Mobile DSP follow along - Match my SEO income by March 2017 (10)
Mobile DSP follow along - Match my SEO income by March 2017
Alright Ladies and Gents - Here's my plan/goal: Match my SEO income of $300/day profit by March 2017
I'm new to AM but not IM, so I'm hoping this goal will motivate me enough while being relatively modest.
To be honest, $300/day is not my true goal.
My REAL goal is to master the skills I need to launch and optimise a campaign from red to green. Because once I know how to do that, well the sky is the limit. The figure of $300 day is not so important; its just a benchmark to aim for.
STM: The Goldmine…the Overwhelming Goldmine.
Jesus Mary and Joseph there is a shit tonne of information in the STM forums!!! (I say that in a good way)
I’ve spent the month of August diving into the STM threads and learning the basics. It’s left me anxious, and excited. Anxious because I realise I have a huge fucking learning curve ahead of me; excited because I know these skills will pay dividends for life.
Here’s a tip I have for any other newbies who are just starting this research phase: Always have a problem/question in mind you’re trying to answer while browsing the forum, rather than aimlessly browsing the forum.
I noticed that I became overwhelmed in the beginning while browsing STM.
I’d go down a rabbit hole of information, and after clicking 20+ links I’d have gotten nowhere and spent 3 hours getting there. Not a good feeling!
So I made a pledge to myself: I’ll search the forums only when I need specific information to a question/problem. I’l search the forum with a goal.
And I must be taking notes in Evernote while I’m researching the answers to my questions. There are so many threads littered with golden nuggets, you’ve just got to put those nuggets somewhere where you can use them later.
No more scrolling while eating dinner or sitting on the can. This just leads to overwhelm.
Lesson learnt: Use the forum the powerful tool that it is, not another platform like Facebook where you aimlessly scroll through for entertainment.
My Plan in a Nutshell
Im going to start with the Mobile Appetiser cookbook…more or less… because its something to follow along with. (Thanks @Caurman for creating it, and anyone else who was involved)
Traffic source: Go2Mobi - Mobile DSP
Vertical(s): Undecided yet. I'm going to start by quickly testing a large handful of offers (20+) to find something with potential. The winner(s) will determine my vertical. At this point Im not tied to anything.
GEO: i’m going to start with SE Asia and latin American Countries - because I’ve heard the traffic is cheaper.
Phase 1 - BLINDLY picking offers from affiliate networks for testing
I’m adding an extra step to the appetiser. I’ll be testing 20+ offers quickly from the beginning to try and find something (or a few, hopefully) with potential to continue with.
I’m blindly picking these offers from the database of offers in the affiliate networks I’m part of (YepAds, ClickDealer, PeerFly and F5 Media)
Before you rip me to shreds for choosing an offer this way, let me explain:
Everyone recommends that the best way to choose an offer is to ask your AM for a list of whats working. That makes complete sense. Start with something thats already working.
In my experience, however, AM’s tend to ignore such requests from brand-spanking new affiliates who have zero track record.
I don’t blame them - they have quotas to hit, and they want to focus their efforts of their affiliates who are actually getting results. So its my goal to ‘get on the radar’ by being proactive and sending them what they want: Traffic.
My hypothesis is that once Im sending traffic and conversions, it will be harder for them to ignore my emails.
Or, someone will tell me im doing it all wrong, and advise me otherwise (Thats what follow alongs are for, right guys?) 
The Goal: Find 3 offers with potential to knuckle down on.
The Hypothesis: If you use generic banners to direct link to 20 different offers, X offers out of the 20 will convert more than the majority. (X being 1-3 hopefully, but I’m aware it could be 0 meaning I have to test 20 more offers). These ‘winners’ are worth investing more time and effort into optimising since they have just shown some potential.
Therefore, I'll be following these steps:
- Choose 10 PIN submit offers with payouts <$1
- Create generic banners, trying to keep the banners the same as much as possible (Same layout, same Icon, same colors, same/similar CTA)
- Direct link and spend $20/offer…….OR…..send X clicks/visits to each offer landing page. The reason I say this is: If I send $20 to offer A in Brazil and $20 to offer B in Mexico, I may get more traffic to offer A due to the GEO or traffic source, so the conversion data may not be an accurate indicator of the offers potential.
- Rinse and repeat until I have 3 offers that show potential. At this point, I’ll continue along with the mobile Appetiser strategy (i.e. start creating angles, blacklisting placements etc)
What i've done so far
- Learnt about tracking and how to set up
Voluum , and offer, and a traffic source. This took me a while, but I'm quite confident with it now. If you're struggling with this, here is a thread that I found VERY helpful - Made some generic banners using the free banners on STM (I used google translate for this. Even though there is minimal text, I know there's a good chance that there will be mistakes so in the future I'll use fiverr)
- Tested 4 offers. I had a few conversions (although I went wayyyy over my $20 budget), but I'm going to wait until I have 10 offers tested so I can show everyone all of the banners and the data.
