Home > > Newbie Follow-Alongs

---

2000$ and 3 months to make some profit– my journey (57)

---

Hello,
In the early 2000s I worked as an affiliate for a long time (promoting mainly adult) but about 8 years ago I decided to move to something else.

Now I'm back, with a 2000$ budget and 3 months to catch up with the current trends and technologies and to make some profit. The scenario and the market are completely different now and I want to learn everything from the beginning.

In this thread you will read my journey, spending, questions, learnings and discoveries.
I am following the Vortex' 40-Day Tutorial step by step and I think it is one of the best tutorial I ever read. Thank you so much vortex for writing it!

I'm still working full time on my day job. I want to learn from this journey as much as I can and I have the peace of mind of having a good salary and investing in this side project without the risk of going bankrupt.

MY GOALS:
Short term: learn as much as I can and launch my first profitable and scalable campaign in 3 months.
Medium term: double my 2000$ investment in 6 months.
Long term: make at least 2000$/month in profit consistently in 12 months.

What do you think about my plan?

Day 0: I bought a subscription to STMForum.

Spent:
99$ - STMForum subscription
(1901$ left)

Learned:

1. payment gateways aren't always working well. My payment was processed but I had to contact the STM support to have my account activated. They were really helpful and nice and everything was fixed in a couple of minute

Day 1: Applying to Various Networks (Voluum Version)

Spent:
75.64$ - Voluum subscription
1.18$ - Namecheap domain registration
(1824.18$ left)

Learned:

1. my Gotzha registration is still pending. Not all the affiliate networks are working the same and have the same timing.
2. Getting in touch with the AM is really helpful. Building a good relationship with them is a really nice idea
3. A cheap domain is good enough to start with. XYZ TLDs are really cheap on namecheap
4. I have some problem with my VAT number. For whatever reason Voluum can't properly track my European VAT ID and they added the taxes on top of the regular price. I will try to figure out why later.

Day 2-8: a) Picking Offers – Mobidea

Spent: 0$ (1824.18$ left)

Learned:

1. the mobidea panel is different from the one in the tutorial screenshots. Nothing major but the interface is not exactly the same.
2. The link to the developing Countries in the tutorial is not working anymore. An archived version of it is available at webarchive https://web.archive.org/web/20190717084547/https://isi-web.org/index.php/resources/developing-countries
3. The EPC is not easy to find in mobidea because they changed the interface. You have to open the single offers one by one and check it. It was a blocker for me and that's the reason why I decided to skip this lesson and move to the next one. Later I discovered (thanks to my AM) that they have a page easier to use for my search https://affiliates.mobidea.com/opportunities/offer . I will pick a mobidea offer to test later in this journey
4. The "conversion flow" column in the excel file template is something I have to investigate better. The wording of the different conversions flows are different from one network to another and I need a glossary.

QUESTION: Do you know where can I find a conversion flow list and glossary?


Day 2-8: b)Picking Offers – Clickdealer

Spent: 0$ (1824.18$ left)

Learned:

1. I made a mistake and I chose a "pin submit (mo flow)" offer instead of a "DCB (1-click flow)" or "DCB (2-click flow)" as suggested in the tutorial. Probably I set the wrong filters and I didnÂ’t notice it. Next time I will pay more attention to it. Using the clickdealer filters for the first time can be quite confusing
2. I found quite confusing that some offers accept traffic from all countries but are written just in a specific language and contain a language tag in the title (ie: /FR). QUESTION: Is it common? What does it means and how are we supposed to work with those offers?

Day 9: Setting Up Tracking on Voluum

Spent: 0$ (1824.18$ left)

Learned:

1. The naming is different from what suggested in the tutorial because Voluum now force Country name, Traffic Source and Affiliate network names at the beginning of every name. Not a big problem but good to know. My campaign is named PropellerAds - COUNTRY - DATE - Main - CARRIER – OFFER NAME
2. I forgot to set the voluum postback url into clickdealer. I realized it when I made my first conversion and it didn't show in Voluum (SPOILER: I already had a conversion in the first 6 hours after launching the campaign!) and to fix it I learned how to manually upload a conversion to Voluum (https://doc.voluum.com/en/conversion_upload.html –advanced reading but useful if you want to learn something new).
   After realizing it I set the postback URL properly and Voluum recorded the other conversions properly (SPOILER AGAIN: I made THREE conversions before the end of the first test campaign!!!)

Day 10: Setting Up a Campaign on PropellerAds – ACTION

Spent:
100$ - PropellerAds deposit
(1724.18$ left)

Learned:

1. Carrier names are not always up to date. Thankfully, the offer I chose have 2 carriers and I used the one I recognized in the propellerads list. The other one was not there because I think it changed name. I did not bother checking and I just chose the one I recognized in the list.
2. There is a 4$ service fee when depositing 100$ with credit card.
3. There is a discrepancy between the spent displayed in propellerads (10.049$+vat) and the spent in Voluum (8.54). QUESTION: How can I fix it?

Day 11: First Lesson on Testing & Optimization

Spent: 0$ (1724.18$ left)

Campaign results:



Learned:

1. The campaign ran for 24 hours and I made 3 conversion! I was really happy! The profit is negative as expected but I already learned a lot.
2. I set a bid 15% higher than the average recommended in PropellerAds to be sure to receive enough traffic to test the campaign. It worked well.
3. As recommended I played a lot with the Voluum filters (way more than the 15 minutes suggested). Sometimes I had to check their documentation to fully understand how does it works. The documentation is really well done and answer lots of questions.

