Home > Paid Traffic Sources > Mobile

Manzie, Mr Green and other Mobile pros. ....Efficient Mobile optimization??? (25)

02-28-2013 01:50 PM #1 marketone (Member)
Manzie, Mr Green and other Mobile pros. ....Efficient Mobile optimization???

Ok So i had to call out some of the big boys...Sorry guys, lol. What im wondering and im sure the answers may vary from one person to the next.

What is the most efficient way to optimize a mobile campaign?

You have many variables to optimize....Like:
* Creatives
* landers
* offers
* carriers
* operating systems
* handsets
* pubs

I know if you optimize some variables before others that you can skew some data and cut your potential on your campaign.

So maybe i can get you mobile pros to chime in and share your systematic approach to optimizing your campaigns efficiently.

Im sure most everyone here would learn something from it. Thanks

02-28-2013 02:06 PM #2 creathinker (Member)

Hey bro, this is the sequence of variables I cut in mobile when optimize :

1. Carriers
2. OS
3. handsets
4. pubs (if available)
5. creatives

You have to spent some bucks to get more data and eliminate the variables.. Almost of my mobile campaigns is direct link (this is the beauty of mobile), but you can to optimize lander after optimize your creatives..

Hope can help..

02-28-2013 05:17 PM #3 stackman (Administrator)

Hey where's my name!

Optimizing variables will depend on the offer your'e running, ie: you may run an offer that converts well with ALL carriers but only converts well with Android. In this case handsets are more important variables then carriers. At the end of the day you need to track/cut everything anyways.

Mobile offers are very dependent on Carriers, OS and handsets. If you've got a good offer, then creatives/landing pages come as step 2. I look to put the process in 3 main steps.

STEP 1:
- Run the offer direct, cut carriers/handsets/os at the same time. This is what i track first and this is the most important step to finding your profitable handsets/carriers. I only cut ones that are 100% garbage (this is important, because some of the smaller ones/ ones you are unsure about may convert in step 3).

STEP 2:
- Run a couple angles/creatives etc... and see what concept you come up with works best. Fine that winning creative angle (and landing page if it converts better then direct).

STEP 3:
- Run your awesome angle with the carriers/handsets that you haven't cut. Run a good chunk of volume to this page, get a lot of data. THEN cut all the bad carriers/handsets. Now you should be nicely profitable if you have the right offer etc..

If there was a step 4, it's to now take your winning angle and make more angles around it, and split test those. Step 5 is to SCALE to new traffic sources, and Step 6 is to SCALE to new countries (if possible).

02-28-2013 08:58 PM #4 seank (Member)

This ^^^^ is GOLD.

Nice meeting you in Dam a few months ago, although briefly

Sean K.

02-28-2013 10:05 PM #5 maynzie (Moderator)

Jordon just the simplicity of that Optimisation process is worth forum sub over and over, you should repost it as a post so everyone sees it haha

02-28-2013 10:32 PM #6 stackman (Administrator)

haha I'll include it in the BIGGEST guide I've ever written which is underway right now.

It explains absolutely everything you need to promote ALL mobile offers, it trims the fat, cuts through the BS and tells you exactly what you need to know to get started that day. Done in approx a week

@sean, you coming back in May?

02-28-2013 10:42 PM #7 Mr Green (Administrator)

Yeah I agree with Stackman, except...in quite a few cases I wouldn't start direct linkin since some offers tank so much direct linking it would be too expensive to get enough data.

A lot of networks now have data on what carriers and handsets work for mobile offers, if they don't have it themselves, ask them to ask the advertiser, you can move forward a lot faster this way.

02-28-2013 10:51 PM #8 stackman (Administrator)

True, it 100% depends on the offer.

ie:
NO DIRECT = Sweepstakes
NO DIRECT = Adult dating
DIRECT = Ringtones
DIRECT = Games

By saying direct i don't mean forever, but you can test certain niches by direct linking.. and then sometimes direct linking will actually work better than any landing page (proven by my own tests)

03-01-2013 12:34 AM #9 marketone (Member)

Sorry Stacks didn't mean to leave you out man!! I appreciate the insight. I'm sure everyone else appreciates it as well. If any of you other mobile pros have anything to add....please do. I know there are more of you out there lurking.

