Home > Mobile > Follow-along Campaigns

---

From Professional Poker to Affiliate Marketing - Staying out of the rat race (29)

---

Welcome mate!

I've met a lot of ex-poker players that have turned STMers and become mega affiliates. There definitely seems to be the right mindset in poker needed to make things work with affiliate marketing! You should read this guys story http://stmforum.com/forum/showthread...y-life-forever

Regarding your banners, there is waaaay too much going on. You will struggle to get the CTRs you need to make your campaign profitable. If I were you I would strongly advise to invest in an ad spy tool, so you can get a better idea of what creatives are working.

This is an example from adplexity on what is working for Apus on Avazu this month.



For 90% of the ads you may be thinking WTF? Why on earth would someone download the Apus app just by clicking a Download or Play button. The name of the game is CTR here. Since you are paying CPM you want to generate as many clicks as possible, and then let your landing page or APUS landing page do the sales work.

---

Hey mbml :P

---

@manu: Hey manu!

@mrgreen: Rio is my personal friend and witnessing his transformation and rapid success has motivated me to consider switching over as well. thanks for the useful tips! I guess I'll spend some time with a spy tool, probably better to learn from others rather than to learn the more expensive way. Testing out 10 campaigns with a budget of $50 is probably going to burn through my funds at an alarming rate.

Your point about achieving CTR is interesting. To be honest I didn't even consider the metric for payouts, didn't even know if I would be paid based on CPM or CPA. Was just trying to run my first campaign ASAP but I guess I should be more methodical and deliberate about it.

---

Now for some useful advice too, haha

First thing I'd say is think of it like poker and think of the initial few campaigns purely as an investment in stats. You then use those stats to improve your EV (or ROI), until you reach green.

Definitely a spy tool like AdPlexity can get you quite a bit ahead. Instead of testing 100 ideas, you can choose top 5 that seem to have some promise and see which is best for your traffic source + geo + offer.

Also, you have a good situation that you have the funds to scale the campaigns once they go green so always keep in mind you will have an advantage compared to many once you do learn what you need.

---

Great post! Will follow this one for sure!

Regarding the banners: indeed you have followed the Appetizer guide to the letter in your creatives. I was planning to do the same. Now I wonder why the appetizer would suggest us making banners like that when there's too much going on, as Mr. Green stated?
Was the Appetizer guide initially written for a PPC-based source maybe?

---

My previous banners were finally approved and I topped up my account with some cash to play around with. So now I know it takes approximately 2.5 days for go2mobi to process the creatives. I might just run one of them for fun, as a basis for comparison, but skipping the rest as it would be an absolute waste of money. Decided to take Mrgreen's advice, bought 1 month of adplexity. I'm not exactly sure what variables to use to narrow down my search.

Trends I spotted while analyzing the data

• Many of the banners are to be animated. i guess I'm not doing that for now.
• Majority of the banners have Download/Play in them. Seems to me that the banners are just meant to imitate the buttons commonly found in video streaming sites and deceive the users into clicking on them accidentally.
• Primary colours for background? Mostly saw blue/green buttons with white text, some had red backgrounds as well. Initially I thought warm/bright colours were more suitable for Calls to Action but maybe I'm wrong.
• Mostly English, but I spotted some translated banners (Turkish, Thai, Spanish)
• Fonts tend to be simple, and words were bolded. Nothing too fancy. Most of the banners also had the symbols for download and play accompanying the words

Should I try Indonesia instead? There were 20 hits in the spy tool as opposed to the Philippines which only had 5. Does that count for something?
I also searched for the translation of Download and Play for Filipino and Indonesian. The word for Download was the same, though when it came to play I wasn't sure what an accurate translation would be (since the word play has many different meanings). I think that there's really no point in testing a translated banner for now.

Messaged my Affiliate manager for some help with regards to making sure that the links are working properly. Do I have to start tracking stuff with Voluum now? Still feel a little confused.

I downloaded Paint.net so I'll remake some new banners with the same Download/Play words

---

Seems you got a bit too detailed into spying. This is not an exact science, it's just an overall picture. The way to learn the details is to test yourself. CTA colours are not the main thing you should look at, that's for sure.

