Home > Push Traffic >

---

The beginning of a new career (55)

---

I signed up less than a month ago, and I feel like I'm a different type of marketer. The knowledge I've learned from all of you guys is unbelievable. I've signed up to other forums (you know which they are), and they don't compare to STM. My biggest respect to everyone who makes this forum so good—Twinaxe, Vortex, Caurmen (RIP), and so many others.

Thanks to all the knowledge I've earned, I want to give back and share with all of you the beginning of a new career as a CPA marketer. I've been planning for almost a month my vertical, geos, and sources; now, it's time to get started.

Since I have a history of abandoning everything and not following through, I've developed a set of goals I'd like to achieve in the next six months. Here they are:

1. Make $1 profit in a day
2. Make $30 profit in a month (avg. $1/day profit)
3. Make $5 profit in a day
4. Make $150 profit in a month (avg. $5/day profit)
5. Make $10 profit in a day
6. Make $300 profit in a month (avg. $10/day profit)
7. Make $20 profit in a day
8. Make $600 profit in a month (avg. $20/day profit)
9. Make $50 profit in a day
10. Make $1500 profit in a month (avg. $50/day profit)

End Date: 31st March, 2021

I know this may seem like a very ambitious six-months plan for some, but I rather aim for the stars and fall on the moon than aiming to the sky and fall on the floor.

Investment: $500 per month for the first month, double that every month. Also, every dollar of profit earned compounds to this fund.

September Goal: Make at least 10 campaigns to learn the basics.
October Goal: Make an average of 1 campaign a day for 30 days.
November Goal: Make an average of 2 campaigns a day for 30 days.

Note: I have defined these goals before I created my first campaign this week. I may have to lower this goal a bit, or better yet, change my definition of a campaign.

Budget per Campaign: # Offers * Payout * # Landers * 10. While this may be the goal, most payouts are too low to work, so I may have to spend more than 10x payout or cut some campaigns within a day.

GEOs:

• LATAM
• SEA

Offers:

• Sweeps
• Push Subs
• Surveys
• Carrier Billing (maybe, I'm still not 100% sold on this one)
• Mobile Content

Type of Ad: Push ads.

Traffic Sources:

• PropellerAds
• ZeroPark
• Pushground (maybe)

Anyway, that's all for now guys, let's get this follow-along started!

---

Good luck my man, in my experience I went from making a small loss/break even for the last 3 months to now making like $25-40 per day, of course still chump change to most and I still consider myself a newbie.

Progress is definitely not linear so don't expect it to be, else you may get unnecessarily burned out.

---

Great stuff man - love the goals! I think that's the perfect approach to focus on those incremental steps like that... that's exactly what I did when I started... $1 profit... $10 profit... $100 profit... etc

---

Thanks for the advice, Fastaj! I know progress isn't linear; it frustrates me a bit, but that's how life goes.

Yup, one step at a time. I want to take it slow and learn without making too many big mistakes.

---

Thanks for the shoutout @chilldude and I wish you the best of luck in achieving your goals!

I don't run push but I'm sure twinaxe will be stepping in soon - being the tremendously helpful and energetic person he is!

Have fun! I'll be watching and rooting from the sidelines!


Amy


Sent from my iPhone using STM Forums

---

You are more than welcome, @vortex! You have taught me so much, I'm eternally grateful. This forum make this journey much more enjoyable than it could be otherwise.

---

I really love your goals because this is how a beginner should start.

In my first Beginner Series Post I also talked about setting realistic goals.

Right now I am trying to finish the next post in the series were I tell a bit about how to achieve these goals.

I hope it will be interesting and helpful for you as well

All the best for your journey.

---

Thank you! Your kind words are very encouraging for me. I will keep an eye on your post.

---

Progress Report #1

I launched the three campaigns, and after a few hours, I followed @vortex's advice and paused the two because:

1. One campaign spent 10X payout and didn't generate a conversion
2. One campaign spent 10X payout and generate one conversion

The third campaign had generated two conversions at 10x payout, so I left that one and kept it running.

Here are the results so far:



Yes, it's not good, but not as bad for a beginner. I've heard -85% ROI is pretty common at this stage. But the main reason why I know this is worse than it could have been, is because I made a big mistake.

