Home > > Newbie Follow-Alongs

---

Learning and Building from First Principles (31)

---

A first principle is a basic assumption that cannot be deduced any further.

Building from first principles essentially involves being a scientist, figuring out fundamental truths and using that to build an understanding of the world.

Elon Musk explains it best here:



This is a documentation of my journey to understanding affiliate marketing from first principles.

I'm following Amy's 40 day tutorial and reading all the materials I can get (including all of Finch's stuff). I really appreciate @vortex for writing such a detailed and comprehensive guide with little to no loose ends. Thus allowing me to truly understand why something works the way it does.

I'm currently on Day 31

Summary of Situation

• I'm running a Malaysian Iphone offer currently as reccomened to me by my AM.
• I have 5 offers, although most are the same merchant but with on different AN with different stages (pre-lander, 1click, deeplink) which I think are still worth testing as separate offers. The offers are worth 0.24 and 0.28
• I've ripped, cleaned and optimized 7 landers. 4 in English, 3 in Malaysian. The LP speeds are all under 1 second.
• One offer got to 2 conversions first (3 of the others have 1 conversion), so I've paused all offers and am using that to test the landers
• I'm running on PopAds and have cut about 10 placements which have 2x payout in spend with no conversions.

Day 31 Analysis

Today I am comparing the results of different landers using the inbuilt Bayesian statistics calculator in Voluum as seen below



Day 31 Action

I am going to be cutting lander MY - Branded survey, EN - Quiz and EN - Branded survey as they all have 0 conversions with 12x payout in spend.

I will let the remaining 4 landers run until they break the 10x payout in spend or shows a statistically significant negative result.

Summary to Date

• 9 conversions
• $24.24 in adspend
• $2.13 in revenue

---

Day 32 Data






Day 32 Analysis

• Obvious point is MY - Spinner is doing pretty bad, only a 3% chance so I cut it earlier today.
• Only 3 landers, seems like the black spinner and the FB theme spinner are both very close. Interestingly that the Malaysian landers out perform the English ones generally. But we'll wait for more statistical significance
• Cut a handful of placements
• Vast majority of conversions come from Andriod, I guess people with Iphones don't want another?

Next steps:

• Wait for the 3 remaining landers to duke it out, quiz & giftbox likely to be eliminated next based on 10x payout per convr rule.
• Look for more offers to test once lander is found

---

Day 32 Data:



Day 32 Analysis:

• Again all landers are neck and neck 10/9/8 conversions respectively
• Cut some placements

Camp Summary to Date:

• 29 conversions at $6.93 revenue
• Spent $44.81
• Profit: -$37.88

Hmm, this is not going so well. My hypothesis is on my current path I'll keep bleeding money while the lander keep performing similarly for a while.
Not that I have anything against losing money at this point - it's just buying data and more importantly but I'm not sure what to do. Perhaps start another parallel camp?

---

I know I'm losing money and yada yada, BUT this whole affiliate marketing game is just so fascinating, so fascinating in-fact that I turned down going out with friends today and worked and read for 10 hours straight with nothing but a cup of coffee and some toilet breaks.

Honestly feels like back when I used to play WoW, sitting on my computer for hours upon hours, often forgetting to eat.

Anyway let's get into today's result:

Day 33 Data


Day 33 Analysis

• Cut Spinner black as it's chance of being a winner dropped to just 8%, now I'm left with just 2 landers both again going neck and neck at 12/13 conversions
• -65% ROI isn't too bad but I'll wait for maybe 1 or 2 more days before just picking 1 and starting to test offers.

Camp Summary to Date:

• 35 conversions at $8.37 revenue
• Spent $53.67
• Profit: -$45.30

Also I started a second camp, again Iphone 11 but in a different geo, spent all day but I went HARD taking everything I read, learned and know building this new camp

• Talked with 2 AMs and did hours of intense spying to confirm that these offers do indeed work
• Ripped, cleaned and optimized 7 landers, 6 of them loading under 0.5s. Went more aggressive with them this time

Let's see how it plays out, really did try my best this 2nd time round to use every trick and to take no shortcut.

---

Just a quick update as I've been quite busy with my day-job.