What i'll have done for the next update:
- Spent $20 on 10 offers via direct linking
- Report on my fuckups (there has been a few so far)
- Figured out how to target carriers more accurately on Go2Mobi. I noticed that a few of my offers where only allowed to be run on specific carriers, yet I could NOT select those carriers in G2M. I got in contact and they said....and i'm paraphrasing: "our system has floors, the most accurate way to target carriers is to find the IP ranges they run on and whitelist them that way". I did a little research on this and it looks a little tricky so im going to figure it out and have something to say about it for next time.
Thanks for reading my rant! I'm pumped
Originally Posted by thetailend
A few hours delay in reply if you're newer is normal... but if a manager doesn't respond at all, they don't deserve your traffic - newbie or experienced. Ask for a new one or ditch the network.
Originally Posted by stitch
Ok - Time for an update.
I have conversions!
I’m not sure if I have offers with ‘potential’ yet, but I’ve got something to work with:
What I did, in a nutshell:
> Created generic banners and direct linked them to the offers.
> Used the Smaato exchange for all offers
> Used "autobidding"
> No targetting apart from what was specified in the offer's terms (e.g. carriers, IOS etc)
Yes - it looks like I did F all. BUT I learnt about tracking and cake (using volume) and how to use photoshop in the process, so its a win in my books.
My goal was to ‘test offers’ and find a few with potential.
I wanted the only variable in my campaigns to be the offers. So I used generic banners and tried to keep them as similar as possible. This is the outline I used:
Nothing special at all.
You get the point.
I’ve hit a few roadblocks and hence have not done all I said I would since last time, but I’m making progress.
Here are a few of the issues I faced, and why:
Problem - I have 2x Thai campaigns that are getting almost no traffic
WHY - I'm not quite sure - it could be becuase I'm just using one exchange and one banner.
Solution - G2M support told me to whitelist more exchanges (I was only sing Smaato) and use more creative sizes. Lets see what happens
Problem - Carrier targeting in smaller GEO’s is limited on Go2Mobi
WHY: G2M’s carrier targeting tool is flawed and not so accurate (according to them)
Solution - You can target carriers using IP ranges which is supposedly more accurate. I looked into buying IP ranges from Maxmind. This is a little too technical for me ATM and hence, overwhelming. I’d prefer to focus just on the basics for now (learning how to create angles, landing pages and optimise) so i’ll avoid GEO’s where I need to use IP’s for now.
Problem: Some carriers appear in Go2Mobi as crossed out red lines
WHY - I asked Support and in a nutshell, its another limitation of their carrier targeting tool.
Solution: Learn to target carriers with IP ranges, once again.
Problem: Viewing a mobile landing page without getting redirected.
WHY: It makes sense to visit an offer page before running the offer to research it, and see if its working/live. I use VPN to connect to the local GEO, but i’d still be re-directed to another generic offer for a small number of offers. Does this happen because Im on a desktop and/or NOT using the allowed carrier/device/OS etc?
Solution - The obvious answer: buy a few different old handsets, maybe even orders certain carrier sim cards online and use them with a VPN, but is there another, less time intensive way? I'd love some tips.
What I'll be doing for the next update:
1. Ask my non-responsive AM’s about hot offers (now that i’ve sent some traffic - maybe they will take notice)
2. At the end of this week I should have my ‘top 3’ offers from this batch of testing, so I'll choose those 3 and;
3. Start running them according to the Go2Mobi appetiser (i.e brainstorm angles, make more banners and drill down on them)
Hello thetailend,
you've done a couple mistakes with your campaigns, so let me address some of the issues.
1. choice of offers - I understand your concerns, but as stitch told you, there is no reason for this. Find a network with responsive AM and stick to it for now. It's a good idea to test plenty of offers, but a better way of doing this is to pick one or two vertical and look for similar offers within it, then test those. What you've done is quite a mess and won't really push you further as the data is fragmented too much.
2. GEOs - same goes for the GEOs, you picked too many and didn't have the chance to properly test any of them. As witht he offers, it's better to focus on 1 or 2 for now and rather test more offers.
3. banners - when we talk about GENERIC banners, we usually mean something without any angle, simply a "play" or "download" button and nothing more. These are misleading banners, but that's the point - to get cheap traffic fast, without targeting it in any way. The banners you have done already have angles on them. Also the style isn't the best, these look like virus alert banners, I'm not surprised they bombed totally with iphone or whatsapp offers. Do you have access to some spytool? I'd recommend to get some banners there, it will be easier to get an idea about the possible performance this way.
You also need to test more banner styles, even if you decide to continue with generics. Small changes in layout and color can make a big difference. Test more banners in a single campaign.