The Big F*#k Up
Here's where I messed up:

I was reading @twinaxe's post on traffic sources, and I read he recommends bidding lower to begin with. My problem is that I bid almost twice as high as the recommend bid because, I don't know.

So right now here's where I'd need your help:

Should I duplicate this campaign and test a much lower bid? What else would you recommend doing to fix this mistake?

Killing Landers
I know that I should wait until each lander has at least 5 conversions to start checking their statistical significance, but what do you think?



The Good News
I got my first conversion!


To-Do's

1. I will prepare and launch a similar new campaign in Singapore.
2. I will get Voluum's Profit plan to get my custom domain's SSL certificate and anti-fraud kit. Any opinions?
3. I may clone my current campaign with a lower bid, depending on your thoughts.

---

This happens, we've all made bidding mistakes, don't worry about it Once I've bid 8 bucks per click instead of 8... the budget went fast LOL

As for duplicating the campaign with a lower bid... you can lower the bid directly in the campaign you have, but since the CPC sources usually allocate a burst of "test" impressions to new campaigns, it might be better to start a new one (or clone it). This way, it might stand a better chance at reaching the desired CTRs and get more traffic.

---

Oh yes, when I started with paid traffic one of my first trafficsources was Mediatraffic.

Once I ran a campaign in US and set the bid by accident to $1 per pop.

As you can imagine the budget was spent pretty fast

Same opinion as matuloo, it´s better to create a new campaign to get the boost.

---

Thank you @matuloo and @twinaxe for your handy advice. I have created a new campaign...and no conversions whatsoever. Today was the worst day of all.



The high-CPC campaign was so bad that PropellerAds stopped it for me. When a company that makes money from fools like me pities you, you know you're doing something wrong. 😂

Since I have spent almost 30x payout and got nowhere, I've decided to stop these campaigns. They were pointless. I don't know what I did wrong, but getting so many clicks and NO conversions shows I did something very badly; either the lander was off, or the offer wasn't good. Whatever the case, I'm on to the next campaign.

Some learnings:

• Test landers before choosing them.
• Create one lander with its offers instead of one offer with all its landers. It's faster and more motivating than the other way around.
• If one lander keeps crashing, move on to the next one. I re-learned this one today; one lander took me like an hour to fix, and yet it continued to crash. I moved to a much better one and got it done in less than half an hour. That includes adapting it to five different offers. It's the path of least resistane, yo.
• Test offers faster. I already have great landers for some offers I have in my backlog. Creating a campaign for them would take me a few hours. Why don't I start there? I already realized this when I had started with a separate campaign. But this is one idea to take into consideration for next week.
• Create an inventory of my landers. I want to systematize my lander creation, so if I have an inventory with all the unique aspects of them (e.g., "the copy is generated through JS not HTML"), I can create them faster.
• I will also add the landers I want to try for a new campaign to my AWS and test them to see how they work. Sometimes, landers seem to work great locally, but not really.
• USE THE CONSOLE. I have found that using the console allows me to find problems with landers much faster than if I try to guess. This is obvious for you, but not for me. 😅

Anyway, as I said earlier, I'm off to a new campaign. It includes 5 similar offers for Singapore. Today, I translated, tested, and created 15 landers—5 landers for 3 offers. I still have 2 more landers to create (10 in total). These remaining ones use a wheel, so I need to guess how to make them work; I still don't figure how to make the spin stop in a different offer. 🤔

In either case, tomorrow I will be done with this, and I will launch 5 new campaigns. This time, I will use a lower bid.

Questions:

• Affiliate networks send their top offers every few days. Today I got one from a network whose offers seemed really interesting. Are these offer suggestions any good? If so, next time I will be prepared to launch a blitzreig of campaigns as soon as I get these emails; the faster I get them, the less populted they may be, right? (If I'm wrong, please let me know!)

---

Open the wheel image in photoshop (or whatever) and rotate it by 90/180/270 degrees to the offer you need to land on

You don't always have to mess with the code to get what you want.

---

Well, your whole post is about landers but it´s much more important to focus on testing offers.

Just grab few different LP styles for your vertical, clean them, change links, make sure that they work and test them.

Then you use the best style then to test offers.

