networks

Mobidea auto-approved - no fun, you didn't say hi to me :/ Picked 5 offers that got auto-accepted immediately.

ClickDealer - got pinged my AM via Skype after a day. We quickly discussed what I wanna do and then I got accepted. Didn't ask for any "converting" offers yet, just picked a random one. (as suggested by the guide)

Gotzha - nobody hasn't replied in 5 days (or 3 biz days), so I searched for a contact person for Gotzha, got his skype credentials and politely pinged him.

Mundo - quick 'n easy talk and got accepted. This is where I actually asked for some good offers. What I got was an excel sheet (I asked for sweeps in t3/4 geos) with a bunch of campaign names + landing pages (??? - see my follow-up questions) There was like 20-30 offers, all of them could then be found via their platform.

To:do

Setup Redtrack.
Learn how to navigate and use Adplexity.
Get back to Mundo AM about some of the offers I want to do.

Questions:

So I am a bit lost about the landing pages and their purpose. When Mundo AM gave me the list, each offer contained a landing page - why? I thought I was supposed to make my "own" (obviously, I would rip it from adplexia). This list was for sweeps offers. Is it more like an example landing page running the offer or I can literally just copy it? (and why would I do that?)

---

Hi daliu!

Thanks for starting this follow-along!

Great self-intro! And congratulations on following your heart. It takes courage and determination. You will doubt yourself now and then. But if you never try, you'll always wonder "what if...?"

As long as you don't give up, you'll get somewhere eventually! To me, failure doesn't exist - it simply means someone has given up before reaching success.

It's just a matter of time really.

Good stuff! That's the intention the tutorial was written based on.

After getting some practical experience, you can always reread the tutorial plus other valuable threads - by then you'll have an existing base to assimilate new knowledge into.


Wowsers!!

Please be sure to drop by the STM booth to say hi!


When networks say "landing pages" they're usually referring to "offer pages".

To them, what we put in front of the offer page is a "pre-lander".

Just semantics.

Some offers actually INCLUDE pre-landers built-in - for sweeps, for example, some offers already ask survey questions. For those, you obviously don't want to put a survey lander in front - visitors would get hella pissed at answering 2 sets of survey questions!

For those ones, try direct-linking (without landers). Or try REALLY short, one-page landers - such as a page that asks the visitor to pick their favorite color for the prize (iphone or whatever).

You're progressing nicely - looking forward to your next update!



Amy

---

Day 9-27

Well shit, I had (and still having) one of the worst viruses I've caught in my life, since I last posted. Had to take a break from taking action for the past month or so. The mentioned deadline I've created for myself before quitting my job is very near, so I already gave my resignation notice (gotta do here 20 calendar days in advance). Even while writing this post, I am having a massive headache and my right eye feels like it could pop out any moment - apparently, sinuses are the culprit, at least that's what the good doctor says.

Anyway, enough 'Oh woe is me!', here's a summary of what I've done according to the 40-day tutorial index. Questions are marked in bold

• Setup hosting via AWS S3/Cloudfront
• Picked an offer from AM for sweeps pin-submit in greece (web+mob)
• Ripped a lander I found in greek language from AdPlexity (wheelspin, no idea if language is authentically correct)
• Created a campaign on Propeller ads with 15$ budget via PopUnder / SmartCPM
• Setup everything via RedTrack.io tracker

..and the results?

PropellerAds



Tracker
RedTrack.PNG  Views: 37  Size: 54.3 KB  ID: 20933" class="thumbnail" style="float:CONFIG" />

ClickDealer offer


utter crap. I think I let the traffic run for around 2 hours, before it accumulated 10k impressions and I stopped it. I didn't perform any optimizations, because I didn't get any conversions. But these 13$ taught me enough to prepare better next time (and it's better than going to the movies, the suspense was killing me if I could get at least 1 conversion)

So my here is my short retrospective, following up with questions in bold:

After checking adplexity it shows that my 10k impressions barely got picked up as a sum of 5 hits! If that's correct, this definitely gives me a better perspective and correlation about the numbers on adplexity. I think there are several factors that contributed to me getting 0 conversions (please add more reasons aswell)
1) After being out of AM for more than a month, I rushed as I wanted to get a campaign going ASAP - this means cheating a little bit and skipping day 9-27 :P
2) Since this was a single traffic campaign, only later did I read up that I should have not grouped these together - wifi/3g , desktop/mobile (as this was web/mob offer)
3) I think probably biggest offender was that I did SmartCPM for Greece and recommended price was around 2$ , whereas I bid for exactly 1$. The Scrooge in me probably fucked me over and I got shit traffic in return. I did not use 'Advanced View' so I did not analyze the forecasts properly.
4) Another reason I think this sucked a lot was that my lander had spinning wheel and after examining the offer lander, its flow was too complex - submit phone nr, enter pin AND then answer 10 questions!!! I am guessing this is too much time for a person to work with and this should of probably been done as direct-linking (obviously I am total noob here and just speaking non-sense, but I'd love to hear anyone's thoughts on this matter.

NOW, for the questions:

• Why did on adplexity my landing pages showed in stats that I was using PopCash? I only used PropellerAds - does that mean PropellerAds sent me traffic via PopCash?
  
• There's a 2k impression difference between my tracker and PropellerAds - how did this traffic source manage to generate 2k more and my tracker didn't log it? Is this normal?
  
• I calculated that there was only 3% rate from lander to offer. Is that super shit or what?
  
• Is it normal to run an offer that is web/mob?
  
• Why do I need to setup conversion tracking both in affiliate and traffic network? Am I correct to understand this is only needed in per CPA payment in traffic networks?
  
• Is there any point optimizing a campaign if its got 0 conversions?
  
• Why are clicks not tracked in traffic source? In PropellerAds case, are clicks == CPC?
  

Bonus questions about shadiness - so I actually was ready to run an offer from ClickDealer on sweeps for iPhone XS - but just before I started to send traffic, I noticed that it changed its offer lander to samsung s9 - and even after more than 24 hours its still like that - the URL didn't even change with its params. The advertiser literally swapped offer landers under the same url. How do you deal with this if you're actually running a live campaign and this happens?

---

1. I wouldn't count on Adplexity to display the stats of your campaigns correctly. As a matter of fact it would be nice if no stats were ever displayed there from my campaigns
They use crawlers for their data, so it will not be exact, not even close.

2. It's normal. Search click-loss on the forums

3. In my experience 3% is actually pretty good for POP traffic.

4. Yes, Web/Mob Sweepstake offers on Clickdealer work amazingly

5. You don't need to setup conversion tracking on traffic source. Many affiliates will argue that you shouldn't, in order to prevent traffic source from seeing what's working for you and stealing your campaigns. I wouldn't worry about it at this stage, especially with Propeller. If you ever want to run a SmartCPA campaign (wink wink) you HAVE to setup conversion tracking with them.

6. There is nothing to optimize. You need more data. I haven't read your entire thread, just the last post, but how many landing pages are you running? Just one? That you don't even know if it works? Are you sure the code is clean? What is the CVR of the offer for the past 7 days? Has your AM told you? Does it convert for others for sure?

7. Traffic source only track impressions and conversions (if setup), it has no way of knowing how many users clicked through on your lander to the offer.
BONUS: Have you checked the lander using VPN in the GEO that you are advertising? If, advertiser did in fact swap the offer landing page, just change your lander to Samsung or even leave it as is. The effect on conversion rate will be negligible on pop traffic.

---

Thanks for replying, affiliatecase,

I ran a single lander and it had 0 conversions after 8-10k impressions. (a total of 304 clicks from my lander to offer) I have checked the code - it's clean. There is no public CVR info on Clickdealer offers. My AM literally just dropped me 8 offers with no more info (I noticed the trend of all AM's doing that to me) after asking for pin-submit sweeps in t3/4 geos.

Do you think pin-submit conversions are harder for a newbie than leadgen (like email submit) ? Could you re-read and give your feedback about my retrospective points - am I forming my observations correctly?

---

Day 27-40

Applying all the knowledge I've gathered from the follow-along I got to launch 7 more campaigns. I am thinking of enhancing the thread topic and calling this "the hunt for the green" as I am trying to find my first performing lander+offer.

The two posts below will be about running sweeps offers for an iphone (3 campaigns) and a gift voucher (4 campaigns).

I have only skipped the part in the 40day guide about bot testing and I will get to it very soon (in the next campaign).

---

POP Sweeps Offer for an iPhone in Greece.

