Home > General > Affiliate Marketing Forum

mobile cpi angle analysis (11)

11-09-2014 05:26 AM #1 acepowermarketing (AMC Alumnus)
mobile cpi angle analysis

i have tested a variety of angles, essentially i have ripped everything i can find and tested them against each other and these are the observations

1. the hardcore upfront scare tactics do not work the best

2. the general messages, improve now! dont work either, and they are worse than hardcore tactics

3. the balance seems to be a balance - able to stir up emotions, yet not make them shit their pants, and feels like something is wrong, some sense of a system notification, yet not explicitly saying it such that traffic sources start to take issue with the "misleading" ads.

4. angle is zoomed into a specific problem, for example in the antivirus offer, instead of saying you have a virus, a zoom in will be TROJAN2X7A found that compromises loophole. your credit card information is at risk. it states a more specific problem, the virus, and targets a specific security issue, your credit card.

11-09-2014 01:14 PM #2 bbrock32 (Administrator)

Yeah the more specific the better it works in my experience.

Also mimicking the UI / font of the OS you target helps boost the CR a lot.

11-09-2014 04:27 PM #3 Mr Green (Administrator)

The way you are stepping back an analysing things will definitely keep you ahead of the pack. Even when some of your test are not profitable you will still be getting ahead by paying attention to these details.

Nice work!

11-09-2014 05:31 PM #4 acepowermarketing (AMC Alumnus)

Originally Posted by bbrock32

Yeah the more specific the better it works in my experience.

Also mimicking the UI / font of the OS you target helps boost the CR a lot.

The issue with mimicking the ui is that if it looks exactly the same, it's denied by many traffic sources as misleading. So the mimicking seems to have to pass the "intelligent man" test. Where the traffic source representative cannot exactly point out which part of it is like a system notification.

Yet it should fail the "grandma test" where a very ignorant user should not be able to tell the difference between a real system notification and your ad.

At the same time I'm not sure why direct system notifications are ineffective. Maybe it's overused? Maybe the user feels scammed? A non direct one cannot cause the user to feel scammed because anyone smart enough to think will not be able to classify the ad as misleading.

11-09-2014 05:33 PM #5 acepowermarketing (AMC Alumnus)

Originally Posted by Mr Green

Thanks for the encouragement! I'm still seeking to systemize and classify the most important elements of an advertisement to go around improving it

11-09-2014 06:08 PM #6 iAmAttila (Veteran Member)

I like all of these tests you are doing.

Here's one you should try that worked very well for me on pops/redirects.
Instead of scaring them, advise them politely using words like please, did you know, etc and include testimonials which are the features of the offer you promote, and turn them into a benefit...

ie I downloaded this app because my friend recommended it, and it really helped me get rid of a nasty trojan I didn't even know I had! -Maurice Stedman, AOLer

11-09-2014 06:13 PM #7 acepowermarketing (AMC Alumnus)

Originally Posted by iAmAttila

At first I turned this away due to the sheer number of words and brain processing required. I thought it needs to be simple but somehow I under estimated their ability to understand. Somehow the testimonials did not work for me, maybe I did not frame it in a relevant manner, or tested it along many other variables or structured an unfair test.

I'll go back again with your recommendation of
1. Polite phrases vs normal phrasing
2. Different credibility sources like testimonials, I tested the Google play store logo, the many thumbs up, likes, but again the tests were all over the place.

11-09-2014 06:15 PM #8 iAmAttila (Veteran Member)

Originally Posted by xavierfok

it could be the offer, remember just because it works in ONE country, or for ONE offer doesn't mean its a blanket rule that it'll work everywhere

that is why we test a shit ton of angles/creatives everytime we enter a new geo, can't just make assumptions

11-09-2014 06:20 PM #9 acepowermarketing (AMC Alumnus)

I thought that for cpi offers that are international the angles should be pretty much similar? Or go around 2-3 main angles? If we have to do the entire process all over then moving to another country is not considered scaling, more like working on a new offer?

11-09-2014 06:36 PM #10 iAmAttila (Veteran Member)

Originally Posted by xavierfok

not exactly... i've spent over $x,xxx,xxx on cpi media buys this year and i can tell you based on the data - that every geo is different

as a REAL example

when the offer clean master was around

in one geo, battery dying? angle won
in geo two, phone slow? angle won
in geo three, phone freezing? angle won

11-09-2014 06:38 PM #11 acepowermarketing (AMC Alumnus)

Originally Posted by iAmAttila

I ripped your phone freezing and it died. I'm going to try on another geo!
Battery dying for a clean master angle? I was expecting battery saver lols

Home > General > Affiliate Marketing Forum