Don´t waste too much time on building a landing page inventory yet or trying to create the perfect lander.

On the spinwheel image just change the prize on the field where the wheel stops.

Then you can use it for all kinds of different prizes.

A top offer report usually is a report that shows what offers are making most revenue in the CPA network.

When an offer is at the top of the revenue it already received lots of traffic so it´s mostly not the freshest offers in the top offer report.

---

Thank you @jaybot for your advice. You are right, fixing the wheel was that easy. I have a tendency of overcomplicating things, so it's great I have you guys.

@twinaxe, I agree with your comments. The point of creating an inventory of the landers I have edited is to write the elements I need to customize so I can edit the landers faster in the future. Otherwise, I'd have to guess how each page works from scratch, and that takes me a lot of time. It's just an idea I have to help systematize my campaign creation process.

September 23rd: Rookie Mistake #2
Today I finished editing the last two landers and uploading them to AWS...and then I realized I may have screwed up the campaign once again.

It turns out, the 5 offers I want to promote have an age restriction: they only accept people between 25 to 60 years old. And my landers don't have any type of age verification. I thought I could target people by age in PropellerAds, but then I remembered that's Facebook lol.

What do I do? Do I have to add some type of age verification in the lander? How do you do it? I have never seen that type of verification in sweepstakes' landers before, so I don't know how to fix this.

---

Almost all offers are 20+

You can’t really target on pops/push. But.

You can use a quiz lander and add the question in there (are you over 20?). Or stick it near your CTA on a spinner/gift box.

Or

Be creative with your push creative “Millennial’s over 20 are using this weird trick to get free groceries!”

Or

Ignore it and see if the advertiser gives a shit or not.

---

The thing is that in worst case you waste lots of time when you edit and prepare landers that you then never use.

You wouldn´t be the first one where that happens

It´s definitely good to download landers for different verticals in different languages so that you already have them when you need them.

But I wouldn´t work on them when I don´t need them for campaigns.

When you want more practise in cleaning and editing landers just run more tests

---

This is a normal start for a beginner. Hope you’ll make good profit very soon. Before launching next campaigns, we recommend that you read this article, as well as join our Telegram chat. There we share useful tips and even bonuses and we’ll be glad to answer your questions

---

@twinaxe: Oh, I agree. I won't do that for every lander, only for the ones I've edited and used. So right now, that'd be around 15-20 landers, all the other ones I have in my backlog I will ignore for the time being.

September 24th: 2nd Campaign Launch and Change in Traffic Source
I have finished editing all the landers for my new campaign and launched it in PropellerAds. Once again, the inconsistency in the approval kill me. One campaign gets accepted, three more don't. I asked why I got rejected, and I'm told it's because my lander says, "You may win a gift card from BRAND 1, BRAND 2, and more".

Am I the only one who thinks this is nonsense? One thing is to get rejected because I'm showing Amazon's logo, it's another to mention the name of the brand of the sweepstake. How will the user participate in the sweepstake if they don't know what they will win? The support guy told me "that offer won't be approved if changes are not done" without specifying what I had to change. An awful experience.

One thing I was surpirsed was @twinaxe's response to another post of mine where he told me he never changes his logos and he uses "landers that look somewhat like big and trusted websites." If he does that and he gets approved in PA, then I don't know what I'm doing wrong. Either way, given my tendency to overcomplicate things, I will start using other sources like Pushground and Richads. I have to test more and not spend so much time with these issues.

Tomorrow I will launch my campaign in Pushground and get ready to launch my third campaign on the weekend in Richads. More tests, less drama.

---

Not really correct, I said that I never made fake logos in my whole life.

I rather go with a lander with no logo but with the look of a big and trusted website instead of using fake logos of fantasy brands or random design.

---

September 28th: Little Progress

I've spent the last few days trying to get my five campaigns approved in PropellerAds—I had to remove 2 landers because they didn't meet their guidelines. In the future, I need to be more careful about this, so I don't get so bogged down in such a dumb mistake. So far, I removed one offer due to bad conversions, and left four more. Here's the statistics for the past few days:



This raises a problem I'm not sure how to solve: the four campaigns I kept running in PropellerAds are within the 1 conversion-10x payout rule, but are still not generating enough conversions for me to cut any landers and/or placements based on statistical significance. Should I wait until each lander has at least five conversions, or should I start doing some cutting? FYI, this campaign has a lower-than-average bid, so this may be a reason for the low conversions.

