Home > Mobile > Follow-along Campaigns

---

Going mobile (15)

---

Hi there,

I have been working for a month on some casual dating campaigns (WEB) and it's going pretty well - not profitable yet, but very close to break even. I will keep working on those campaigns "after dark" But I have a home office, so I need to do something else in the daytime when the kids are awake... I guess you can imagine that most of my casual dating banners are not too child-friendly

Therefore, I will spend my day hours on mobile campaigns instead.

I have no experience with mobile, so I read the mobile cookbook and a lot of other posts and blogs, to understand how to do it right.

I am pretty good with graphics and web programming - My weak points are:

English angle research/copywriting
Spying and defining the target audience
Analyzing the stats and optimizing the campaigns.

I know those are very important things, so I concider to try to pay a mentor, who can spend an hour or two each week to help me improve on those weaknesses - Please send me a PM if you are interested

My goal for the next month is:

To launch 1-2 campaigns per week.
To get a solid understanding of how to optimize campaigns.
To post my results and progress at least every second day.
To find out which offers work and which don't.
To create a "standard" procedure for launching campaigns.
To end up with at least 1 campaigns with some potential (I don't expect a positive ROI after 1 month)

So far i have:
A $1,500 budget
Tracking (Voluum)
VPS (Beyond Hosting)
Affiliate accounts (approved in 15 networks)
Alexa account
WRW account
Traffic source accounts (Decisive, Airpush, Avazu, Leadbolt, Mmedia, TapIt, RevMob)

I also picked my first offer, found 2 angles, created 10 banners and 1 lander for each angle (translated by onehourtranslation) and got those approved by the advertiser.

The way I picked the offer: I looked at the weekly emails from the affiliate networks to see if there was something, that seemed to have some potential. Then I spoke to my affiliate managers. They told me that the offer is converting very well in other GEOs but doesn't have much traffic in Sweden. I know from previous experience, that this type of offer should work in Sweden, so I decided to give it a chance. Other reasons: I am Danish and Swedish language is very similar, I have done a lot of E-mail marketing in Sweden and the offer runs in quite a few countries, so if it works, I could be able to scale it.

The first offer is:
[WAP] Responsive - Grocery Voucher - SOI - SE

The angles are:
"You have been selected"
"You deserve to win"
(hope i get this angle thing right - ANY comments will be appreciated)

I will split test the offer on 3 different networks:
F5: $3.35
Adsimilis: €2.80
C2M: $3.30

Since, I've read that there's a lot of bot traffic on Decisive, I will use caurmens method to check for bot traffic (http://stmforum.com/forum/showthread...hlight=destroy). I hope to find some good placements using this method and to be able to create a reusable list of reliable and bad placements for future campaigns.

I've uploaded the javascript redirect and created 8 campaigns in Voluum (the method from the Mobile Cookbook)

Next step is to create the campaigns in Decisive and wait for approval.

What do you guys think so far?

Cheers,
Jakob

---

Hey Jakob, thanks for posting all that detailed info.

From your description looks like you are doing it right.

Now just start the campaigns and let the data guide you.

Keep us updated!

---

For angles, think more along the lines of:

"Fill the shopping trolley this week without emptying the wallet"

"Family to feed? We can help!"

"Need to stock the cupboards? Win grocery vouchers!"

---

[WAP] Responsive - Grocery Voucher - SOI - SE wasn't the best for me across multiple traffic sources. For some reason, the DK version did quite well though... Might have been the translator I used

Try "[WAP] Supermarket vouchers - SOI - AU" on F5 media - that does quite well.

Also, $1,500 for 7 traffic sources is way too little. Pick one. I'd avoid AirPush (quality is suspect and you'll need to block a lot of zones) and MM (good quality, but lots of volume and you'll need to block a lot of zones).

Leadbolt is great for beginners, so easy to get a profitable campaign, but optimisation is a pain because you have to send a list of pubs to block, you can't block them yourself

---

Thanks for explaining angles AGAIN (I think I am getting closer to some kind of understanding)

Thanks for the advise - did you run the campaign on Desktop or Mobile?.

About the budget: I am not planning to use all traffic sources. I will use 1 or 2 to get to know them well before jumping to another one. So for this month only 1 or 2 traffic sources - sorry if i didn't make myself clear.

-----

So the first campaign was approved while I was asleep last night. I started it this morning - I've attached the stats so far.

As I wrote yesterday, I am direct linking through caurmens bot detection script. I don't have much data yet, but as you can see from my Voluum stats the CTR is not even close to 100% (as it should be) - Can i be sure that the missing CTs are bots or could this be caused by something else?