03-01-2013 04:07 AM #10 szmudo (Member)

Quote Originally Posted by stackman View Post
Hey where's my name!

Optimizing variables will depend on the offer your'e running, ie: you may run an offer that converts well with ALL carriers but only converts well with Android. In this case handsets are more important variables then carriers. At the end of the day you need to track/cut everything anyways.

Mobile offers are very dependent on Carriers, OS and handsets. If you've got a good offer, then creatives/landing pages come as step 2. I look to put the process in 3 main steps.

STEP 1:
- Run the offer direct, cut carriers/handsets/os at the same time. This is what i track first and this is the most important step to finding your profitable handsets/carriers. I only cut ones that are 100% garbage (this is important, because some of the smaller ones/ ones you are unsure about may convert in step 3).

STEP 2:
- Run a couple angles/creatives etc... and see what concept you come up with works best. Fine that winning creative angle (and landing page if it converts better then direct).

STEP 3:
- Run your awesome angle with the carriers/handsets that you haven't cut. Run a good chunk of volume to this page, get a lot of data. THEN cut all the bad carriers/handsets. Now you should be nicely profitable if you have the right offer etc..

If there was a step 4, it's to now take your winning angle and make more angles around it, and split test those. Step 5 is to SCALE to new traffic sources, and Step 6 is to SCALE to new countries (if possible).
Actually, step xy (and also inbetween scaling sources and countries) should be landing page optimization, meaning creating more landers, testing them again in every country and for every new traffic source.

If you don't do that, you will leave a lot of money on the table, I have many landing pages that convert different for traffic sources, carriers, countries etc. So I always test them again, and also throw in new landers into the mix (always test 10+ new landers, if you're too lazy to do that, here is what seperates you from really successful people ,o) )

Talking 5% conversion rate and $0.10 epc or 12% conversion rate and $0.25 epc... Or $500 profit, or $2000 profit per day

03-01-2013 05:10 PM #11 stackman (Administrator)

Quote Originally Posted by szmudo View Post
Actually, step xy (and also inbetween scaling sources and countries) should be landing page optimization, meaning creating more landers, testing them again in every country and for every new traffic source.

If you don't do that, you will leave a lot of money on the table, I have many landing pages that convert different for traffic sources, carriers, countries etc. So I always test them again, and also throw in new landers into the mix (always test 10+ new landers, if you're too lazy to do that, here is what seperates you from really successful people ,o) )

Talking 5% conversion rate and $0.10 epc or 12% conversion rate and $0.25 epc... Or $500 profit, or $2000 profit per day
That's a given, was explaining the basic variables to look at first
Just wrote about the math behind optimizing before scaling :P http://stmforum.com/forum/showthread...e-2688-PER-DAY

03-02-2013 01:15 AM #12 szmudo (Member)

hehe wasn't trying to lecture or offend you :P Just threw in an important factor that should be part of any step by step optimization guide imho, and used your answer as a starting point. Because this is what takes a campaign from barely profitable to highly profitable more often than not.

03-02-2013 04:51 AM #13 marketone (Member)

I appreciate any and all input. Thank you guys. That's why I love this forum.

04-11-2013 06:55 AM #14 stian (Member)

Quote Originally Posted by stackman View Post
Hey where's my name!

Optimizing variables will depend on the offer your'e running, ie: you may run an offer that converts well with ALL carriers but only converts well with Android. In this case handsets are more important variables then carriers. At the end of the day you need to track/cut everything anyways.

Mobile offers are very dependent on Carriers, OS and handsets. If you've got a good offer, then creatives/landing pages come as step 2. I look to put the process in 3 main steps.

STEP 1:
- Run the offer direct, cut carriers/handsets/os at the same time. This is what i track first and this is the most important step to finding your profitable handsets/carriers. I only cut ones that are 100% garbage (this is important, because some of the smaller ones/ ones you are unsure about may convert in step 3).