I have also launched the same four campaigns in Pushground, and so far, the results aren't promising at all:



They amost hit the 10x payout threshold, but they aren't just there yet. I will wait one day more, but I think this test wasn't worth it. What are your thoughts?

New Campaigns
I have also prepared 5 new campaigns to launch in Thailand, which I will launch directly in Pushground (the landers are more aggresive, so I know PropellerAds will not approve them).

I have applied to a lot of new offers in Lemonads, Rainmaker, and Zeydoo, some of which are identical, so I'm hoping to test these offers soon! I just need to get more efficient in my campaign creation process, especially within the lander creation and campaign testing phase.

Anyhow, things aren't working, but I won't quit! This is just the beginning, and I'm learning a lot thanks to you guys

---

Looking forward to your TH campaigns, this geo was so far bleeding the most money out of me. When you find winning lander you could try a Fiverr translation, This will boost it even further and it’s about 6€ (tax incl.) for 400 words.

Good luck and I hope u crush it

P.S what vertical are you working on?


Sent from my iPhone using STM Forums mobile app

---

How many landers do you test in these campaigns?

I also wouldn´t cut placements based on statistical signifance.

You use statistical significance calculators to find the lander or offer that is best from the ones you test so that you can cut the other ones.

For placements however you don´t try to find the best placement so that you can stop the others.

Find the best lander and offer combination first and then cut placements by profitability but not in relation to the other placements.

It´s better to first get campaigns running profitable on one source before you move them to another source.

You don´t want to spend money on losing campaigns.

For sources like Pushground it can also help alot when you ask for a supply ID whitelist and run BL campaigns on these.

About the campaigns, are these all different campaugns or the same with just slightly different setup?

Because it seems that offer payout is same for all campaigns.

Also check/post placements in your stats and not only the overall performance.

---

@plutus Thanks for the encouraging words. I'm working on sweeps, that's all I'll say.
@twinaxe I'm not looking to reach statistical confidence for my placements; I'm trying to reach it for my landers. The problem is, the campaign is bleeding money, and the landers are not generating 5 conversions each; I'm not even getting 5 conversions for the whole campaign, spending +10x payout.

I've created a blacklist for the only campaign I'm still running with all the data I took from my original campaign to a new, higher-bid one. I know I've not found the ideal offer+lander yet, at least not the lander, but I want to test and see it for myself. There's a 90% chance it will fail, but let's see.

Considering I've spent +10x payout and didn't get more than 5 conversions total, would you recommend cutting it?

Also, I've tested Pushground just to see how it goes and to test new landers I couldn't test on PropellerAds. I've made big mistakes I will explain soon, so I don't think it was in vain.

---

September Report
September was my first month doing CPA marketing. In my first month here, I learned as much as possible, taking notes of dozens of articles, absorbing everything that could help me start with the right foot. Around 45 days after signing up here, I applied everything I knew (or thought I knew), and while the results have not been promising, the learnings have taught me more than one year of full reading and studying.


Here's what I did:

• I opened an account with dozens of traffic sources, affiliate networks, trackers, and spy tools, among others.
• Set up my AWS account—I also spent way too much time and money with lander tools that were nothing but a disappointment.
• Picked my vertical, geos, and offers.
• Created 14 campaigns, 50 landers, and 70 ads.

Most of my mistakes—and learnings—have come from my inefficency around setting up landers and campaigns. I think most of these mistakes won't be reocurring so much now, but boy have they been painful.

Now I need to have my first profitable offer (I think I may have a winning one in my hands ), optimizing them properly, and learning how to scale them.

For October, I'm planning on:

• Launching 1 campaign per day—3 landers to start, 3 traffic sources—PropellerAds, Pushground, and maybe, RichAds—and 5 ads per campaign.
• Find better landers with Anstrex (AdPlexity wasn't too good for my needs, I barely found any sweeps landers there).
• Get my first profitable day.

Let's get October started!

---

Your first month looks so similar to mine I see that you are also doing sweeps with Propeller and are on the same stage in AM as myself, I left you PM with my skype id and I would be glad if we could talk and learn together

---

Alright, got it.