• The original MY campaign really took a turn after I found a couple new offers (one makes 2x the EPV of the previous "winner" and tried a new lander (which almost doubled my CR%).
• Right now I'm just letting 2 offers battle it out until I get statistical significance.
• One has 65% chance the other has 35% chance, their ROI's are -30% and -40% respectively which is a HUGE jump from the -75 to -90% I've been having to deal with.
• Hope I can begin optimizing once the winning offer is chosen , seems like -30% ROI is a nice really starting point

Also a quick progress update on the 2nd camp I started:

• I think iPhone was a little overplayed in the GEO so I pivoted to vouchers, couldn't find any lander so I ripped some iPhone ones and photo shopped vouchers on them and changed the colours to match the brands.
• It immediately was at -70% ROI overall with 5 offers and 7 landers
• I cut down to 2 landers and as they were neck and neck after 20 conversions ea I just picked one, since if they perform so similarly it wouldn't matter which one I picked.
• Now I'm cutting down offers, 4 were pretty bad with 0 conversions at 5x spend so I cut em quick
• Now 3 are battling it out, man it takes SO MUCH DATA to get a statistically significant EPV with a 5% cut threshold.
• Their ROI's are at -13%, -24% and -60% which is crazy promising, I just want to CUT CUT the 2 -24% and -60% and jump to optimising the -13% campaign since I think I can turn it green in no time, BUT the stats show the likelihood of being best is currently 74% 19% and 6.7% respectively so I will wait, and be patient - we are doing things from first principles after all

Crazy how I can go from a literally dogsh*t -80% as my best lander+offer combo to -13% by just reading, testing and applying what I learn here in the span of a weekend.

I'm learning SO much about testing, aggressive advertising and optimization. The D33-35 guide @vortex has written about optimisation is "end-to-end" logically complete. I've read it about 3 times and the info in it could easily sell for $497 in a course - AND it would be worth the money unlike most courses.

---

Oh yes, I remember I felt the same when I started with paid traffic, it was fascinating to watch the numbers all day long, the data comes in quickly and there is always something to do with it

Sometimes it was so hard to wait for significance and hold myself back from making changes too early.

And I see you're in the same stage now:

But try to wait some more so you don't make a premature decision

It's all about what you do with the info we have here on the forum, when applied correctly, it works as you described... but not everyone is able to do it, so kudos for progressing this fast!

This will, or should be, the breaking point since you're running in a very competitive vertical... one has to aggressive in order to stand a chance. You literally need to learn how to walk on the edge... aggressive enough to convert, but not too much in order to get banned.

You're doing very well so far, keep it going!

---

Thank you @matuloo you have no idea how much that means to me as you're one of the people who I've picked up the most from, a lot of my understanding of AM came from gold nuggets you've dropped on STM.

Daily Update

Now I'm not sure if this was the right decision but I have decided to pause/stop my original MY iPhone campaign.

Summary of Campaign:

• 93 conversions
• $21.48 revenue
• $103.26 cost
• -$81.78 profit
• 7 offers tested
• 7 landers tested

I was rereading @vortex's guide and decided to just do a quick estimated daily earnings by summing positive ROI placements and got only $2, now I'm not sure if I'm 100% certain this was the right figure as I could bid higher and get more traffic or try other traffic sources other than PopAds but continuing to run a -65% (cumulative performance of best offer) campaign for only a potential of $2 per day did not seem worth how much it would cost to make green (if possible) or how much time it would take

But this was my first affiliate marketing campaign so I still consider it a MASSIVE success, although it sucks to lose $80, I consider it an investment and would like to summarize for any newbie who decides to read this FA a year from now (like I have done )


First Campaign Debrief

• I learnt how to apply to networks
• I learnt how to setup tracking
• I learnt how to communicate with affiliate managers
• I learnt how to spy and rip landers
• I learnt how to cut placements
• I learnt how to evaluate landers
• I learnt how to evaluate offers
• I learnt how to estimate daily profit
• I learnt how PopAds works in some detail
• I learnt how to setup a FAST AF website with AWS/CDN/LP optimization
• I made my first $1 in revenue
• I made my first $5 in revenue
• I made my first $10 in revenue
• I made my first $20 in revenue

When put like that, how could I say I got a BAD deal from spending $80? Probably the best $80 I ever spent , especially since I know for every person who wanted to try affiliate marketing there is likely a HUGE drop-off at every bullet point above.