Day 12: Setting Up a Campaign on PropellerAds – EXPLANATION

Spent: 0$ (1724.18$ left)

Learned:

1. The exact definition of zone. It was quite confused in my mind
   A Zone is a website where our ads are displayed for that website's visitors to see. It's also referred to as a publisher site, or an ad placement or simply a placement. PropellerAds pays these placement site owners for showing their visitors our ads, and in turn sells this traffic to us with a markup. I'll be using the words "zones" and "placements" interchangeably in the rest of this tutorial
2. I knew most of the details explained but it is always really good to refresh my mind with a well written tutorial

Day 13: Setting Up a Campaign on PopAds

Spent:
10$ - PopAds deposit
(1714.18$ left)

Campaign results:


Learned:

1. I set the minimum budget allowed (2.5$). It completed in 1 day with 0 conversions. I used the same offer as in propellerads but a different carrier.
2. PopAds and PropellerAds works almost the same in terms of campaign setting. It was easy to set up.
3. Prefetching is really interesting. I will use this feature in the future
4. I really like this quote from the tutorial:
   As long as your bid is average or above, that should be enough to ensure traffic quality, even without restricting any of these other settings.
5. There is a discrepancy between the spent displayed in popads and in Voluum. It isnÂ’t huge (2. 21$ VS 2.5$) but I will have to find a fix for it later

Day 14-15: Introducing Landing Pages

Spent:
129$ Adplexy Mobile
(1591.18$ left)

Learned:

1. There is an extra layer of complexity and I never worked with landing pages before. It is something exciting and scaring.
2. If you search online you can find some really good discount codes for Adplexy Mobile.
3. I decided to buy Adplexy Mobile to speed up my learning. I consider it an investment
4. I know the basic of HTML, CSS and JAVASCRIPT but the links provided in the tutorial are really useful to learn something more.

---

Hi barbarollo, welcome to STM.

In the last few years probably many things changed from what you did before.

I merged your posts, would be great when you try to edit your posts rather than creating new posts.
That helps to keep the thread cleaner

These are pretty realistic and doable goals.
In the beginning don´t try to focus too much on earning 6 figures month or so.

Use the first time to run rather lower payouts to learn the traffic and gather data.
On Pop traffic the placements often can be used for many different Geos (good ones as wella s bad ones).

So when you gather enough stats on lower payouts or lower tier campaigns you can use a lot of the data as well for other campaigns.

That way you can save good money in the beginning.

Good question, HEREhttps://charlesngo.com/mobilecontentoffers/ you can read a bit about mobile content offers and the different conversion flows.

Don´t really know if I get what you mean
When offers accept all countries then you should be able to send traffic from all countries.
But usually when it´s tagged with something like /FR it only accepts tryffic from that specific country.

Do you have a specific offer?
Then I could check it.
Or just ask your AM for clarification.
Sometimes it also can happen that offer title and description doesn´t match 100%.

Check for clickloss.
Especially on Pop traffic you can have rather high clickloss compared to other traffic.
That´s just normal.

The trafficsource shows cost for all traffic that was generated in the campaign.
In the tracker you can only see the cost for the traffic that reached the tracker.
That´s why such discrepancies can happen, because of trafficloss.

In some trackers you can set a % for trafficloss so that the costs are more accurate.
Just ask Voluum support if there is an option for it as well.

All the best on your journey.

---

Thank you for your reply! It's good to know that my expectation and budget are realistic. And thank you so much for the links and suggestions. I'll check them and ask to Voluum about the clickloss fix.

Following an example from clickdealer

---

Alright, I see what you mean.

In that situation the title is right.
The offer is only for Thailand traffic and only for carrier Truemove.

When you click on that offer you can also se more info about it.



As I said, sometimes such stuff can happen.
But usually the country from the title and in offer description is correct.

---

Day 16: Picking a Vertical and Applying to Offers

Spent: 0$ (1591.18$ left)
Learned:

1. Targeting just a specific Country or language in adplexity will restrict too much the number of landing displayed. Using a more broader search filter helps a lot (but translation will be needed)
2. I like sweeps. They looks easy to promote and start with. I’ll start with sweeps and two Tier 3 Countries.
3. I don’t think I will be really aggressive with the landers at the beginning. I prefer to learn and to avoid any ban or problem with affiliate networks, advertisers and traffic sources
4. In some tier 2/3 Countries there aren’t a lot of SOI and DOI sweeps offers available on clickdealer. I selected Voucher/Leadgen as vertical and I see more offers now. QUESTION: Is it correct?
5. I don’t see a lot of offers for sweeps SOI/DOI for tier 3 Countries on clickdealer. QUESTION: Is it normal?

Day 17: Downloading Landers from Adplexity

Spent: 0$ (1591.18$ left)
Learned:

1. I browsed a lot of adplexity landing suggestions. Lots of them are carrier specific and I don’t want to target just a specific carrier right now. I want something quickly adaptable to all the needs.

Day 18: Setting Up Hosting and CDN

Spent:
1.18$ - Namecheap domain registration
(1590$ left)

Learned:

1. I followed the guide step by step. I have some IT knowledge and it was really easy to follow the setup instructions and the explanation. The first landing page is showing properly, the DNS resolve fine, CDN and SSL is set up. It is working!

Day 19-25: Fixing Up Landers

Spent:0$ (1590$ left)

Learned:

1. I tried to use crossftp and sublime text but it was more complex than I tought. After trying to make it work for 15 minutes I decided to skip this setting for now and to upload the files to Amazon S3 using the browser. I will come back to this configuration later but I don’t think it is mandatory right now.
2. I fixed 2 landers. It’s not easy to find in adplexity landers in the language I’m trying to target and with the right offers
3. I didn’t understand if the getURLParameter Functions are mandatory or not. I didn’t find them in the landers I ripped. QUESTION: Should I add those parameters for tracking purposes? Is it mandatory to perform the first test campaign or is it just a fine tuning I can add later?