I thought you were talking about placements as well because you posted "landers and/or placements".

Then maybe the campaign is just not worth it.

How did you create the blacklist?

When you are still testing landers and/or offers you can´t really build a BL based on the placements quality.

You can only build a real blacklist when you have a good working funnel, otherwise you can´t be 100% sure if it´s the placement or offer/lander that isn´t converting.

Hard to tell because I don´t work with a 10x payout rule.

When you work with 10x payout you work with a static value but campaigns are dynamic.

As an example, when you have a campaign with only 1 offer and 1 landing page then each offer and each landingpage would receive 10x offer payout.

But now let´s imagine you have a campaign with 4 landing pages and 5 offers.

Then each LP woulod receive only traffic for 2.5x payout and each offer for only 2x payout.

This is too low when you have a new campaign in test mode, especially when you don´t have a good BL/WL yet.

Go go go

---

Had a quick peak at your posts, seems like You are doing too many things.

Pick one or two traffic sources, since you are doing PUSH -- Propeller & ZP is the best choices.

On PUSH generally, male is the target audience. Dating works great ie CD, TopOffers; finance offers from M500/Supremedia; and ecom stuff thats broad interest from DFO/Verve.. for example a winner on PUSH was the Portable Air Conditioner in the summer, it was big money on PUSH.

Next, you need campaign structure. Always separate desktop & smart phones, within smart phones always separate iPhone and Android. For some reason iPhone does almost as good as desktop for us, and android does 1/3rd worse. No idea why, maybe because iPhones cost more now that AT&T isn't giving "MY IPHONEEEEE" away for $50... and they are now giving away android phones.

Another actual tip that works, always target the latest 2 OS versions...

Your campaign should be testing images first on PUSH because thats what catches attention, for us warning sign, or inbox works best, but if its finance offers from M500/Supremedia then stacks of cash is great too..

Use as aggressive to the point headlines as possible...

Test 3-4 landing pages...monitor landing page CTR in your tracker.. RedTrack or Voluum or whatever u use.. the higher the LP CTR the more people get to offer... obv.. the conversions matters the most but you get the drill..

i always run 3x landing page so if payout is $10, and i have 4 landers, that means $10 *3 *4 = $120 to see if that campaign has any potential or not.

---

I don´t agree with this.

Chilldude is just starting in affiliate marketing and Crypto offers with few hundred Dollar payouts are definitely no good match for a beginner.

As we can see from the posts in this thread he started his first campaign about 2 weeks ago and still has to learn how to work with a tracker, landers, how to test, optimize, scale and all that stuff.

Long story short, the tip with Crypto offers doesn´t help at all and in worst case he just risks to lose all his money without a single conversion.

I also like Media500 but beginners should run something else, something easier.

@chilldude You already have a good idea about what to run, better stick with it

So you recommend him to run In-Page push?

Because web push is not even available for iPhone.

For iPhone you can only buy in-page push and there the volume is mostly so low that it doesn´t make sense to focus on it.

Just for comparison, here is worldwide web push volume on PropellerAds for Android



And here is worldwide in-page push on Propeller for iOS



You see, 14 Million vs 500k

And considering that he wants to run campaigns in

we can expect that the iOS volume there is even alot lower.

---

Thank you, @twinaxe for your advice. As I said in my first post, I create my budgets based on the rule: Payout * # Landers * 10.

I think it was you (or Vortex, or some of the other masters in here) who recommended using the rule. So each lander gets 10x payout everytime.

Also, @IAmAttila, thank you so much for this amazing advice. I haven't been segmeting my campaigns as you suggested, but that's something I've been thinking lately. Since it's something I didn't read from the guides I read by @vortex and @twinaxe, I thought it wasn't worth it, but I think I will do it now, at least as a test. How much budget should each campaign should get?

So, if I create one campaign for desktop, and one for mobile, how much should I invest at most? 10x payout each, or 5x?

Honestly, you guys, the advice I'm receiving is worth hundreds of dollars, I'm so humbled to have your guidance here. Thank you!!