STEP 2:
- Run a couple angles/creatives etc... and see what concept you come up with works best. Fine that winning creative angle (and landing page if it converts better then direct).

STEP 3:
- Run your awesome angle with the carriers/handsets that you haven't cut. Run a good chunk of volume to this page, get a lot of data. THEN cut all the bad carriers/handsets. Now you should be nicely profitable if you have the right offer etc..

If there was a step 4, it's to now take your winning angle and make more angles around it, and split test those. Step 5 is to SCALE to new traffic sources, and Step 6 is to SCALE to new countries (if possible).
In STEP 1, when cutting OSes - are you referring to the different Android/iOS versions i.e. 4.1.2? Or Android/iOS/Windows - the OS as a "whole"?

Stian

04-11-2013 11:07 AM #15 bbrock32 (Administrator)

I think he meant cutting the OS as a whole , cutting versions would be very time consuming and generally I've noticed versions don't make a huge change.

04-11-2013 11:20 AM #16 deondup (Member)

^^ The exception would be with certain Apps. that just don't work on certain versions of the OS - especially on Android.

04-11-2013 12:58 PM #17 stian (Member)

Ok, thanks guys!

What if;

In step 3 - you've been running a single handset - let's say Samsung Galaxy S III for 10 X payout total spend. The handset shows a few conversions. But still sitting at -50% ROI. Would you cut it? Or keep tweaking lander and try to make that handset profitable?

Of course you would be running loads of handsets at this point, and this is just the stats for that particular one.

04-20-2013 11:44 AM #18 stackman (Administrator)

@stian

If other handsets are converting well at this point and that specific handset is -50% roi after spending 10x the payout i would cut it unless its a handset like iPhone4 or a GalaxyS3 that brings a lot of traffic. If it's a "mega" traffic handset then i would try optimizing further before i block it.

04-30-2013 04:48 PM #19 neyphil (Member)

have you guys tried using sitescout to buy mobile traffic?

04-30-2013 05:14 PM #20 nefig (Member)

We tried, it sucked. mostly app traffic, very little WAP placements, not a lot of volume. It has a lot of potential imho, but inventory is lacking at the moment. Would be great to choose same placements + Android as OS for example but it's not possible. Device targeting only becomes optional when "mobile" is chosen and it limits the available placements a lot, which doesnt make much sense to me: there's a ton of responsive sites out there that have 300x250 units placed prominently which can deliver nice CTR..

04-30-2013 05:43 PM #21 The Angry Russian (Moderator)

Jordan as always great post.

I think for most n00bs even that is a bit complex, I'll try to sum it up to the following line.

Cut the shit first, look at whats working, break it out, optimize.

I don't know if there is an "order" of optimization since sometimes one site ID can kill your whole campaign. So I always look for something that is completely not converting whether its OS, Site ID, Carrier, Handset it doesn't matter if something is eating clicks and doesn't convert axe it. Then I see what is working and scale along those lines. I try not to over optimize because you tend to kill off volume so instead I focus on creative, day parting, and post click optimization to try and produce a higher yield.

Side Note: If something is sucking up tons of clicks and converts at a low % doesn't mean block it entirely, just break it out into a new campaign and bid at a price where it will back out.

Just my $.02

02-17-2015 05:12 PM #22 seanyboy (Member)

post click optimization to try and produce a higher yield.
Great thread here guys.

From this quote, do you essentially mean collecting email and monetizing on the backend? Do you find it simple enough to collect emails via display or search mobile advertising?

Thanks!

02-18-2015 12:45 AM #23 globejohan (AMC Alumnus)

Great tread, will read these advices every morning i waking up, Maybee i put a big note aswell abow the bed.

02-22-2015 11:06 PM #24 walt_white (Member)

By "post click optimization" I think he means optimizing the funnel after the ad is clicked, so landing pages and offers.

02-24-2015 11:05 AM #25 seanyboy (Member)

@ Walt_White.. thanks for clearing that up dude

Home > Paid Traffic Sources > Mobile