As a brief update on my 2nd camp:

• Gaining more statistical significance, however the other offers are catching up.
• Cumulative ROI of 3 offers shown above: -40%, -53%, -58%
• Thus the ROI's have dropped a bit since the weekend
• Increased my bid my 15% as I was quite low in the bidding position and I need more traffic

Right now the plan is just to wait and see which offer will prevail and then begin to try and optimise to turn the camp green.

I'll also be looking at starting another camp in the next day or 2. If anyone has any geo suggestions for sweeps I'd appreciate them

---

This one needs more eyeballs I think

---

Nice! Your first campaign results absolutely smoked my first ever campaign.

Also I need to check out this inbuilt Bayesian stats calculator in Voluum

---

Haha, thanks @stickupkid you're too kind.

Haha man, I learnt from your mistakes - standing on the shoulders of other newbies is the name of the game

Yeah the Voluum stats calc is awesome, can do ROI, EPV and CVR checks in seconds w/ no data entry. Result is the same as the other web-tools mentioned (I checked).

Daily Update


Today is looking good, things have gotten a lot more clear and by tomorrow I think I'll have a winner (no assumptions of course). Cumulative ROIs are -31% , -53% and -54%

After cutting a couple of bad placements and increasing the bid by 15% I've been getting more traffic AND a higher CR in the last 24 hrs.

The last 24 hours ROI of my best offer is -15% w/ 15 conversions. Perhaps not 100% statistically significant it's getting me kind of excited since I've done no optimization other than cutting a handful of bad placements yesterday.

I haven't been able to do Amy's D33-35 guide yet, as the winner offer + lander hasn't emerged, but hopefully I can make a bigger post when I do all the optimizations in the next few days.

Considering I'm still targetting iOS (which is 1/3rd of my traffic and has a -85% ROI) I could probably turn it green but just cutting it, but I'll stay disciplined and follow the process.

For when we win we party but when we lose we ponder, and all greatness comes from pondering.

---

iOS is often not performing well with certain offers, so if you see such a big negative %, it is actually a good idea to cut it. There is pretty much no way to turn a -85% target into a green one.

I've had MANY campaigns where I targeted android only, don't be afraid to try it. It's also a good idea to take a closer look at the data to see whether this is a general pattern or just certain placements are sending such poor iOS visits. That happens too and in such cases it's better to cut such placements instead of the whole OS.

---

Ok fair enough, did a drilldown and found 3 positive placements with 1-2 conversions all the rest being negative. I don't think it's worth trying to target those down so I'll blacklist iOS.

Daily Update


Today there is more statistical significance and I'm pretty happy to finally cut my 3rd offer due to it dipping below 5%. Now just the 2 offers are left to duke it out, although to me it's a clear winner as the difference is -25% ROI and -45% ROI cumulative I'll let that 9% hit 5% before cutting.

Otherwise nothing too much to report, I've cut a handful of placements in the morning and the 24hr ROI of my best offer is now at -11% with 16 conversions.

I have just blacked-listed iOS so I guess we'll see how the performance plays out tomorrow, stay tuned for the next episode of DBZ

---

When I see difference as big as you have here... so same traffic and 48 vs 29/22 conversions I don't wait for any additional significance and go for the clear winner which is the offer with 48 conversions. Assuming that the payouts are the same or very similar on all offers, of course.

---

There is a 30% payout differences, with the lower conversion offer having the higher payout. Thus I ran the test on EPV rather than CVR. But you were right in the end.

Daily Update

So today a quite a bit more exciting than the past few days because I have FINALLY narrowed down to a winner lander + offer .

Thus I was able to follow the D33-D35 guide and optimize. So I cut out any segments that had a lower than -85% ROI and had at least $0.50 in spend.

Just observing right after all the optimizations (cutting desktop, cutting iOS and cutting a bunch of smaller segments) it seems traffic has slowed down to quite a trickle, at my current bid.



My spreadsheet is showing an estimated daily profit of $15, but I guess we'll see how accurate that is.

Next Steps:

• Test out Monetizer as it seems quite a lot of people are having success with squeezing out additional ROI - worried about load speed so I'll split test a lander both w and w/o it
• Test out a staggered bid approach in PopAds via duplication, I can get about 70% more traffic by increasing bid 34% - but I'll do it in a separate parallel campaign.
• Try run on PropellarAds
• Test out 2 new offers I found
• Also I'm going to try launching a new campaign in a T1 English speaking geo, from all the advice given to newbies this is going to be harder than giving birth so god bless my soul

Quite a busy weekend planned - most likely another 10 hr / 10 hr work weekend.