---

Day 26: Optimizing Landers for Speed


Spent:0$ (1590$ left)

Learned:

1. Following the tutorial and after some optimizations my two landing pages are loading in 1sec or less. Being the first landing pages I’m happy with the results

Landing 1


Landing 2




Day 27: Introduction to Offer+Lander Testing

Spent:0$ (1590$ left)

Learned:

1. I set the first campaign following step by step the tutorial: PropellerAds (10$ budget, Mexico, 1 carrier, Android, 3G only) + 2 landing pages + 1 offer.
2. I’m still not sure if I have to configure the tracking parameters adding the javascript codes in the landings. So far I didn’t add them. QUESTION: Should I? What is the purpose?

3. I followed the following rules as explained in the tutorial


Day 28: Doing a Bot Test


Spent:0$ (1590$ left)


Learned:

1. bot testing looks a little too technical for me so far. I decided to follow this suggestion and to skip the bot testing for now.

Day 29: Verifying Tracking Etc.


Spent:0$
Revenue: 1.8$
Total: 1591.8$ left


Learned:

1. Following the tutorial I tested everything and it is fine
2. I started a campaign journal. It looks like a great advice
3. When the campaign ended I manually updated the cost because it was about 15% lower in Voluum compared to PropellerAds. I learned that clickloss is real and it’s better to manually update the cost of a campaign before analyzing it

Following the first campaign results!

---

Yes, this is normal.

It´s just the nature of such offers, leadgens vs sales.
Let me explain it in a very basic way.
On sales the advertiser makes money as soon as a conversion happens.
On leadgen the advertiser pays your revenue when someone just gives some info.
At this moment the advertiser didn´t make any money from your conversion yet.

And you can imagine that info about people from Tier1 Geos are much more interesting for advertisers because they can make much more money from it compared to info from people from Tier 3/4 Geos.

It´s necessary to add them to the URL when you call out some dynamic values from the parameters on your landing page.
For example when there is something on your lander like "Your Samsung Galaxy S9 just won a price" then probably the Samsung Galaxy S9 is called from the values in the URL like &brand=Samsung&model=Galaxy S9 and then you need to add these parameters.

When you don ´t use any values from the URL on your lander you don´t need to add the parameters to your landing page URL

---

@barbarollo Great Follow Along, I'm very excited to see your results bro, keep going
Goodl luck in your journey.

---

Checking my stats trying to understand why there were differences between the payout tracked in voluum and the one tracked in clickdealer I discovered that some offers have payout in Euro in the description but clickdealer is showing it in USD and Voluum is showing the Euro payout but in the wrong currency. QUESTION: How can I fix it in Voluum? Should I use the currency setting in the Voluum offer configuration? Will it fix the past stats too?

Clickdealer report


Offer description


Voluum report

---

Thanks for the nice words! Are you following the tutorial too?

---

Quick update about the first campaign: I stopped the landing page less performing and blocked one zone that wasn’t performing well. I will spend another 5$ to collect some more information

In the meantime I launched a second campaing with another offer (10$ budget, 1 Country, 1 carrier, Android, 3G only)
The traffic is slow but performing well. I just cut the landing page less performing and one zone that wasn’t converting. I will run it for another day to see how it will change. Following the stats of the second campaign

---

@barbarollo Looks like you're making tremendous progress! Apologies for my absence - I've been in Bangkok for AWA and then vacation. Finally back home in Canada and have caught up on sleep and the most urgent work.

But you're in very good hands - twinaxe is extremely experienced and has a passion for helping newbies. We're so happy to have him join the mods team!

To answer your latest questions...

Yes - you can specify the currency of payouts in Voluum settings. You can change the currency of an offer going forward, but can't change it for conversions that are already posted.

Please see: https://doc.Voluum.com/en/adding_offer.html





-30% ROI before significant optimization - looking hopeful!

Just wanted to check: Are you using a stats calculator to make lander-cutting decisions?

Which traffic network are you running on? And are you throttling traffic or running on a limited budget (i.e. reaching your budget daily)? I'm wanting to gauge whether you're getting enough traffic to warrant continuing. If traffic is already very limited even before you start cutting, and optimizing further will only dwarf the volume even more, then it may not be worth it to continue - unless there are other sources with more traffic in the same geo+carrier+etc.




Amy

---

welcome back! :-) I hope AWA was great!

Thanks for the the help with the currency and Voluum. I will use it soon.

I'm using the stats calculator to make lander-cutting decisions. I'm working with propellerads traffic, not throttling it. The campaign had low traffic and I decided to stop it. The second campaign was more promising and had lots more traffic but I decided to stop it anyway.

I spent the last week reading all the tutorial again and re-thinking about the strategy to use before running any new campaign and before optimizing the ones I discussed in this topic.

I had the feeling I was missing something and I needed a more structured approach. I also learned how to use adplexity better and I made a couple more landing pages. I also applied to other affiliate networks and found 2 more offers I want to test.

Today I will start the campaign again from scratch: same geo, same carrier (the one previously tested that was bringing more traffic), same devices, more landing pages, more offers from different networks to test and a more structured approach. I'll update you soon!

---

What are you running and how many landers do you test?

I just ask because you should keep it as simple as possible for your tests.
Test rather just few different landing page styles instead of many different variations of the same style.
There are not so many real different LP styles out there.

For example for sweeps you basically only need to test 3 landing pages maximum (wheel of fortune, questionnaire, gift box) to see if it works or not.
When you find a working combination you can test variations of it.

But first task should be to find something that´s working ASAP.
Then you can work on fine tuning and testing with rather minor changes.

---

I'm testing 4 landings and 2 offers. Do you think i should remove one of the landers?

---

It depends.

When these are 4 completely different landing page styles it´s perfectly fine to split test them.

But I personally wouldn´t test different variations of the same style in the beginning.

There you should try to keep your setup as simple as possible.

And different landing page styles can make a huge difference in performance.
Different variations of the same style mostly only make minor difference.

So it´s more important to find a working style instead of testing variations of a style where you don´t even know if it´s good or not.