4. bidding - don't use auto-bidding unless you really have to - it's ok to launch with it to see what kinda bid is needed to reach what WR etc ... but it will usually be higher than needed and you will be loosing money this way. You don't want 70% WR either, shoot for 20-30% for best price/value ratio. Make sure you start with tight capping too - 1/12 or 1/24.
5. Running too long - when you see results like $15 or so CPA for offers that pay "cents", STOP running such campaign immediately. You spent $104 in BR to get 72 cents back, this is a horrible result. It's absolutely impossible to optimize something like this into profit, so it's a waste of time and money to continue with it. The data is worthless too as there is obviously some problem in the funnel or the offer is a really bad one. I have seen campaigns that started at -70% get into profit, but those you have are faaar from that, I'm sorry. Stop all of them.
So what next? 
You are right about the carrier targeting at go2mobi - it's far from being perfect. However, it works well with some carriers, so I would suggest to check your data and see what campaigns actually sent you properly targeted traffic and maybe continue with those GEOs. You can buy the IP ranges from maxmind but they are known for being "NOT" so accurate There are also better providers but the DBs are rather expensive. IP ranges of mobile carriers are still one big mess so to speak.
I would definitely recommend to get a spytool (adplexity is a good one), take a look at what is running right now and what creatives people use and rip some of them. Rip some landing pages too, many verticals are impossible to run without them. Pick 1 or 2 GEOs and 1 or 2 VERTICALS and get as many offers for those as possible. You will need to test a lot of them until you find the right one, these days it's a LOT about the offers. Your first goal is to start getting stable conversions without loosing an arm and leg, then we can move on and get to optimization.
You should also look for more traffic sources, if go2mobi doesn't work for you, check other sources. If banner traffic isn't working, you can try POPs ... I'm not telling you to jump from one thing to another day after day, but also don't stick to something for too long when it's simply not working out for you
Looking forward to your next posts.
Thank's, Matuloo!
In hindsight (for me as a newbie) I can see how my first attempt has been all over the place.
It makes so much sense to focus on 1-2 geo's and verticals for testing. Thanks so much for your advice.
I bought Adplexity and decided to focus on the following 2 Geo's and verticals for now:
> Vietnam and Brazil
> Antivirus and mobile content downloads
I did as you suggested and looked at carriers in Go2Mobi that were sending traffic (although I'm not completely sure on how to determine whether its targeted or not without many conversions - I guess by looking at clicks, right?), and thats how I came to choose these 2 GEO's.
Back to the drawing board!
Azza
Brazil and Vietnam are popular choices for sure. From some reason I wasn't able to make Brazil profitable myself, it's one the weird GEO's where you either hit it or not. I was doing well with appinstalls in BR some time ago tho. If I had to suggest one GEO from latam, I'd vote for Mexico, that always worked better than BR for me.
To determine whether go2mobi can properly target the carriers you want, you have to make a campaign and target just that one carrier. Then check "mobile carrier" in
This mobile carrier targeting thing seems to be a hurdle but I like your idea of sticking to what is working Matuloo, thanks!
Had a few issues with offers expiring/being taken down (to be honest I was just a little too slow) so in the mean time I'm running some more traffic to one of the offers I tried testing in my initial batch.
Offer: Phone cleaner pin submit (recommended by AM)
Geo: Iraq
Traffic: Go2Mobi
Spend: $34.88
Revenue: $2.88
After brainstorming a bunch of angles and problems I chose one randomly, as suggested in the mobile appetizer. "Your phone may be hacked, clean now"
I bought Adplexity but could not find any good banners to rip (I did find some landing pages and popups however. Maybe I should try this offer on pops?)
These are the banners I created and ran, and the results:
Now, since i've had a few conversions the appetizer suggests:
"now allow your campaign to reach $60 spend, which should take 2 days.
DO NOT mess with your campaign during this time. No matter how much you might be tempted to cut banners, add new ones, or do anything else with your campaigns, resist the urge until it has reached a $60 spend. "
I will do that, but first:
[SIZE=4]I have 2 questions and would love some guidance
[/B]
(1) I'm doing autobidding, and from the advice I got above, I know I should NOT be doing autobidding, and aiming for a WR of closer to 20%. My campaign is now paused: should I let it run to $60 an autobid, or change it to a manual bid?
(2) For next time, how can I determine the right bid amount to get me to a respectable WR? is it a matter of just picking a random low number and adjusting until the WR is good?
Muchos gracias!
Originally Posted by thetailend
So the choice is yours. Truth to be told, the results don't look too well for now.1: I wouldn't autobid, it simply means you are overpaying. Change to manual and bid lower.
2: I always start low and drive up the bid to reach approx 20%-30% winrate. That's not a rule carved in stone, but it's the level where I usually see the best results. So consider it a tip based on my experience.
I've taken everyone's advice and moved to pops!
My new follow along is here, for anyone interested: http://stmforum.com/forum/showthread...348#post295348
Home > Mobile > Follow-along Campaigns