But you know what they say,

Workin' on the weekend, like usual
Way off in the deep end, like usual

---

Ladies and gentlemen, WE ARE GREEN!



Of course let's not get too attached to green.

In fact, I think I think the first green day is a day to be extra cautious and perhaps a little scared as we're reaching one of the most devious and devastating Great Filters.

In brief terms, a great filter is an evolutionary hurdle which wipes out the majority of those trying to cross it. Here's the wiki-link.



I have read perhaps a 100 follow-alongs, and one of the primary things I look for is not how someone succeeds, but rather what causes someone to fail and get wiped out.

It seems that shortly (although not immediately) after getting a green day the likelihood of giving up increases greatly.

I didn't understand it at first, it didn't make much sense to me - surely people would be more motivated after learning that affiliate marketing works.

I think the problem lies in over-attachment, because the first green day/campaign is so hard-won it causes people to not want to go back to red, ever - at the cost of not trying new things. Which is of course spells death in fast moving environments such as pop-traffic and even affiliate marketing in general.

Daily Summary:
Today was a pretty productive day and I aimed to test as many things as possible.

One of the main things I tested was bids, part way through the day I launched a duplicated campaign on PopAds with a 34% higher bid in hopes of capturing 70% more traffic (as the PopAds interface indicated)

The result of that is:



A -20% ROI isn't too bad on it's own, however it pretty much stole ALL the traffic from the lower bid campaign (the green one above).

It makes sense as it has identical targeting and a higher bid but I don't understand why @matuloo and @vortex recommend it, as it seems to cannibalize my earlier profitable campaign for a negative ROI campaign. Perhaps there is a strategy I have missed.

I'll leave it running for now to gather more data, but really it does not make much sense as I am bidding more for largely the same traffic it seems.

The second thing I tested was PropellarAds, with a smart CPA 2.0 goal.



It's performance was/is pretty crazy. It started off deeply red for the first $7 or so in spend at -60% ROI then shotup and crossed breakeven in the next $7. If I just look at the last couple hours of data instead of the whole day it's at a 65% ROI. WTF, is this black magic?

With such craziness I've decided NOT to scale super hard without confirmation of lead quality from my AM, one I get it (+ hopefully a pay bump) I'll unthrottle it.

The third thing I wanted to test is Monetizer

I've signed up but am not super familiar with all the methods available, plus it seems they have to add my sub-domain to their SSL cert which will take 24-72hrs, without it a big fat red unsecured page shows up on chrome - which will probably kill conversion rates.

From what I can understand, it seems Monetizer works in 2 main ways to boost ROI:

• Collection of push subs on your LP to sell as push traffic on Advertizer (gives you 80% as residual revshare)
• Back button redirect to a smartlink which goes HARD trying to get the poor sucka to convert.

Seems interesting and I can't wait to try it.

Did not get time to launch a new campaign today as I thought it would be wiser to focus on scaling and understanding what works first and I've read the last couple of Amy's posts about scaling.

I'll probably launch 1 or 2 more campaigns on pop traffic and suck out all the lessons I can get from running pop before trying another source, perhaps push or native? I could probably just take my winner offer + lander straight to push without much modification it seems.

---

Daily Update

Just a quick update today. I let the high bid campaign run for most of today but ended up cutting it due to poor performance at -15% ROI as the majority of conversions were coming from placements that the low bid campaign was once getting. Low bid campaign, due to picking up shitty scraps was at -70% ROI before I paused the high big campaign now seems to have recovered a bit and is at -40%.

I'm not going to mess with bids as it doesn't seem worth bearing a $50 loss to gather placement data to blacklist in my high bid campaign just to make maybe $5 extra per day it seems. I may be wrong but I think it's more prudent to focus on higher ROI activities.

I finally got Monetizer working and now am collecting push subs and money from the back-button redirect. Made like $0.04 lol, but the split-test shows NO drop in conversion rate when compared to the previous default lander which is interesting for sure. Infact it shows a slight increase, but may be due to chance alone.