---

Thanks! I'm testing 4 completely different landing page styles. I'll update this thread as soon as I have some updates

---

This looks very promising! I'm following Vortex's $1 guide but I'm stuck because it's been over a week since I signed up for Mobipium and still haven't heard from them... Have you tried that guide I'm talking about?

---

I dindn't try that guide. I'm following the 40 days tutorial you can find here: https://stmforum.com/forum/showthrea...ntro-and-Index

---

Following my stats after 4 hours. Am I doing something wrong? The traffic looks slow but I'm bidding more than average on a geo+carrier with a lot of traffic available



Edit:

I collected some more data and I have an offer with 2 conversions and a CPL <10*payout. I paused the second offer



I also paused the landers 3 and 4. The first 2 are still running both because I don't have a clear winner. As soon as I will have 500 visits/each I'll select the best one and pause the other






Edit 2:
I reached 500 visits per lander. I decided to pause the lander #1 because of lower CTR. QUESTION: is it the right approach? Higher CTR means more people seeing the offer



I unpaused all the offers I tested at the beginning. Now I have a succesful lander and it's time to check if the previously paused offers are good.

I manually updated the campaign cost and checked if there is any zone to cut. I didn't find any zone with a cost > 2*offerpayout. I'm collecting more data and check the stats again in 1 hour

---

18 hours ago I launched a second campaign. After 18 hours the results are the following. I'm stopping the campaign because I don't see potential. Not enough traffic to justify investing time in it. QUESTION: Is it correct?






In the meantime the first campaign is still running. Both the offers are still running because there isn't a clear winner. According to the Voluum settings it should be splitted 50%-50% but the number of clicks to the first and the second campaign are very different. QUESTION: Why Voluum is sending double the traffic to the second offer if the setting is 50%-50%?



I also excluded 3 zones on propellerads that were costing twice the offer payout.

---

After 12 hours I'm checking the stats again



No new conversion
Why the offers aren't converting anymore?

Following a screenshot of the zones. Following the tutorial I think I should cut all the zones with profit < 0.66$ (the offer payout is 0.33$) but I will cut all the zones with at least 1 conversion. Is it the right approach?

---

Oh dude, so many updates already

First question, you are running only Pops for these campaigns, right?

Then 500 visits to the lander are way too low to decide anything.

Do you judge just by gut feeling or do you use something like a statistical significance calculator?
If not you should do it, especially in the beginning it can help you alot to make the right decisions.
Here is one where you can splittest several elemnts each other, in your case landing pages https://marketing.dynamicyield.com/bayesian-calculator/

You are right that higher CTR means that more people see the offer.

But that doesn´t always mean that you make more profit from it.

Example:

LP 1 = 50% CTR and offer converts at 5% -> 1000 LP visits = 500 clicks to offer = 25 conversions

LP 2 = 10% CTR and offer converts at 30% -> 1000 LP visits = 100 clicks to offer = 30 conversions

Basically correct, but in the very beginning it´s absolutely ok to run such campaigns just to learn how to run campaigns.
The process of running such small campaigns and big campaigns is bascially exactly the same.
Only difference is the volume and money you spend on it.

Such small campaigns can help alot to learn how to setup campaigns, learn how to setup tracking links properly and so on.
When the something goes wrong you only loose a few Dollars.
When something goes wrong in a very high volume campaign (maybe even in a very expensive Geo) then a small mistake can easily cost $xx-$xxx.

So as long as you still learn how to run the stuff it´s perfectly fine to use such small campaigns for learning.
But as soon as you know properly how to run campaigns you should go for bigger ones because these small campaigns won´t make you enough profit to be worth it.

What exactly is this screenshot all about?

Because as you can see one element has 3.10% CTR, the other element has 6.27% CTR.
That´s why one offer has double the visits because CTR is also double.

This happens, that´s why it´s always good to spend the campaign buget over a longer timeframe like 2 days and NOT spend it as fast as possible.
Offers can convert different from mmorning to evening, from weekday to weekend and so on.

Fluctuations in performance can happen and are normal so it´s better to have stats from a longer timeframe.
That way you can get a much better impression about the average performance.

---

I'm doing my best trying to learn quickly

Yes, 100% pops on propellerads

What is the minimum of visits before deciding on a lander?


I'm following the tutorial and I'm using this tool to decide http://www.peakconversion.com/2012/0...al-calculator/ (down now, a backup version is here https://web.archive.org/web/20190404...al-calculator/)

You are right but I have to cut landers early to avoid spending a lot of budget in testing and without a lot of conversion is hard to have the full picture clear. How do you suggest to proceed?


My goal is to have at least 10 profitable campaigns before moving to a bigget GEO

In the screenshots you can see the two offers. The users are supposed to see every offer 50% of the time but the clicks of the first offer are 107 and the clicks of the second offer are 212. Aren't they supposed to be the same amount of clicks if the campaign is set to split the offers 50%?


Thank you so much for all the feedbacks and help!

---

After 6 hours I checked the stats again.
I decided to pause the first offer according to the following stats. The second offer is the winner


Now I have 1 winner offer, 1 winner lander, 6 zones excluded

I tried to understand if there is anything else I can focus on but I can’t find anything consistenly green to optimize. I broke down the report for zoneid, browser, OS, language but I just found some green for some user agents (following screen). The numbers are really low and I don’t think it make sense to focus on optimizing according to this information


Should I keep the campaign active?

---

Ciao @barbarollo

Here's a great thread about DOs and DONTs of Optimisation.

There are also links that point to other interesting articles on cutting placements... I'm sure you will find them useful!

---

This is hard to tell because there is no fixed number.

But on Pops it´s usually more visits than on other traffic types because you send the users directly to your lander = your landers receive more volume anyway.

For my Pop campaigns I use a rough rule of thumb.