Today I found offers for my next campaign. I spoke to my AM at Yeahmobi to get some inspiration, and i picked a couple of AntiVirus pinsubmit for Malaysia.

I created a list of 15 angles and picked 2 - more about that later. I also created 10 different banners so far - I've created all banners in both English and Malay (It didn't take much longer, and i suppose i can use the English ones for another campaign in the future)

This was just a short update. I am going out for dinner, but I will post the details for Campaign #2 later.

Best,
Jakob

---

OK guys time for an update.

Last night i saw zenos video tutorial on Adobe Muse - I have always used photoshop/dreamweaver and coded things by hand - but I guess i could save some time using Muse for creating landing pages - at least I will try it out - it seems pretty cool and easy to use.

Today I finished the banners for my new campaign, which is a Android Antirus campaign targeting Malaysia. I created the campaigns in Voluum and Decisive and now they are pending approval.

The offers I will be using are all from YeahMobi:

Shoopadoo - Antivirus - Payout $4
Antivirus - Celcom - Exclusive - Payout $3.50 (There are other offers for other carriers but I will begin with Celcom only, since it's the biggest carrier in Malaysia and also has a better CR than other carriers)

I will use the Mobile cookbook approach again, but will only target Mobile users using Celcom - so only 2 campaigns per angle (APP/SITE)

My angles are:
Kill The Virus before it Kills Your mobile
Massive Virus Outbreak in Malaysia - Is your phone protected?

The idea is to scare the prospects - to "force" them to take action. Let's see if it works.

Banners are translated to Malay - however the offers are in English, so I am not sure how it will work - but according to my AM the offers have a good volume.

Since my Swedish campaign indicated a lot of bot traffic, I will use the same method for this campaign.

The Swedish campaign is paused right now.

The total stats are:

Impressions: 135,004
Clicks: 884
Spend: $40.98
Landing Page Clicks: 632 (Should be 100% since traffic is just redirected after 300MS. However most of the traffic that never redirected came from 1 specific placement (Mediashakers.com), where nearly 1/4 of the hits came from and not a single one got redirected)
Conversions: 1
Earnings: $3.30

I don't think this campaign is worth spending too much time on? Do you?
I know the offer convert for some affiliates, but I don't know if it's web traffic or mobile - I made a big mistake before starting the campaign - I didn't go through the offer - and it's a real pain in the ass on a mobile - a survey with nearly 10 pages of questions and a pretty messy user interface - at least on mobile.

So what I've learned so far:
ALWAYS try to complete the offer before picking it (if possible - guess it's not possible for pinsubmits)
ALWAYS be aware of bot traffic (start every new campaign using the bot detection script)
USE Pomodoro timer to stay focused and to push my own limits

---

If the offer is a minefield from your perspective then what do you think less tech savvy users are going to feel?

Drop it.

---

Excactly what i thougth after going through the offer... Stupid I didn't do it before spending time on banners and landers... But I will NEVER forget this important step again..

My Malaysian antivirus campaign launched this morning... So now i am collecting data, while creating landers for the campaign

---

Hi Jakob,
you are doing good job! wait for your good news.

---

what do you mean by leadbold is easy to get a profitable campaign ? what kind of offers is suiteable for that network ?

---

As native swed i have had no success in my own market. I feel a bit ashamed. I really dont know. I started well in Norway, Finland with similar offers

---

I too have just started using Muse, could someone link me Zeno's Muse tutorial, had a search but couldn't find it.

---

http://stmforum.com/forum/showthread.php?15421-Quick-Guide-to-Desktop-Landers

---

So guys... Time for another update:

Stats from the last 3 days:



As you can see, there are 2 main problems: Traffic is very limited and there haven't been any conversions.

So far I killed 1 placement in the 2 site campaigns and 1 app in the 2 app campaigns, because they had sent 60 clicks without a single one getting redirected on my bot detection page.

I paused the 2 campaigns which have spent > $14

Question is... What to do now?

I am thinking

1) should I copy the campaigns to new ones using a lander and let the bot detection campaigns run (they are direct linking so that could be the reason for 0 conversions)
or
2) should I kill the campaign - the offer had 330 hits without a conversion direct linking - does it have any potential at all or should I move on?

What do you think?

/Jakob

---

I would leave the bot detection campaign running - it's a smart move - and start a new campaign that sends traffic from the non-bot placements to a lander(s) for further testing.

LPs often outperform direct-linking so it's definitely worth a test.

Just blacklist placements from the bot campaign regularly (and in other campaign according) so that it's always testing fresh traffic, not stuff you know is good/bad.

---

Home > Mobile > Follow-along Campaigns