Thirdly, I've let my PropellarAds campaign continuously run, it was doing quite well for most of tody at a +30% ROI but now it's running at only +5% ROI. Don't want to get a tin-foil hat on but I feel by passing back both the conversion AND the payout it's simply uses the payout to optimize (and not my set conversion value) and thus trends towards breakeven. I've now set the passback payout to 1 cent to allow it to register converions BUT to not know how much I make. May require a restart of the campaign but we'll see.

Also I've been working on a new campaign on a geo with much more potential, with some more aggressive landers - pretty interested to see how it turns out once my offers are approved.

---

You are right, the first "green" is awesome but it can cause trouble to some peoples mindsets... as you mentioned, it literally hurts to return to the red area again, once you've seen a green campaign. But that the way it is and no attaching yourself to the green color is a wise decision. More red days are about to come and it's absolutely normal.

You have to realize one thing, you're not the only affiliate bidding in a certain spot. It's a fighting pit and if you don't raise the bid, someone else likely will... so they will get "your" traffic anyways. Your goal is to bid as high as possible, while still maintaining some ROI. With some campaigns, you will be able to be the top bidder and grab most of the traffic, but some campaigns won't perform that well so you will have to settle with whats left from the top dogs.

Higher bids often equal to better traffic quality, since the higher bidders are served first, which means the users didn't see that many ads yet and should be more prone to converting. It's not a definitive rule, but quite often it's the case.

You need to test several bid levels to see what works the best for your current campaign.

---

Daily Summary:

• Built out my 3rd camp now and will begin running traffic to it, payouts are much higher has it is T1, but traffic costs are also a bit higher it seems. It should balance out but will be interesting as it has much more potential to scale.
• Got a reply from my AM saying the lead quality is good and that I'm uncapped - exciting. Let's see how hard I can scale this. Maybe I should ask for a pay bump?
• Noticed my Monetizer-enabled LP takes twice as long to load, optimized and redid another test and found a massive difference of -60% ROI and 120% ROI

• With such a crazy difference I immediately canned the monetize lander and am now back to my normal lander
• Also spent a lot of today messing around with PropellerAds, mostly losing money while tinkering with the settings. I think the smart CPA 2.0 is a bit like FB where if you mess with it at all it immediately tanks. My monetizer test causes the cr% drop and traffic reduces down to a trickle so I had to restart another campaign and wait until it optimists again.
• Also I'm going to be testing 2 new offers against my best offer from a new network I joined.

---

Looking through other peeps' follow alongs I was considering adding this...but that's a huge difference in load speed

---

This is possible, at least on the first load. But once the CDN kicks in, I imagine the js loads relatively faster. I haven't noticed a big performance difference in my campaigns.

Perhaps, @Monetizer can chime in?

---

Looks like you posted a preview of the image, it's too small to read... can you post the full resolution one please?

---

Thanks @matuloo that really clears up why testing bids is important. Really awesome to hear the strategic thinking behind it all.

@nitrousoxide I DID manage to get the load speeds quite similar, by not following Monetizer instructions and just hacking it my own way haha. Despite that it still unperformed quite terribly.

Thanks for the insight @jaybot , that interesting. I wasn't too sure if the Monetizer had a CDN as the js script is hosted on their end - which does slow everything down. My workaround was just to very carefully rewrite the code so that it would effectively do no external calls - and it still worked and was similar in speed.

I think maybe the push notification pop-up dissuaded people. I'm not really sure but the massive difference (in the image below) with only 1 hour of data was decisive for me to can it. Based upon your recommendation I'll give it another try.

Ah sorry @matuloo didn't know why it shows such as crap low rez verson. Here is it reuploaded:

---

Daily Summary:

So I tested the 2 offers from the new network overnight. I would post a screen shot but it seems the post-back was screwed up so here's the raw data (each with the same spend)

• Current Offer: $16.8
• New offer 1: $11.3
• New offer 2: $14.20

Given that they under-performed and that I don't want split up ALL my payouts among a lot of affiliate networks I'll just stick to my current offer.

Also it sucks that my worst AM is on the network with the winning offer, the dude barely responds to my messages, often says he going to do something and never does. Meanwhile I have other amazing AMs that respond instantly, are proactive and even do a bit of spying for me.

Also I asked for a pay bump and he said he wants to see what my breakeven point is and a screenshot of my tracker, is this normal?

(EDIT I got a 7% pay bump, not the biggest but still decent I guess)

Can the big boys (and girls) chime in here - would it be terrible if I asked the network for a different affiliate manager? Trying to scale my offer is so much more painful than it needs to be with this guy.