Look at angle, offers and traffic sources, in combination. This way you get a good idea of what offers run where and using what angle(s).

I wrote a case study about my first profitable campaign. Maybe it's helpful for you as well to understand what mistakes I made that you shouldn't repeat.

In terms of Voluum/tracking in general, certainly; the sooner the better. There are other tracking solutions available, all have something different but for the overall package, I'd say Voluum is the best by a significant margin.

As for creating banners, I used GIMP2 on my very first banners, but then I learned the basics for Photoshop. First batch of banners was difficult, always googling the simplest things but then it got much quicker and better than with GIMP2/other tools. I'd suggest sucking it up, and spending 2-3 hours in Photoshop.

---

Welcome to the affiliate world Galvin.

Why not try both? Keep testing till you find something.


Split test your own banners with those ripped banners, and see how they perform, this would be a better initial test.

If you are not using landing page, Voluum is not really necessary at the moment. But you will definitely need a tracker later on.

All the best!

-Jon See

---

Hell yea thats awesome

---

How can you 'spy' on other people's angles if it's just all generic Download/Play buttons? That indeed looks like imitators for accidental clicks, as @gosubay already noted. As I already said in an earlier post: why would the Appetizer suggest using banners with proper angles, when apparently that doesn't work and we should all be using generic Download/Play buttons (like all our competitors)?

---

They will usually have a landing page after the banner.

---

Haha fair enough. It's just that the Appetizer promotes direct linking directly to the appstore/playstore, but I guess we'll have to create a LP then Thanks for your response manu!

---

@shepherd - Good work on the spying!

With regards to the banners you're seeing: there are several ways to skin any available cat

A lot of people will indeed use the "get the click any way possible, then transition to the angle on the lander" approach. If you're comfortable with getting tracking set up, that's something you can try too using landers.

As the Appetiser mentions, it's designed as an introductory approach that can generate profitable campaigns. Landers definitely improve your chances, but they also add a lot more complexity to the mix, and it's been my experience that people who attempt to do everything at once in AM often end up very confused or with big holes in their methodology.

However, if you're already familiar with HTML, webdev and/or data analysis, then there's nothing wrong with going straight to the tracker-and-lander approach. Just be sure to rigorously check and learn everything twice. In that case, you'll find "The Mobile Cookbook: The Main Course" useful.

Alternatively, stick with the Appetiser approach. You can test out the "Download" buttons going straight to the app store - they may work surprisingly well if the app store page for what you're promoting is well-written. And also look for banners that aren't using that approach: if you can't find them in the geo you're targeting, spy on other, bigger geos and see what's working there.

("Spy on Geo A, use approach in Geo B" is a very good way to find new angles for a given territory, particularly if you're porting from larger-and-more-competitive to smaller. Likewise, "Spy on traffic source A, use angle on traffic source B" works very well too.)

Good luck, and looking forward to your updates.

---

I also started with the Appetiser but I never thought about making money with the simple techniques explained there. Not to be misunderstood, they are important but they are just the base of the xxx/day or more campaigns, not the whole package. My main focus was seeing -95% go to -70% , to -50% and higher and higher after every few tests, adding complexity when there's nothing left to test in that level.

It wasn't so obvious then, but when looking back at it now, the focus on IMPROVEMENT not on PROFIT was a huge help for me personally. Red and green is the black and white of AM, but it's just not how reality works. It's a very large spectrum between red and green, or black and white.

Also, on the topic: DOWNLOAD | PLAY is also an angle. It's just very far from the marketing one would learn in school, that people outside of AM don't ever think about it.

---

a) Always depends, just don't think of them as one or the other, split test designs with and without animations.
b) Kind of same as a). As you gain more experience, you will see what are the elements that generate 80% of the revenue/profit/conversions and you will focus on getting those right. Colour swapping vs flashing is a small element which you can always test.
c) Depends on geo. For Philippines, definitely English. For Indonesia, I never even tried English, they don't speak it well at all. Malaysia is a geo where you can consider split testing, Hong Kong too... I generally google the official and/or most common language and initially test with that. If there's enough volume in the geo, and the offer showed at least some potential, you can add languages to a split test and see which is best.

---

Manu_adefy explains the purpose of the Appetiser perfectly above.