Number of landers x number of offers x average offer payout x 10 = test budget

This is the formula for completely new campaigns, when you don´t have any data about blacklists and stuff.
It´s rather rough but especially in the beginning it can help alot when you work with such formulas so that you have one steady approach to use.

This also prevents from jumping too much or too fast around with your stuff.

When I don´t see any signs of success after 50% of the test budget I also stop the campaign and test something else.
You still can test again later with new offers (or landers).
That way I never risk to lose all of the test budget.

And you are right that it´s hard to get a clear picture without many conversions.
But cutting elements too fast is also a bad decision.
Better stick with some rules that you use for all of your campaigns.

When you don´t let the traffic run long enough or when you act just by gut feeling you risk to destroy all your test stats.
Then you also just wasted money.

Everyone should have a good system for testing that can be used again and again.
That´s the only way to get significant stats sooner or later.

When you just run stuff without a good system and exclude landers, offers or placements just by gut feeling you will have it much harder to learn about what works and what not because you won´t see patterns there because in the end than every campaign is just random.

Before you start your campaign make a quick check.
How many landers, how many offers, calculate budget and then run the tests.

Also keep in your mind that by far not all campaigns are winners.
In fact most campaigns will fail.

That´s just normal, otherwise we all would be Billionaires.
So when you calculated your test budget make another check if the campaign is worth running.

For example, when you calculate a test budget of $200 but the volume for your campaign is very low and offer payout is very low as well then it would take ages to get the test budget back AND also make profit then.
In such case it´s just not worth running the campaign.

Do you run the 2 offers on the same lander or on different landers?
Because as you can see, one of the offers has twice the LP CTR than the other offer.

So it rather looks like you run the offers on different landers.

Anyway, can you post a screenshot of the campaign settings?

Because only with the screens provided it´s impossible to say what the issue is.

You decided by your gut feeling again.
I just checked the performance in a statistical significance calculator.



You should have at least 90% or even 95% confidence that one offer is a clear winner.
Otherwise it didn´t reach statistical significance yet.

I know that such stuff can be really annoying.
But when you are new to these things you really should rely on such tools because you just don´t have enough experience yet to judge by your own estimates.

And this will cost you more money in the longrun.

In my Pop campaigns I never check or optimize by browsers, user agents, device brands, OS versions or such stuff.
When I have to optimize these things to make the campaign profitable I stop the campaign.

Because then you have to do micro optimization and this is really time consuming and in most cases not worth it.

A good campaign will be profitable without it.

---

You can try hitting up Flor on skype:

https://stmforum.com/forum/showthrea...pium-joins-stm

Sorry - didn't mean to hijack this thread! I have nothing to add, as @twinaxe has said everything I wanted to say while reading all the updates! Keep up the great progress @barbarollo!




Amy

---

After the end-of-the-year holidays break, I am back! Following the suggestions in another thread I will focus just on this follow-along instead of spending time on the 1$ tutorial too.

MY CURRENT SITUATION
Spent: 601.64$
Revenue: 37.16$
Total: 1435.52 $ left
Time Left: 58 days

I’m trying to do the bot testing as suggested in the tutorial. In my previous testing I skipped it but it’s time to learn it now!

- First discovery: If you have a javascript warning in the landing page, you have to add the script code for the bot testing BEFORE the script code of the warning or it will not trigger properly
- Second discovery: I do not know how to properly access to the data in Voluum to read the results of this testing. I opened a request for help on this topic here: https://stmforum.com/forum/showthrea...l=1#post386948

EDIT: I just discovered that I can't use the bot testing method with the Voluum Discover plan and I have to upgrade to the Profit plan (more than double the price). Do you think it is worth it?

---

Not sure if Custom Conversions are available in the Discovery plan. Maybe someone from @Voluum can shed some light on it.
If you have custom conversions available, you may give a shot to this method, has worked wonders for me
https://stmforum.com/forum/showthrea...ffic-in-voluum

Also, I wouldn't suggest focusing too much on bot detection rather then the campaign itself. I've made this mistake myself in the past and that's one of the things I regret doing. Try focusing on the following:
- Publisher level performance
If a publisher doesn't show any promising sign, just kill it. You have plenty of publishers to get traffic from.

- Offer level performance
If you have an offer that sucks on its conversion rate, test it against another one. If the other one is better, ditch the first.

- Landing page performance
Landing page click-through rate is quite an important metric. Try finding some stable long-running landing pages on Adplexity and get it to a stable CTR. If the CTR backs up to the offer conversion rate, then you're on a good path. In terms of performance I see you already have a good page loading time.

---

Sorry for confusion, let me go a bit more into detail here

You can use the formula as a rough rule of thumb for each step in the testing progress.
Just calculate it again for each new test phase.

So the process would be:

1. Stage: Finding an offer that is good enough to test landers

4 landers x 4 offers x $0.30 average offer paout x 10 = $48 test budget

BUT as soon as an offer has 2-3 conversions stop the test and use this offer to test for the best lander.

When you have 2-3 conversions on an offer after only $2 adspend then you don´t need to continue running because you already have an offer to use for testing landers.

The test budget is only the maximum to spend.

Also when you don´t see any chance to succeed after 50% of the test budget you can cancel the test and don´t need to lose more money on it.

That way the most you can lose is 50% of your test budget, you never risk to lose all money.

2. Find the best lander

Use the previous converting offer and test it with the landers.

4 landers x 1 offer x $0.30 payout x 10 = $12

The $12 are the maximum to spend in that test.

This would be $3 for each lander, when a lander has no conversions with the good offer after 50% of the budget take it from rotation and continue with the better landers.

That prevents you again from losing all money.

3. Use the best lander to test all offers

After the lander test you should have a winner, use it now to test all offers again.

1 winning lander x 4 offers x $0.30 offer payout x 10 = $12 test budget

Run the test with the winning lander and the 4 offers you have.

This time it would be $3 budget per offer.

When an offer has no conversions after 50% of the allocated budget take the offer from rotation.