---

@vortex I did get to track visits, but not conversions. I didn't track any in either side. One AM told me he did see conversions, but somehow they didn't get properly passed to my account. The other two ones didn't track anything. Considering that I sent them over 600 visits, and given my past (albeit limited) experience, it's surprising that I got 0 conversions. All my SOI offers got conversions, even if it was one, for that volume of clicks.

Thank you so much for offering doing this for me, Amy! I will PM you now.

For the rest of you, one AN is fixed—the test was successful—the other two haven't, so I'm still working on it.

---

Thanks for the PM - added you on skype. Let's get this sorted soon!

Hmm....if your AM said he saw conversions but they weren't recorded in your account, then it has NOTHING to do with your tracking. It's their internal system being flawed.

I would advise not to run offers with that network for a while (or ever) - at least not until they fix that issue.



Amy

---

Is this a screenshot from live visits or live clicks? If you have a landing page in your campaign funnel, Voluum's click ID will be assigned at a click - if you're checking live visits stats, the click ID won't be assigned yet. It might work in a similar way with offer stats (but I'd have to check as I'm not 100% sure if that's the reason or if there's something else wrong).


Karolina

---

Actually, from a tracker perspective, only {clickid} is essential. The other two ({trafficsource.id} and {campaign.id}) are not necessary for tracking to work, you can pass them to the affiliate network but it's completely up to you.


Karolina

---

November 16th: Overworked and Re-Energized
I'm sure some of you must have wondered what happened to me after all this time. Well, to start, I took some time off to fix the issues I had with some of the networks that weren't working properly. The good news is one network is working fine, the other one still doesn't work very well—they track my conversions, but Voluum doesn't register them for some reason—and the last one simply doesn't work. I've tried a lot of things to make it work, but conversions aren't being properly tracked in Voluum. @vortex looked at my URL and everything seems fine, so it's strange this is happening. The AM has stopped messaging me, so I guess I will have to stop working with this network. ��*♂️

Besides being a bit sad that my past campaigns haven't performed well and that I was losing money in my campaigns (due to the technical problems described previously), the main reason why I stopped working with my campaigns is that I was super busy with work. I got a lot of new clients for my marketing services, and so I spent most of my time working with them.

I took this time off to rethink some parts of what I was doing, and gain new energies to start once again with new campaigns. In these past few weeks, I've applied to four new networks which specialize in sweeps, and I've created four new campaigns. One got rejected immediately by PropellerAds—weird, as the other three campaigns were exactly the same and hadn't been rejected—two of which were duds, and one that seems to be a dud but hasn't spent all of its money yet to make a decision.



The first one is the one that got rejected. Actually, it didn't get bad numbers, so I will have to create new landers and relaunch it. The second and fourth one are the duds I referenced to previously, and the third one is the most promising one. As a note, they all target Singapore, so the audience is quite small. Is that an issue? I can imagine SG campaigns tend to be not scalable.

Also, note that the last three offers are missing conversions, as my network tracked them but Voluum didn't. Their ROI still sucks, but hey, every conversion counts, right?

This week, I will create two new campaigns, one for Portugal, and one for Brazil. In the next few weeks, I will be testing new campaigns in LATAM, a great region for me thanks to my native Spanish speaking. I promise I will keep you posted. ��

---

@chilldude have you reached out to our support to check why these conversions aren't being reported in Voluum? Ask the network to provide you with the Voluum click ids you passed to them for these conversions and include them in the email pls. There must be something causing it and we can look at the postback urls we're receiving (also, check if we've received any).


Karolina

---

Hi Karolina, thank you very much for your help. I have contacted your support, we're still figuring it out.

---

November 18th: Promising Campaign
After launching over a dozen campaigns, I've finally caught my break. My latest campaign is showing promising results, with a ROI -45%, which is amazing to me.



Originally, I tested 2 offers, one which I stopped promoting because it seemed to be a dud. Yesterday, I got a new offer accepted, so last night, I decided to add it. To my surprise, this last offer (offer #1 in the image below) has a great CR, which drove a better ROI compared to the other two.



Now, I'm running my campaign with this one offer, still waiting to spend all the remaining budget to see if I can start cutting landers. So far, I still need 5 or so conversions to run them in the statistical confidence tool. I also checked my placements, and I don't have enough data, so I'll wait.