Don't focus on money when you're starting. Focus on learning. You need the learning before you can make the money.

---

Good answers that I don't really feel would need any more commentary

Let me just add one thing - if you decide to translate something, then translate both banner copy and the text on LPs. I had good results with both english and translated creatives but when I mixed them, it usually hurt the conversion. For example Malay banner and English LP didnt work. But english banner and english LP can work even in Malaysia. If I should make an average from the past campaigns I've done in non-english speaking GEOS, 100% transtaled funnel works best in 80%+ of cases, 100% english was like 50/50 for me, but mixed funnels usually failed - the worst was translated banner and english LP, english banner and translated LP worked pretty often.

---

Interesting insights, and come to think of it I never did mixed, it always seemed to not "fit".

Testing still best for this topic but: Perhaps English banners attracts people who speak English well enough + it's just a headlines usually, while when you get into details, it's harder to beat their best language. While if you do native/best language first, then details in a foreign language, the message isn't received well enough by the user.

---

Hi Matuloo, where do you get your banners and LP translated?

Do you use Google translator? or do you hire someone to do it for you?

Thank you!
Nicole

---

You could get your copy translated at some outsourcing platforms. Take a look at http://fiverr.com, http://upwork.com or https://www.onehourtranslation.com.

Hope this helps!

---

I use google translate when the copy is VERY simple. So phrases that simply cant be translated in a wrong way, like "I agree" or single words. Once it becomes more complicated I always look for native speakers to do it for me. I started with onehourtranslations, but they were too expensive for my liking, so I moved to odesk which works well for me - just make sure you look for native speakers. Its very cheap and usually pretty fast.

---

I think the reason is more simple, if you start with an english banner, chances are it gets displayed on a large % of sites that are actually in english, so it kinda blends with the content and the users take it as part of the site and doesn't find it weird. But when they see a translated banner and then the LP is in english, you loose the GEO targeted effect and it looks weird to them. That's how I see it

---

@gosubay - what's the spend been on the banner?

Given you're only running one banner, I'd not keep running the campaign. Pause it, get all your banners up, then run it.

---

Go2Mobi is very quick to answer compliance questions in my experience.

I would message support asking what's the line for each exchange, in terms of compliance, and also kindly ask if they can provide you an account manager to help you out. They are (or used to be) very open to connecting account managers with new users to help onboarding. Especially if you can explain you are serious about it, I think they will likely respond with a yes to the account manager request.

---

MoPub will not approve play/download banners until the day the internet dies. Banners must be branded with at the very least the app/adv logo and usually the name somewhere as well... and they are pretty sketchy on the whole "fake" button thing in general.

Headlined banners are the way to go on MoPub - from experience they do work. Not every display campaign has to be play/download based.

---

Exactly, dont even bother submitting play/download banners to mopub and google adx. Mopub will take a play button if there is also some imagery and a headline or logo on the banner.

I'm with stich on this "Not every display campaign has to be play/download based." ... not sure when banner creativity turned into ripping play/download buttons from a spytool only

---

You should have all your banners in one campaign though. Launching a campaign for each is not a split test, or not as accurate one as you can get.

Furthermore, linking the 2 is important for tracking. It's not vital right now, these are just your first steps here, seeing how each part of a campaign looks like but Voluum/Thrive/other tracking tool is your Holdem Manager. As soon as you want to look at numbers and make decisions based off them, you will have to either connect G2M and Avazu when you direct link or get a tracker when you add landing pages.

---

Ahhh ok, i feel stupid for not realizing that. I thought that should be the way initially, but got confused with the definition of "a single campaign"

---

Nice work!

Job #1 here is definitely, DEFINITELY to fix your conversion tracking. Without the conversions showing up on Go2Mobi it's very difficult to perform any significant analysis of the data: CPM and CTR will only get you so far and won't really let you optimise.

I'd actually say that your tracker is even more important than Holdem Manager / Poker Tracker / whatever. You can play poker without those and still improve your skills - but running a campaign without being able to figure out where your conversions came from is very, very hard. (I only know of one person who did it reliably.) It's more like trying to play online poker with the GUI disabled, just using the sounds

---

Home > Mobile > Follow-along Campaigns