Also I launched a campaign on PropellerAds - spent $5 got 2 conversions then traffic slowed down to like 5 cents per hour, did this again and again. 3 times.

I managed to get like +30% ROI with Smart CPA 2.0 with 80 conversions (but I messed with blacklisting and campaign tanked) in the weekend so this algorithm is indeed powerful. But mysterious, and punishing.

Did a bit of reading (man I should readup a lot more) - turns out it uses machine learning similar to FB so not touchy touchy it seems.

So I launched a SmartCPM campaign with auto-optimization enabled, with all blacklisting done up front and with a whitelist based on the placements that converted well from the 80 convr campaign. Here's the results: -44% ROI but I'm still soo happy that it's actually spending haha, hopefully it'll auto-optimize and I won't be touching it this time.




Also, I think you guys are going to get a kick out of this - my first foray into a T1 geo:

• Payout is about $1
• Spent $30
• Literally am using all the tricks I know for the whitehat landers and even threw in a black~ish hat lander into the mix just to see what'd happen.
• All landers load in under 0.5s
• All offers thoroughly vetted via spying and AM

NO CONVERSIONS

Man that's SO brutal - wouldn't expect this.

Tested all offer links twice, test landers thrice, checked in with my AM - traffic WAS indeed coming.

Scratched my head, maybe it was my traffic source?

So I went and checked.

I had everything I wanted blacklisted as whitelisted...

Man, oh man, but I'm glad this happened. Truly blessed - really and honestly.

Kelly Ballard in his 2017 AWE talk said he launched a campaign late as night in a rush to get the camp out (similar to me) and burned $10,000 overnight.

Always double check your black/white lists. I learnt the same lesson for $20 as he did for $10k.



Fixed and threw another $20 spend at it, let's see if it converts.

But more importantly today, railing helplessly against the PropellarAds algorithm and the targeting mistake (spent an hour trying to fix it) has taught me something important.

That I really don't know anything. Despite having done all 40 days of @vortex's tutorial, I still don't really know anything.

and THAT ladies and gentlemen is a great thing.

I have ascended an important step.

I have gone up from unconscious incompetence, to conscious incompetence.



To many more future steps.

---

Daily Summary:
Not much to report today. My T1 campaign in Spain after $20 in additional spend had 0 conversions so I've shelved it for now.

• Top is my PopAds campaign, happily chugging away between +10~25 ROI
• Second a new PropellarAds campiagn I launched with Smart CPA with a higher bid (due to yesterdays pay bump), -60% ROI but I'll let it keep spending and see how it goes
• Third is my white-list PropellarAds campaign with Smart CPM w/ auto optimisation, -50% ROI but again I'll let it keep spending.
• Also the budget are like $50 and the bids are set to auto (whites is about 5x average) but the camps only spend like 20% of that. Not sure why.

Think I'll give pop one last try, on a T3 geo before moving on to another traffic source.

Also I'll try my current campaign on another pop traffic source, to see how that goes.

---

AM's often ask about how big payout you need in order to become profitable, so they can decide whether such a bump is realistic as that would likely mean instant volume increase from your side. In the end, the more volume you send, the bigger bonuses they get But I would definitely not show a screen from my tracker to an AM just like that... would have to be someone I have VERY good relationship with.

BTW: Do not push for payout bumps too soon, it's better to send stable volumes for a while, then think about bumps. 7% is decent btw, nothing special, but not bad after such short period of time.

It's quite common, especially when the AM is in a different time zone ... just do it in a polite way, no need to harm anyones feelings

I didnt test their auto-optimization, but if it's not bleeding heavily, try to leave it on for a while and maybe it will show some promise.

Lesson learned, right? I had similar ones too... once I've set my bid to 80 cents per click instead of 8 ... the $$$ went pretty fast ))

BTW: when testing a new GEO, I rarely start with just 1 offer, there is still the chance of hitting a poor performer, while some other offer could have worked good. Having a strong offer is the most important part of the puzzle these days.

---

Yeah for sure, I did ask for a quality check to which the merchant said the quality was good and had delivered about 350 leads over 10 days at that point before asking. Perhaps it was a bit too early, but I guess I can only learn that by testing asking earlier and later.