Used only 1 lander and 1 offer (pin-submit)

So being the hothead I am and just wanting to experience the pain immediately, I straight up just re-used the same lander from the previous post I talked about and then split up its traffic into these three:

PropellerAds MOB 3G/LTE
PropellerAds MOB Wifi
PopCash MOB 3G/LTE



Because I already have the insight from the next campaign, the results here are horrendous - BUT! I finally got my first conversion, which totally made me happy and figure out that everything was working, at least from technical perspective.

Total spendings:

PropellerAds MOB 3G/LTE - spent $19.66 / Revenue: $8.99 / ROI: -54.27%
PropellerAds MOB Wifi - spent $21.09 / Revenue: nada / ROI: -100.00%
PopCash MOB 3G/LTE - spent $10.00 / Revenue: nada / ROI: -100.00%

Total cash spent on traffic: $50.75 / total Revenue: $8.99 / ROI: -82.29%

Decision: STOP running ALL campaigns

Maybe I should have continued longer with the 3G/LTE traffic from PropellerAds. But since I don't know the gist of when to stop and start, I just got happy this converted once and thought testing with just one lander and one offer is insane and would burn more money.

If you look at the picture, its actually insane, how many clicks from lander to offer went through via popcash - and yet no conversions. Is this because of getting spooked of being asked to submit a pin (I warmed up to this in my lander)? The 10 bucks from popcash almost brought in 2.5 times more than propellerads combined. Am I getting fucked here with shit traffic?

---

POP Sweeps for a supermarket voucher in Germany

This time I got more aggressive with several of my AM's and asked for best converting sweeps offers. Most of them came back with US / UK offers. I then asked for t3/t4 offers, which later the networks gave me the notice they are being turned off.

So I decided I will venture into Germany and try to run a voucher offer. But I went with a hunch and first explored the offer I got from AM via AdPlexity and noticed a couple of its voucher variants floating in australian market. So I ripped and cleaned in total 6 landers and ended up using 4 of them. Except for 1 lander I ripped, I chose to translate via google-translate into german :/

So my campaigns were set up as is:

4 Landers for 1 Offer

PropellerAds Mobile Wifi
PropellerAds Mobile 3G/LTE
PropellerAds Desktop
PopAds Mobile Wifi

The results:



PropellerAds Mobile Wifi - spent $155.53 / Revenue: $39.39 (13 conversions) / ROI: -74.67%
PropellerAds Mobile 3G/LTE - spent $19.81 / Revenue: $9.09 (3 conversions) / ROI: -54.11%
PropellerAds Desktop - spent $15.62 / Revenue: nada / ROI: -100%
PopAds Mobile Wifi - spent $6.90 / Revenue: nada / ROI: -100%

Total cash spent on traffic: $197.86 / total Revenue: $48.18 / ROI: -75.65%

These are the results I have gathered in the 10 hours since I've launched all of the campaigns.

The Desktop campaign - it had 8k impressions and only 11 offer-to-lander jumps. I stopped it.

For the 3G campaign - the only "optimization I've done" is that I've looked at the carriers and one of them had 2k impressions with only 3 clicks to an offer, so I excluded it from traffic.

So let's focus on the most expensive campaign mobileWIFI:

Please let me know if what I am doing is correct:

1) Early in the campaign, I blacklisted two zones with 11k impressions / 12 clicks & 6k impressions 2 clicks
2)
After 4th conversion, I noticed this

Two of my landers got both 8k impressions and around 25~ offer-to-lander jumps, but no conversions. I decided to turn them off and focus on the remaining two (which at the time, both had 2 conversions each) (this was around 4 hours since the campaigns got launched).
3) I proceeded blacklisting two more zones - 3.5k impressions / 3 clicks & 1.2k impressions / 0 clicks (am I over-optimizing here?)
4) I did not alter any targeting inclusions/exclusions for the propellerAds or popAds traffic.
5) My bidding was very close to the "recommended" bid and looking at the propellerAds graph, I was definitely jumping over at least 3 slopes (referring to the bidding tutorial in 40day guide).
6) Not sure how to handle popAds bidding yet, I also bid very close the recommended suggestion. (no conversions yet)

I topped up another $100 for propellerAds and set enough to hopefully last until I wake up in the morning.
No one's probably going to share wisdom before I make my new decisions, but here's what I am thinking:

1) I noticed a lot of traffic and conversions (a.k.a. 2 conversions for one OS phones, rest of the conversions came from a particular OS) - I will only target that specific OS devices.
2) Both landers I kept running are around the same amount of conversions - I will continue running them both. (is it normal for two landers to perform equally?)