After this test you hopefully have a good working combination.

When you don´t have a clear winner amongst the offers but the offers are converting run a new test round to try to find a winner.

When you have several offers that are converting more or less same just let them run together when they are profitable.

Sooner or later you will have a winner there as well.

I hope it´s clearer now, if not just ask

---

I need to print this out and tape this to my monitor

These are fantastic guidelines for testing!

---

Wow - so many updates in the last few days! Loving all the discussions!

I know of affiliates that will keep starting new campaigns just to take advantage of that initial burst of high-quality traffic.

And this can be done not just with CPA campaigns, but also with CPM campaigns on some networks. (Basically anywhere where you get that initial burst of good traffic and then a decline in conversion rate.)

Not every traffic source will allow you to do that, but if you can find some that do, AND wouldn't mind the trouble of having to duplicate campaigns all the time, then it may be worth a test.


I actually have a different opinion than @twinaxe regarding statistical significance. But of course no two people will do things the same way and I think it's great.

I believe in statistics, and personally I would have cut those landers like you did @barbarollo.

Of course, a couple of conversions either way CAN very well turn the tables, there will always be different possibilities for different outcomes.

But this is exactly why we use statistics: The math tells us the LIKELIHOOD that a lander will end up winning.

And in this specific situation, the math is telling us that it would be highly unlikely for those 2 landers to win out.

It's not a huge deal though. You can choose to run them for longer and cut landers at a higher level of confidence, or just cut them to avoid spending more money. It's the old accuracy vs. efficiency dilemma we're constantly having to deal with when analyzing stats.


The way I see it is that the $48 is like the MAXIMUM amount of money anyone should spend on testing the $0.30 payout offer.

And the testing should only continue if you see some type of promise. For example, if you're confident about the quality of your landers, and the offer doesn't even convert once during the first 10x payout in spend, I wouldn't even bother to keep running.

I'd rather spend the remaining $45 on testing another offer.

Great progress and amazing discussions everyone!!



Amy

---

Haha yes, 5 people = 5 different approaches.

But maybe there is a small misunderstanding.

When I said I would keep them running for a bit it was related to this screenshot where they only spent about $1 and where every next conversion could flip the coin.



And I could imagine that you refer to this screenshot, the last one from the SS calculator.

If you refer to that screenshot then I absolutely agree.

There I also would cut the two bad landers and continue with the better ones.



Exactly, that´s why I better explained it more in detail above.

The $48 are only the maximum for this test round.

But when you find a good offer to test landers after only $2 then stop the test and prepare the next round, then of course you don´t need to wait for the $48 to finish.

---

Thank you so much for all the feedbacks! After reading all the comments I'm starting the campaign from scratch trying to include all your suggestions. Let's do it properly this time!

1 geo, android only, propellerads, smartcpm, exclude android webview, 1 carrier, 2 landing, 4 offers. Total test budget 2x4x0.3x10=24$


I have a doubt: at the beginning of my journey I decided to fix just 4 landing pages, I identified 6 offers (sweeps, SOI) from different AN and 1 tier 3 geo and I'm trying my best to have a almost positive ROI (-30% max consistently) in a campaign. Is it a winning approach? I'm trying to avoid jumping from a geo to another, from a landing to another and from a vertical to another without mastering it first

---

That´s correct, but remember that you don´t need to run through he whole $24.

As soon as you have an offer that converts 2-3 times stop the test and use that offer to test landers, no matter how much you spent already.

And when you have no conversions after 50% of the test budget also stop the test, then it´s not worth it.

It´s good that you stay at one vertical and don´t jump around.

Also sweeps are a good vertical to work with.

But when you feel safe enough you should consider to run offers with a bit higher payouts (probably then in other geos).

The thing with very low payouts is that the offers need to convert very good to make it worth it.
The lower the payout the lower the maximum possible profit.

That means when you run $0.30 payout offers then the possible profit is from $0.00-$0.30

Now let´s say you make $0.10 profit per conversion, then you would need 1k conversions for $100/day.

When you run an offer with let´s say $1.20 payout the possible profit is from $0.00-$1.20, it´s a much wider range.

When you run the campaign at same performance as the example above you would make $0.40 profit per conversion.

In that case you would need only 250 conversions to make $100/day.

Of course these are only example numbers but I hope that helps to understand why it´s often better to run a bit higher payoits.

Low payouts are only really worth it when you have huge volume and when the offer is converting very good.

---

Thanks so much! Your suggestions are helping me a lot.

I have the feeling I'm missing something important but I can't really understand what it is.

Today I launched 5 campaigns, 4 of them on 1 LATAM Geo, the fifth one in a Tier 1 Country following the @twinaxe advice to try higher payout offers.
I don't know if I'm picking the wrong GEO, the wrong carrier, the wrong segment, bidding too low or just using traffic sources without enough traffic but I always need days before collecting enough data to cut landers/offers and to optimize and as soon as I start the optimization process the offers stop converting.

At the beginning of my follow-along I set 90 days as maximum time to see some profitable campaign and the last 30 days are coming soon. I need to quickly accelerate the learning process and I'm ready to invest more money in it.
Do you know any course or any coach that can help me to speed up the learning with some personalized training session?

---

Yesterday I launched some campaigns. I needed to collect more data and to spend some money testing.
Today the results are not the best

Last 24 hours global stats


Last 24 hours campaigns stats


Offer stats of the only campaign with 1 conversion


A campaign with a setting similar to the campaign with 1 conversion was doing way better last week. I really can't understand why it isn't converting anymore. Same landings, same offers, same traffic source and segment

EDIT: I paused all the campaigns and I'm checking the Voluum configuration. I have the feeling that I'm sending the users to the wrong offers

---

It can have several reasons that campaigns suddenly stop converting (apart from the offer).

One thing about my campaigns is that I never change the targeting in a running campaign, I also never change the bid in a running campaign.