Fair point, professionalism is indeed key. I'll hold off asking for another AM for now as he did deliver a pay bump in the end. Also I checked, seems like I was auto-bumped 2 days BEFORE I asked so I'm actually making about 25% more than street. Funny that, no one told me my payout was bumped and here am I getting upset


The problem is that the traffic slows down from spending $5 per hour to 50 cents for all of today. Really a finicky thing but I'll let it run.

Haha man, ouch. I read some stories of people who bid on a CPM basis on a CPV platform which would be super painful.

Yes of course, perhaps I forgot to mention I tested the 4 best ES offers from 4 networks in rotation - still no conversions.

There's obviously some entrenched players at this point so I'll leave ES be for now. I suppose this ties into your earlier point where people cut and raise bids again and again until they have monopolised key converting placements to the point where my bids will never even get me 1 impression from the promised lands.

Daily Summary:

• +66% ROI is pretty cool to see, but I over-optimised and will be trying to get more traffic and cutting back some optimizations
• -77% ROI sucks to see, but the small $2.61 spend sucks even more - especially since I have a $50 budget and a high bid. PropellerAds is perhaps beyond me - seems like you need to launch at the right time, right place and pray to the gods to get some early conversions so the algo ascends your camp to heaven.

---

@fastaj what cpa network are you using?

---

This is pretty common... when we optimize too much, there is no traffic left and I believe the traffic network algos prefer campaigns with less optimization, since it helps them to sell more inventory. So sometimes it's better to keep running at lower ROI but get more volume.

The performace of your campaign probably seems to low, so the algo isn't pushing traffic your way. Increasing the bid should help.

Could have been a default network wide bump, advertisers sometimes decide to bump payout for everyone, in case the leads are backing out nicely for them. It's still weird that the AM didnt tell you, usually they rave about it as it's their main selling point. I guess the AM you have is not dedicating much of their time to you, since you're not sending that much volume yet.

---

Hay, I'm using ClickDealer, Gotzha and Big bang ads.

Yeah I think you're 100% right about the over-optimisation thing. Especially based upon the results I got today.

Also, very interesting point about the network wide bump, I have no real way of observing what the default street payout is on my network, checked the same offer on another network which has the same lower payout and assumed it was just for me. Your theory about the network-wide bump is quite plausible.

Daily Summary:

• Pretty good result today, above row is PopAds, I cut back on some optimisation and got 3x the traffic but at 10% lower ROI - definitely worth it as profit > ROI.
• Second row is PropellerAds which pretty awesome, since I got 4x the traffic but 16X the conversions from PropellerAds - it's at -29% ROI which of course isn't great but damn it brought a smile on my face . Thank you propeller algo gods for shining favorably on me. Now make that sucker +60% ROI @ $100 daily ad-spend and I'll pray to you every night

Otherwise things have been quite busy - been pulling 12 hour days at day job which starves time I spent on AM last couple days. Fortunately the weekend is here - you know what that means, time to pull 12 hour days again but working for myself which is of course is so much more rewarding.

Current plan of action:
- Try my campaign on a bunch of difference pop sources - not sure which are good but Advertizer seems promising.
- Give monetizer another try as per @jaybot suggestion.
- Take a pause to rethink my understanding of affiliate marketing

• Go through @vortex's 40D tutorial again with a much more critical view
• Start reading FA's of more advanced members instead of just newbie FA's
• Spy on US LPs without intention of ripping but simply to understand how the game is being played at the most competitive level.

Perhaps a more reflective and consolidative weekend is due in contrast to wild action taking I've been doing for the past month.

Better to steady your gaze and aim for bulls-eye once than to shoot 100 times blindfolded.

---

Great that you were able to pull more volume, quite often its a hit or miss game with the algos... we never know exactly how they distribute the traffic, so it's about testing and developing a style. For example, what worked great for me on many sources was to simply clone campaigns in order to get more volume. At times, I had 10 campaigns running on propeller, all with the same targeting and ads... and the volume was X times higher compared to running just one campaign. Of course, this doesn't work all the time, but it's another tip you could try.

Yup, take a day or two off from AM, focus on other things, clear your mind ... you will start to see the things from a different perspective again. It's always a good idea to give the thoughts some time to soak in and evaluate what you've achieved so far.

Looking forward to an update from you on monday

---

Hay @antoine, thank you for the kind words, I'm glad you got value out of this FA

---

Home > > Newbie Follow-Alongs