But none of this matters if I don't ask these two questions

1) Reaching around $200 on traffic spending and still continuing to run it with another $100, with revenue around $45~ - am I going in the right direction? If I should have quit, what signs have I missed? I am thinking, that to spend couple of hundred in Germany and gather the data to learn about the particular vertical + offer seems legit, but I might be thinking incorrectly (?)
2) Should I have cut off the two landers out of the 4 so early (with 8k impressions each) in the mobileWifi campaign?
BONUS: Looking at my tracked impressions, I have 10k click loss for the mobileWifi campaign. Is this still normal or should I be concerned and start figuring out what's wrong? I am serving this via AWS S3+CloudFlare and did lander load-speed tests - they are fine.

---

So sorry for the extremely late reply @daliu5 !! Been busy getting out the "What's Working in Affiliate Marketing" report (which is finally done).

Playing catch-up on the forums now!

Will reply to each point in detail below...

It may not be a good idea to hold yourself to that deadline.

That "virus" may actually not be what made you as sick as you were. It could have STARTED the whole ordeal, but I'm suspecting that it's because you're forcing yourself to quit you job before you're ready, that is causing massive stress which is what had led to such violent reaction/rebelling from your body.

Just go with the flow - embrace your job. It's providing you with the stability you're needing. When you're ready, you'll know. There's no need to be giving yourself all this stress. Don't back yourself into a corner to try and make affiliate marketing work out in a short time - doing so can be very detrimental to your mental and physical well-being, not to mention your bank account!


First of all, it's important to test at least a few landers. By testing only one lander, you had NO IDEA whether it was decent enough to get conversions or not.

i.e. When you didn't get conversions, you wouldn't know whether it was because the offer was crap, or the lander.

Whereas, if you had picked 5 (or more) landers that were popular on Adplexity, and optimized them for functionality, display, and speed, then you could be confident that at least one or more of the landers is/are decent. So then if you run 10x payout's worth of quality traffic and don't get one conversion, chances are the offer is a dud.

Secondly: Don't be disappointed if one offer doesn't work out. Pop is ALL about testing MASSIVELY for as cheaply as possible. If an offer doesn't start converting almost right away, ditch it.

It's more expensive when you're breaking into a geo for the first time, because you don't have a proven lander, AND you don't know which placements are bad/good. Once you've run a campaign in a geo and have cut some bad placements and uncovered some good ones, next time you run there you can test for cheaper - either by cutting the same bad placements right off the bat, or by targeting just the good placements. And if you're running in the same vertical as well, then the winning lander from the last campaign can be reused as well.

In other words, in the beginning, you just gotta suck it up and spend that cash.

And don't despair: Pop isn't the only traffic type. In fact, I would suggest that you DON'T stick with pop for too long. You can get better results investing the time and money elsewhere. But pop IS the cheapest and easiest to learn! So look at all your ad spend now as tuition, run some pop camps, then "graduate" to push and/or native, or even start tinkering with FB/Adwords.

Even now, if you see a promising offer+lander from pop camps, you can test them on push traffic, and just test ads (text+image) there - chances are you'd get better conversion rates there.

But I digress. To answer your questions from that campaign:


It's possible! I'm not sure how exactly Adplexity accounts for the traffic.


This is referred to as clickloss. You'll find that you'll almost always see fewer impressions in the tracker than on the traffic source, and fewer clicks in aff network stats than clicks registered in the tracker.

2k "missing" impressions in 10k is around 20%. That's not excessive, but you can potentially improve on it, in 2 ways:

1)Get a faster tracker, or make your tracker faster. For example, if you're using a self-hosted tracker, make sure it has enough server resources to effect fast redirection speeds. Or, switch to a cloud-based tracker to compare the difference.

2)Use a faster DNS service for your tracker domain - the difference in speed would probably be minimal though. I haven't done a comparison there to see if any impact on conversion rates.


See this is yet another reason to test more landers!

When you do test more landers, you'll get a much better idea on what the "average" CTR is - if there's even such a thing.

Different geos and verticals, and different landers, will see different CTRs.

For example: In tier 1 countries, CTRs tend to be lower than tier 3/4 geos. The difference can be night and day - like <5% CTR vs. 30-50%+ CTR.

I don't remember what kinds of results I was getting in Greece - you'd need to test more landers to find out.


Of course!