The only thing that I change in running campaigns is blacklisting placements.

When I want to test another bid I create a new campaign for it, when I want to test different targeting I create a new campaign for it.

I had it some times myself that I had very good converting campaigns and as soon as I increased the bid for example to get more volume the performance dropped alot.

Try to see it from a positive perspective as well.

Do you know the saying "Every failure is one step closer to success"?

You learned a lot already and you made some mistakes already that you don´t need to do again.

With every mistake or every failed test you know one more thing that doesn´t work for you.

So keep track of what you are doing, analyze and learn from it so that you don´t make the same mistakes again.

That way you eliminate your obstacles one by one and sooner or later there is basically no way around to succeed.

Putting more money in it can also help to strengthen the mindset, the more you have to lose the more serious you are with your stuff.

But please take care to not get into panic mode and waste the money for random stuff.

You already make progress so let´s use the next 30 days to get you to the next level.

Well, I did pops coaching for some time but now I better spill my knowledge for everyone here

But one thing, just in case you find someone for 1on1 coaching please make sure that you can trust the person and also that it´s too overprized.

One of the reasons why I started with the coaching was because I wanted to give the people the chance to get honest and good 1on1 coaching without getting scammed or paying too much for too less content.

Unluckily there are quite some honorless people out there who try to exploit the situation of beginners who are looking for help but in the end don´t receive something valuable back for their money.

Also a 1on1 coaching is no guarantee to succeed, I would recommend to stay here on STM and we work on your FA together with you.

That way you also get very good support and the money you would pay for the coaching can better be invested in traffic (what you would have to pay for anyway).

Hard to say what happened, unluckily there is no general rule when things like this happen.

But for example, when you are running SOI sweeps then it can happen that the offers database just fills up so it´s harder to convert the longer the offer is active.

That´s why offers often convert best when they are fresh with empty databases.

It also happens that an offer is converting very good, then it gets paused and when it´s back a week or so later it suddenly has higher payout but doesn´t convert that good anymore.

That can happen when advertisers also optimize their offers.
Sometimes they start an offer without previous stats or so and just keep it running for some time to collect data about what is making them the most money.

Then they pause it, analyze the leads and optimize the offers for their own needs.

Often then they can afford higher payouts because they have a more lucrative offer funnel then.

But when they resume the offer and it doesn´t match that good with your traffic anymore it can happen that exactly the same stuff (traffic, landers, offer and so on) that was doing very good before is then not converting anymore.

---

Thanks for the help! Following a quick update

During the weekend I double checked all the Voluum configuration and discovered that a couple of campaigns were misconfigured (wrong tracking code). I fixed everything and I launched new campaigns. Lesson learned: always double check Voluum configurations and settings

Analyzing the data I decided to apply for the 14 days free trial of the anti-fraud voluum kit to understand if I have a bot traffic problem. I implemented it and tomorrow I will check the first results of the analysis. I want to have the peace of mind of knowing my campaigns are not good because of my settings and not because of bot traffic. As soon as I discovered I can do it for free with the Voluum trial offer I applied.

Talking with a couple of AM over skype I discovered that the geo I was targeting (Mexico) is really hard to convert. One of the network also suspended one of the offers I was using because it was not profitable enough for them.

I decided to capitalize on the landing pages I already have and I moved to another Spanish speaking Country with a better traffic quality: Spain.

I launched a campaign but I think my bid was too low (1$ CPM) and after 24 hours I didn't receive enough traffic or conversion to optimize anything.

Today I decided to invest more and I launched another campaign using the recommended CPM in propellerads (about 4$) for that geo and segment. The offer is sweep, SOI, mobile only, one carrier smartcpm. Let's see if paying more for the traffic will pay in terms of quantity and quality. I doubt I will see any ROI because probably most of the affiliates are buying that traffic to run more lucrative offers (PIN, CC submits, ...) but I want to give it a try.

Today I also tested a couple of campaigns using the monetizer smartlink and the push subscribers offer. I just wanted to understand how does it works and what is the potential of it but I doubt I will see any ROI. Probably I'll implement it in the backbutton for my landing pages.

My main problem at the moment is traffic buying and having enough traffic to start optimizing the campaigns without waiting weeks. Let's see if spending more on CPM is the right next step.

---

No conversions so far. I opened my campaign in propelleads and now the recommended CPM is 1.84. Yesterday it was 4.3$



Why the CPM is less than half now?

Following the stats for that campaign from propellerads. My actual CPM is about 1.63$


Following the last 24 hours stats from Voluum


I'll collect some more data today.

---

Really dude, don´t focus toom much on hunting bots or finding out if bots are the reason for no conversions.

You are running on Propeller Ads, it´s a good trafficsource where many others (me included) also buy traffic and run campaigns.

So it´s probably no bot problem if you succeed there or not, it´s rather that you didn´t find the right combination of geo, lander, offer and targeting.

Bot tests can help to identify if some placements have a bot problem or not.
But usually running bot tests won´t make a failed campaign suddenly successful.

When you have a good combination of geo, lander, offert and targeting then it doesn´t matter that much if there are bot placements or not.
You will see good results anyway.

You should always spend most time with tests to find a winning combination and when you found it to milk it.

Testing for bots is all good and fine but nothing to set your focus on.
There are other elements that have much bigger impact on your campaigns.

Yes, it´s not the best geo.

I know it´s tempting because there is huge volume in MX and cheap bids.
But all that doesn´t help when there are not enough good offers to run.

One thing where you can also check if a geo is good to run or not is to check how many offers are avilable in that geo.

Usually you can say when there are many offers are available then advertisers make money there and then you can do as well

Did you blacklist placements in the
These things can cause such changes in the bid.

But bids usually change from time to time anyway depending on available volume, competition and such stuff. campaign already or did you change anythign on the targeting?