I'm excited to finally have one promising campaign. In the meantime, I'm creating two new campaigns in another geo (same language as this one), both which have similar offers.

---

I am really a bit curious about your setup.

Don´t get me wrong, I don´t want to offend you at all but with all the postback issues on several different platforms I can only imagine the it´s somewhere at your end.

Otherwise it would be hell of a coincidence.

---

It's a bit of a coindidence, and a bit of my own fault. It's almost fixed now—one network still doesn't pass conversions correctly, even though they're being tracked in their side. But in any case, I'm focused on networks that are working properly, and I've learned a lot in the process, so it's water under the bridge for me.

---

All the networks simply have to work properly, without tracking there is no optimization and that would put them out of the business. You must be setting something up in a wrong way.

It's really pointless to spend any money until the tracking works correctly, it's like shooting a gun blindfolded

---

I understand that, and as I've said, I've fixed one of them, which I'm currently using, while I'm still trying to fix the other two, which I'm not using and I won't use for the foreseable future. For the time being, I want to focus on creating more campaigns and optimizing the ones that work.

---

Gotcha, in case you're not ok with the other networks, it's fine to move on... but it's still weird you couldn't make the tracking work with them. The thing is, unless you identify the errors you've made, the problem will repeat itself with other networks too. You're definitely making some mistake with the postback setup... or the AMs at the other 2 networks are "..."

---

You are right, but next time, I will be much more careful with the set up and I will run test conversions before starting a campaign. Also, I've learned a lot about postback setup, so I don't think I will make the same mistakes than before. Right now, I'm just excited I'm starting to see good results with a campaign, and that I've got a lot new ones coming.

---

I keep my fingers crossed that after the hard start it will become smoother now

---

Now that all your trackers' postbacks are in order, let's go!

---

November 22th: Two Campaigns Out, New Ones Coming

This last week has been one of the most interesting ones. To start, I fixed the issue with one of the networks that didn't pass my conversions into Voluum. What was the issue? The network's template in Voluum was wrong; the AN parameter was "clicklid" when it should have been "clickid". Thanks to @jeremie's help, I found the issue and discussed it with my AM. She then talked to Voluum, and they fixed that now. I just happened to be the one who stumbled on that bug. ¯\_(ツ)_/¯

That issue aside, the promising campaign I mentioned previously ended up being a dud. Still, this has been my best campaign so far, reaching a negative ROI of 60%. I launched another campaign last Thursday, but this one quickly became another dud.



Consdering that my promising campaign was from Portugal, I'm thinking on launching a few low-payout campaigns in other tier 1 countries, like Australia or Italy. I'm also considering testing low-payout offers from verticals like finance or utilities. I don't want to change my focus too much, and I don't want to spend too much money on testing, but I just want to test new waters because all my tier-2 and tier-3 campaigns so far have been big duds, with negative ROIs of 80% and above. What are your thoughts on this?

---

I’m glad that it started to click correctly for you on Voluum and that you identified core issue that was causing it.

Hardest thing is behind you, resolving another ones will be way easier from now on as you know where to look for them.

T1 geos are tempting for sure, be careful with that.

You might waste way more comparing to T3 before you get any significant data.

Volume is higher but also is competition.

Given the fact that you want to run lower PO offers, do have in back of your head total cost on given offer up until now.

You dont want to end with -150$ profit before it even starts converting on -80% ROI, and on 1$ PO offers getting all of that money back would require a lot of conversions.

On the other side spending only 20$ might not be enough at all to get any data at all.

I did RON campaign on DE and getting 0 conversions on 30$+ ad spent wasn’t much satysfying to see, trust me




Sent from my iPhone using STM Forums mobile app

---

Ah crap, these "typo" style of errors are so hard to find, glad that jeremie could help you to find it!

It's tempting to attack higher Tiers, because lets' admit it, the offer selection is wider for these GEOs and the payouts look better as well. If you really want to do it, focus on GEOs that are not the most exposed... for example when looking at EN geos, USA is really very competitive, but something like Australia, Ireland or New Zealand tends to be a bit less competitive in some verticals, because the volume is lower. Italy, Spain, Greece, Portugal... these are technically tier-1 GEOs since all of them are in the EU, but in the AM world I'd put them under Tier-2... that might be an option too.