And there's nothing wrong with targeting both if you just want a quick test to see if there's any potential. But when you see potential (e.g. that the offer is converting and campaign can be optimized further), you may want to separate them into 2 camps. It would just be easier when drilling down to the various stats variables and making campaign decisions. Otherwise you'd need to drill down to web/mob as the 1st level and THEN drill down into stuff like OS etc. etc. You can do it that way though if you wouldn't mind the inconvenience.


You DO need to set up conversion postback at the aff network, so that conversions will get posted back to your tracker.

You don't need to set up conversion postback for the traffic source actually. That's optional.

I tend to be paranoid about traffic sources stealing campaigns, so I rarely post conversions to the traffic source.

But there are benefits to doing that also. You'd be able to see conversion data along with ACTUAL costs per placement in traffic source stats, which can save you the task of having to export cost data from the traffic source, export revenue/conversion data from the tracker, and combining them on a spreadsheet to see accurate revenue+cost data per placement. I usually don't try to go for accuracy though - I'd cut placements based on cost data in the tracker, which is rarely accurate no matter if it was sent back from the traffic source token or manually-updated (because for smart CPM, the cost of traffic will vary from placement to placement, and the tracker wouldn't know this when you do a manual cost update). I would just do a rough estimate in my head - for example if the actual cost of traffic is around 30% more than the cost I see in the tracker, I would mentally add that to the cost when I'm looking at placement stats and deciding which ones to cut.

Efficiency trumps accuracy in most cases when it comes to pop, which is mainly about testing massively to find promising offers, then optimizing and scaling quickly. This is because pop camps tend to be short-lived. Spend too long on optimizing and the campaign will die before you recap your losses.


No. Vast majority of the time, optimization decisions are driven by conversions.

The only time when we make decisions based on CTR, is if the CTR is horribly low, such that it would never become profitable. This applies to landers and placements. For placements, you can also cut the ones that have CTRs that are considerably lower or higher than most of the other placements - those typically don't convert well.

Like I've mentioned though, testing a few landers would be good. And if you can find similar offers to test as well, all the better!


This is pop traffic. The traffic source is only responsible for sending visitors to your tracker campaign link. After that, the tracker takes over.

Therefore, there would be no way for the traffic source to know whether the visitor has clicked through your lander, or not.

Only the tracker would know this.

Traffic sources would only charge by clicks if you're running banner ads, where the visitor can either click on your banner, or not. And because the web page where your banner ads is displayed contains the traffic source's tracking code, it will know when someone clicks on a banner.


That's just wrong. If you notice it happening, inform your AM.

But one question needs to be asked: Are you sure you were looking at the right offer? Were you using a VPN set to an/the accepted geo? Did the offer accept wifi traffic (if not, then using the VPN won't work)?

Reason for the question: You may have been redirected by the aff network to other offers.

One way to find out if this is the case (next time you encounter something like this), would be to check your aff networks stats, to see which offer(s) is/are getting traffic. If you're seeing one or more offers in additional to your intended offer, chances are the aff network has redirected you to other offers because you were browsing from outside the target geo without a VPN.

Will reply to your other posts below...



Amy

---

Would you like me to change the title? If so please let me know.


Aiya...(SMH)

Next time PLEASE don't skimp on lander testing!

You even said yourself that you weren't sure if the text was properly translated. What if it wasn't? Then you'd be testing offer after offer - and if the lander is shitty, you'd be pissing in the wind.


You should have kept running that 3g campaign! ROI of -50% without major optimizations is promising!

AND - this is without having tested landers!

BTW - congratulations on that first conversion!! It's a major milestone in terms of expectations and perspective. You'll go on to make many, many more - but this will remain the most important conversion ever.

As for traffic prices: For sure there's a difference in quality between networks. But keep in mind two things:

1)This is why running a bot test can save you money.

2)Traffic cost is driven by the ROI the other bidders are getting. Nobody will continue to pay for traffic that doesn't convert, and nobody will bid high for low-quality traffic. Which means in the end, more expensive traffic means it converts better, which is why bidders are bidding higher to drive up the prices.

This also means that there's a way to make both cheap and expensive traffic work - theoretically. Just run traffic and let the data speak.