It´s normal that your real rate you pay is lower than what you set when you run on SmartCPM.

---

Hi @barbarollo

Very Nice FA. Wish you best of luck in your journey.

Could you please explain me about "excluding Android WebView"?

What's that? How can I exclude from my campaign and what's the benefit of that?

Thanks.

---

Thanks! I'm not focusing on bot testing too much but I hate to have doubts in the background of my mind. Spent 10 minutes on it: done

I didn't blacklist any zone but the CPM dropped by itself.

---

Android Webview is a browser embedded in an app.

See more here https://developer.android.com/refere...webkit/WebView
WebView objects allow you to display web content as part of your activity layout, but lack some of the features of fully-developed browsers. A WebView is useful when you need increased control over the UI and advanced configuration options that will allow you to embed web pages in a specially-designed environment for your app.

I'm excluding it after reading a post here on stm. Those in-app browsers are not standard, I don't know exactly how they display the landing pages and the offers and I don't want to risk anything targeting those users.

---

Thanks for explanation.

Do I need to set that in my lander?

Because I haven't found that option on my propeller traffic source.

Could you please share me that STM post if you don't mind or way to exactly how to exclude that?

Thanks.

---

I can't find the STM post right now. You can set it in propellerads. I don't think it impact much but I don't see any side effect in setting it.

---

Total test budget 4x4x1x10=160$ (4 landing x 4 offers x 1$ average offers payout x 10)

Last 24 hours campaign stats.


Offers breakdown


Paused all the offers with less than 2 conversions. I found a good enough offer to start cutting down the landers. I will re-test all the other offers (and much more) later.
I'm fighting my fear of spending money on traffic. I'll let the campaign run for another day. I increased a lot the maximum daily spending for that campaign. I have to fight my fear of having the campaign out of control!

---

Did you get any insights from your AM about these offers? Carried away from everyday stuff, we usually tend to forget about the feedback we can get from AMs about specific offers.
One of your offers is showing 4.88% conversion rate. Can you check with your AM whats the conversion rate of this offer on the network level? How much traffic is the network getting for this offer and if possible from what channels.


Usually the information you get from your AMs is very important when it comes to evaluating an offer. Without that, you may just throw money hoping something that doesn't stand a chance will convert and bring you some profits.
PM me your email so that I can set you up with an account at TheOptimizer Mobile. This way you'll keep things under control without having to worry much

---

Hey Barbarollo! Awesome follow along!! Just wanted to share some insight with you about the bot traffic since you mentioned it a few posts back and I think I see those numbers in your screenshots above. I am seeing around 25% - 40% in my Propeller POP campaigns (SOI sweeps) in the ZA GEO. I've typically been seeing the higher end of that range at the lower bid's and so far it has seemed to stabilize as my bids have gotten higher.

Sounds like we're both about the same spot in our journey - a few months in and anxious to make some money off of the time and money that we've been putting in to learn. I mention this just to let you know that I'm not an expert, so my numbers could be way off, but at least it gives you another reference point.

The place that I have found the bot numbers to be useful is in cutting zones... I'll let my campaigns run for a day or so to collect data and sometimes it is surprising to see that some zones are crazy infested, like close to 100% bots. And in those zones I never see clickthrus or conversions, so I use it as an indicator which helps me cut those zones with a bit more confidence as sometimes they have huge traffic numbers until you see that its all junk.

All that said, I do agree with twinaxe's advice though to just keep testing more offers and landers. I've been trying to systematically test everything (like bots) to see what changes to a campaign makes the biggest difference, and I'm slowly seeing (I'm a bit stubborn) that just throwing more offers and landers out there makes the biggest difference for me. I'm generally now starting to look for major differences in campaign stats - like 2% CTR vs 5% CTR instead of small things that I thought I could incrementally add up, like doing this one thing might get me 0.5% increase in my CTR. I still that that line of thinking is good, but would better serve me on the back end after I've done all of the big things first.

Anyway hope that info helps! Send me a PM if you want to bounce ideas back and forth some time.

---

Unluckily there is no way around in paid traffic - we have to spend money to make money

Important is that you spend money for important stuff and that you don´t waste your money.

You are not alone with that feeling.

I think everyone can relate to it, especially when you have your first big(ger) campaign running that is spending quite some money and then you keep it running over night when you can´t check or control it.

The best way to run AM is to be as unemotional as possible.

In the end campaigns are all just numbers on your screen.

The worst thing you can do is to get emotional attached to your campaigns, good ones as well as bad ones.

Man, that´s a great move.

Thanks for being so generous.

I also use Optimizer myself and I run lots and lots and lots of campaigns and wouldn´t be able to run my stuff that way without Optimizer.

I really need more time, then I could also write a bit about my personal way to use Optimizer.

That´s basically also how I do it.

I don´t run specific bot tests but when a campaign is running for some time already you mostly get a good feeling for the average performance a placement should have.
When I the see placements that are way off from it like very high or very low CTR and low performance I also block them even when they didn´t reach enough spend already.

But to do so you need enough conversions already to check for such patterns.

Great idea, working together can help alot, epsecially when both are at (more or less) the same level.

---

It's been over six months since the last update. I wonder how things ended up? Some really good information from @twinaxe and @vortex and others though, so glad I read through everything.

---

He was last active on March 28th. Maybe he reached @stickupkid step 9
https://stmforum.com/forum/showthrea...mpquot-summary

I doubt we will ever know... Hope I am wrong.

---

Yup, good thread.

Old but gold.

---

Thanks, haha makes me laugh to read back again! Let's hope he disappeared with a pot of gold and living the life he wanted!

---

It is touching to see your selfless dedication. You are a warm-hearted good man

---

Did you reply to the questions that MOBIPIUM sends to your email when you sign up? If you did and your account is still on hold please ping me on Skype and I'll help (live:flor.savasi)

---

Home > > Newbie Follow-Alongs