Or do the "language approach"... for example the Spanish language is spoken in a large range of GEOs, across several tiers.... with Spanish creatives, you can reach more than 500 million people. So basically, you could work on creating a solid base of ADs and LPs in spanish and then test the many GEOs where it's natively spoken.

---

If you have done creatives/landers in portuguese, I would try Brazil.

---

November 29th: Different Campaign Settings, Different (and Interesting) Results
This last week I launched three campaigns, all of which didn't perform too well. Still, I saw some interesting results.



To start, I launched them using PropellerAds' all user activity groups. Previously, I used only the high-quality activity group. Not only did I see higher conversion rates at lower costs, my overall ROI was less negative than before—~-70% vs. ~-90%. There were times were my ROI got to -40% or so, something I've not seen previously. Quite exciting, for sure.

Since I saw more conversions than before, I got to the point where I cut all my landers and focus my campaign on the best one. My campaigns still aren't working that well, so while I'd like to keep running them, I rather spend my money on better-performing campaigns. This is something @vortex has mentioned before—focus on campaigns that work right away than optimizing for those that could work better but that require higher investments.

Also, I focused my creatives entirely on calls and messages (e.g., "You've Got 1 Message"), instead of more promotional ones (e.g., "Win X Prize Today"), and cut some zones after exceeding +2x payout with negative ROI.

Question: if I want to block a few zones, should I create a separate campaign with the blacklists or can I just add them in my current campaign and keep running it normally?

Anyway, this week I'll be launching new campaigns in English-speaking tier 1 countries with multiple offers each. I'll only use 3 landers that have worked previously. I want to focus on testing the offers; if any of them work surprisingly well, I'll test more landers. I'll keep you posted!

---

Great progress! Sounds like you have a plan!

For blacklisting, no need to create a separate campaign - just keep cutting zones for the current campaign.


Amy



Sent from my iPhone using STM Forums

---

December 6th: Two Duds, Four More Coming, and Some Questions
This past week I launched two new campaigns in AU and NZ, both of which performed horribly. I spent almost 10x payout for the 2-4 offers and 3 landers, yet at no point any of the campaigns performed well.



I just launched two new campaigns for the same geos, and they both seem to go the same route. Almost similar offer, same landers, and despite the acceptable CTR (both for the ads and the prelanders) and CR, they didn't perform well at all. Compared to my other campaigns in LATAM and Portugal, these two geos seem to be a no-go for PropellerAds.

What I saw is that PA doesn't seem to have a lot of traffic for these geos—the volume for mobile traffic is under 10k people, I think, similar to Singapore, a much smaller nation. In this case, would it be prudent to try new sources? ZeroPark and Pushground are the two I've tried before, and I'd not mind trying them again.

Also, I've been thinking about global blacklists. Would it make sense to create a BL for PropellerAds from all the data I already have? I know I've not generated more than 100 conversions so far, but I've spent over $1k with them and I'm sure I already have some data to create one. I think this may help me with my testing.

Anyway, I will keep you posted with any updates later this week.

---

Sounds like normal and necessary testing to me! Good observations and planning!

For geos with smaller traffic volumes, yes by all means start on bigger networks. If you want to save money, you can still do the initial testing on PA - may take longer for sure so pros and cons. If you want to get testing done faster then start testing on ZP directly. All depends on whether speed or money is more of a concern.

As for global blacklists: Absolutely it can be helpful! I've seen many placements perform like shit across all geos. But I would encourage you to retest those placements for each geo once you have a profitable/promising campaign, at least the biggest ones. Because there are placements that have different performance in different geos.



Amy


Sent from my iPhone using STM Forums

---

AU and NZ are really tough places.

Traffic is really expensive and especially in NZ there also isn´t much volume available.

Better test easier geos, I would even go so far to say that US can be easier to run compared to AU and NZ because there you have at least huge volume.

It depends, when these conversions are all from many different campaigns I wouldn´t do it.

To create good blacklists or whitelists you also need good campaigns.

It´s not only about how many conversions a zone generated totally, a zone has to be a solid converter to be considered for a WL or it has to fail on good converting campaigns to be considered for a BL.

---

Home > Push Traffic >