Yeah there seems to be a lack of tier 3/4 cpa offers. Trying tier 1 geos wouldn't be a bad idea. Just keep in mind that you'd need to 1)test lots of offers and landers to find something promising, and 2)cut placements aggressively, and expect to have to do a lot of cutting before seeing green. It takes money to conquer a tier 1 geo - you just need to suck it up while you bleed money on cutting placements. Doing a bot test first to weed out the worst placements helps. Looking at CTRs of all placements, figuring out the average range, and cutting placements with CTRs that fall outside that range can help as well.


One suggestion: If you're testing multiple offers and/or landers, perform this testing on the most reliable traffic source (e.g. propeller/popads), find the winner combo, and THEN take this winning combo to other sources. There's no need to spend money testing on more than one networks.


For an initial run in a tier 1 geo like DE, these are VERY encouraging results!

Especially considering you're still testing 5 landers, and also only running a single offer! After you arrive at a winner, and test a few more offers, and do some cutting/optimization, you may very well end up with your first green campaign!

I'm very glad you're split-testing landers this time! And your efforts were rewarded.


All sounds good! Great analysis and optimizing!


That would be good assuming the rest of the OSs have spent enough money and still showing a very negative ROI.

Let me point out a common newbie mistake - which isn't necessarily the case here - but I still want to warn you about it just in case (and for future reference).

Some newbies would drill down to a variable - OS stats for example - and try to judge the data the same way as lander data, and feel they need to pick a winner and blacklist the rest. This is NOT the way to do it.

It's not a split-test. For traffic segments, you'll want to make as many of them profitable as you can. The more green segments, the more profits for you.

Also keep in mind that as you cutting bad placements (and even other traffic segments in other variables that have very negative ROI), all your OSs' ROIs will increase.

So unless a traffic segment has spent enough money that you can be relatively sure it's not going to be profitable, leave it be.


Yup - that's common unfortunately,

If after say 7-10 conversions (for each lander), they're still going neck to neck, then just pick a random winner (probably the one with more conversions). Then optionally start another split-test - either for landers or offers.


You're doing well! I really don't have anything to add without seeing detailed stats. If you're still running this campaign (or if not, we can still do a post-mortem to learn from it), please feel free to post lots of stats from different tracker variables.


If some landers have really dreadful CTRs, such that it would require impossibly-high conversion rates to break even, then cutting based on CTR is justified, without having to wait for conversion data.

10k clickloss out of how many?

Also, clickloss can be worse when there's a lot of bot traffic. Try to compare the % clickloss you're getting now vs. before you started cutting placements - is the clickloss improving?

Other than that - it boils down to tracker redirection speed as I've talked about in the last post.

I hope you're still running campaigns - you're very logical in your thinking and based on the info you've presented, are doing all the right things. I'm seeing promise in your campaigns as well. You're definitely getting somewhere!

Again, if you want to look at some stats together, please do post them. Just one suggestion though: Upload images to imgur, and copy and paste the bbcode links into your posts. That way they appear nice and big. vBulletin's attachment function doesn't even work half the time. Imgur is way better.

Have fun - looking forward to your next update!





Amy

---

Hi Amy

Thanks for all the answers and encouraging words.

It's flattering to hear all of this!
But I am already jobless and I have never been happier. If you were me, you'd understand how fucking empty it makes you feel when you have a cushion job and forget the feeling of taking decisions and risks for yourself.
I actually developed flash video games in high school and that is the only time I can recall when I also felt free and afraid at the same time. So I am not sure why I suddenly decided I need to get a job with a salary. I guess it was about trying to prove something that I am a good programmer, but I digress
The virus was definitely something weird, but it was not from stress. I no longer have it and I am sleeping like a baby AND I am ready to roll when I am up in the morning

I have paused the germany campaign for the moment and re-evaluating all my data and actions I took. I am in the process of gathering offers in sweeps from lower tier countries (not asking AM's to do that for me, just browsing their platforms and applying). I believe I should venture into tier 1 geos once I at least have one green campaign that I've optimized and understand what the fuck am I really doing

Oh and I am already planning to run few campaigns in push, because that sexy 'What's working 2019...' report is convincing me

So yeah, I will most likely do two things next here:

1) Dump germany data for post-mortem.
2) Show you my next campaigns.

---

Hello Daliu, what do you suggest for beginner to start with ? Should I apply also on Mobidea, Clickdealer and Gotzha? What is Mundo ?

Why dont you use Google ads instead PropellerAds and PopAds - I first time hear for that two ? What model and tools to use to start with AM ?

---

how do you make landers? copy-paste from adplexity?

how do you make split test?

---

Hey @bibanum007

That's a very broad question, to answer what to start with I picked POP's (and now push) because there was already a guide on the forums. Simple as that.

Apply to Mobidea first - they auto-approve. Then aim for Clickdealer and Gotzha, I found they are easy to work with.
Mundo or Mundo Media is just another affiliate network I have applied to.

Not sure if Google ads offers popunder traffic, but I picked PropellerAds and PopAds because of the recommendations I saw from other members of this forum. As for tools and model(?), I would suggest to pick something that has good documentation and beginner guides. This way you can stay laser focused and work out all the kinks and problems along the way.

I do use adplexity for that. Sometimes, because I am in sweeps vertical right now, I copy prelanders I get from offers.

Split testing is basically like this:

Pick a variety of landers and a variety of offers. Combine them, so all of their flows are logical (i.e. if iPhone offer, only use iPhone prelanders), then purchase the traffic and send visitors to your various combinations of landers and offers. After a while or set budget, evaluate data (a.k.a. conversions) and see whats working best of the whole bunch, and you should probably end up with a 1-2 combos that perform the best of the worst (usually). Then, it's time to make just slight variations (changing color, buttons, some text) on those landers and then see if your ROI improves.

Actually, if you feel lost like another newbie I just saw, I posted a bit more detailed answer here: https://stmforum.com/forum/showthrea...l=1#post368842

---

Please start a separate thread to ask questions instead of hijacking daliu's thread.

Also: Please read through the 40 day tutorial before asking questions! You've just been asking a ton of super broad questions over the last few days. This will create a TON of unnecessary work for mods and other helpful members!

You need to at least do your own work first. We appreciate questions and I've never faulted anyone for asking them, but in your case, you need to do your own part first, instead of making everyone do double work.

@daliu5 Thanks so much for helping to provide the answers! You're quickly becoming an inspiration to a lot of other newbies that are just starting out. Thanks for this entire follow along!


Amy

Sent from my SM-G930W8 using STM Forums mobile app

---

Ha! OK then I guess what I said about embracing your job doesn't apply anymore.

I do understand that empty feeling you're talking about. But I also understand what it's like to be jobless and penniless. Was there for years and wouldn't recommend that experience for anybody - but I guess some people are meant to go through this stage, and some DO survive!

At the end of the day, ANYTHING is possible and success WILL prevail if we stick to something and don't give up on our dreams. I'm just always trying to warn people with the hopes that they'd have a less rocky road than I did. But I also know that some people NEED to go through hardship to learn lessons that will benefit them later on.

But I'd give that warning anyways. My mother hen instincts.

Personally, I was thrown in the deep end. I got laid off many years ago when I was still working as an engineer in the automotive industry - the industry was downsizing at the time and the plant was shut down. The day before that, I registered my first domain. The morning of, I was so reluctant to go into work because I was so excited working on my first site. 10 minutes after I got into the office I was notified of the news and was escorted out. I practically floated all the way home - although I knew I'd be living on credit from that point onwards. So yeah I know how that feels. But if I had known what the next few years would be like, I'd never have the courage to go through with that without looking for a job. But then in the end things worked out, and I've learned very valuable lessons I wouldn't otherwise have - most importantly seeking spiritual advancement. Another story for another day though.

And noted on your plans - sounds good!



Amy

---

You mean to start new thread instead of reply on each thread for which I have doubts ?

I read 40 day tutorial, but also it was mentioned that some infos are old. So If I have any doubts I need to create one thread with all questions inside , right? Beacause I dont get it - when I open new thread I got them deleted, when I reply on thread also doesnt work well. What to do when I have questions on particular case like this case ?

---

You opened like 20 threads already and received tons of valuable replies. But you never take the time to actually take the discussion further, but rather start another thread. On top of that, all your questions are very BROAD, it's like walking into a car manufacturing facility and asking "how to make money in the automotive industry" ... it doesnt work like that, nobody will give all the info to you on a plate.

You keep on asking the same questions in various threads, you get an answer, then ask the same thing again in another thread. It looks like you're not even giving it the time to read it and just expect the other forum members to explain everything to you, one question after another.

The answers for most of your questions are already somewhere on the forum, at least take the time to read it. You can't learn AM within a week anyways, it takes time to understand all the concept.

---

Home > > Newbie Follow-Alongs