Home > Mobile > Follow-along Campaigns

---

Mobile Browser Brazil (CPI) (33)

---

I've launched a few campaigns lately.
Most of the time it goes likes this:
Looking at volume reports from the networks => brainstorming angles => creating banners and landers => spending way too long uploading all creatives => spending way too much money on data => pausing the offer because I failed => cry a little => realizing I learned nothing => cry a lot

This is one of my failed campaigns.

Traffic source: Decisive (also waiting for other sources to accept the campaign)
Ad structure: wifi/app, wifi/site, mobile/app, mobile/site.
Bidding: I bid the average price as stated in the adplanner.
Flow: banner -> relevant LP -> Play store (I started testing DL later on).

Region: Brazil
Offer: UC web (android app - browser). Splittested between MundoMedia and ClickDealer.
Tracker: Voluum
Hosting: hosted on rackspace CDN for speed

Banners


Landers





Stats



If anyone knows what I'm doing wrong:
1) please slap me for spending 300 times the offer payout...
2) give me your advice please

One day, I will have a laugh about this thread.

---

You learned at least how to design some banners and landing pages, and you learned that you should have watched your data like a hawk sooner, or set a better budget cap

Just an observation: according to Wiki, only 7.9% of Brazilians speak english.
Maybe you should try to translate your creatives?

---

I tried this offer as well and didn't success to make it with a positive ROI it's ok most of the campaigns will fail . just keep testing

---

Sorry for your failure, but I am running the same offer as my first mobile campaign and have 500 clicks to the offer with ZERO conversions. Maybe it is the offer? My banners are in Spanish and direct linked, so even with the native language I have ZERO conversions.

---

Brazilians speak Portuguese dude.

OP: Did you speak to your AM before pushing the offer? I personally would go with a Hot offer in the beginning

---

Yes I did speak to my AM.
They said it has potential. Maybe thats a synonym for 'it sucks'

Interesting insights about the language! Thanks for all the replies.

---

Know your audience.

This campaign should never have launched in English or Spanish.

For Brazil, you will want to make all banners and landers Portuguese and the offer should be in Portuguese as well.

---

But given the amount of clicks to the offer that both he and I had, shouldnt more than one of them had converted?

---

Would you convert on an offer presented to you in Swahili ?

Also mm just a quick note.
Have you tried the save money angle ? I dont know about the Brazil mobile telco market with regards to data charges,
but perhaps an angle - save money less data charges could work ?

Also the FB banner if that was lets say 3 years ago it would be a bit invalid, since Orkut was the dominant social network in Brazil untill 2012
http://mashable.com/2012/01/17/faceb...-orkut-brazil/

And on another note http://www.emarketer.com/Article/Bad...Brazil/1010436
youtube no2 ? the load videos faster angle looks promising ? Its a more specific benefit than the general net being faster..

---

I wouldnt have clicked on it. I have friends from Brazil. She said there is a large population of English and Spanish speaking people that live there. I assumed the Engish speaking people were clicking on the English ads and the Spanish speaking people were clicking on the Spanish ads.

Still, the offer page is in.... Oh, I get it now. I bet the offer page is in Portuguese, so if english or spanish speaking people click through to the offer, the offer itself isnt in their language. Did I just learn a valuable lesson?

---

@finishseo

I target a lot of smaller eu countries and occasionaly I mess up the geo targeting, the banners lp and offers are for country / language 1, and the geotarget i set in broker is for country 2. What I observed is that the CTR of the mistargeted banners is not much different than if it was properly targeted - people still click the banners even tho they dont understand the text or barely understand it. More interestingly even the prelanders get clicked thru the entire funnel. And the click discrepance / performance is only around 50% lower than if the campaing was properly geo-ed with the adbroker. It all looks like a normal campaing launched on a bad day or with a bit larger % of bad siteids. Untill the actual conversion data is checked , which is 0 .

Just trying to say that from my data.. the assumption that the banners will not be clicked if not in native language is wrong.

---

As others pointed out, use the language of the targeted GEO.

---

im targeting brazil as well and just like any other country the use of their native language produces much better results. think about what the user is thinking if he sees an ad in english, compared to an ad in his native language. the english ad probably tells him something like, oh this ad is from united states and they are just showing it here, does not matter to me. but if he sees it in portuguese, he will probably think, not that many people speak portuguese, probably targeted for brazil? lets see??

the numbers i got were around a 3x higher conversion rate, my data is limited... but thats the same result my friends got.

hope it helps

---

How did you build your landing page?

---

Coded by hand. My main income is a job as a webdeveloper.

---

True. That was the main mistake.

However I still think the offer sucks.
Installs are way too low.
I'll talk about it with my AM.

---

OK, a few suggestions:

1) Watch your spend more closely next time Having said that, spending $90 to test an offer isn't too bad.
2) Yeah, the language thing. Portugese and Spanish are not the same by any means, and that's probably why nothing worked. Remember not to make that mistake again!
3) All your banners use a very similar design approach - text in the middle, small images to the left and right, highly saturated (or white) single-color background. Definitely vary the styles a bit more to see what converts best!
4) Likewise, your landers are very similar - test more widely! Test something with very little or no text, test something highly interactive, test something much more informal, etc.

Hope that helps!

---

I think with these results the campaign is a bit hopeless. It should be worth though using the work you already did and apply it on a different offer. Try another browser in Brazil and update the language + browser brand.

What I usually do when testing these type of offers is that I start by DL from banner to offer. I create 5 different banners with the same design but different angles (I might trow in some random additional banners which I have good experience with). The purpose of this is to see which angle generates clicks and conversions.

If the offer doesn't convert at all then I'll either try a different banner style or just move on to a new offer depending on how much I trust in the potential of the offer. When I do get conversions I'll create more banners (sub-angles & new designs) and I'll then create LP's.

By doing this you can minimize the time spend on making creatives. I would probably spend max $15 if the campaign has 0 conversions (considering CPA = $0.37). No conversions = no chance to optimize

---

Try another browser offer like Dolphin.

---

Well, I went back and ran the offer in the CORRECT language. So far, I have 300+ clicks and only 1 conversion. So maybe the two of us picked a bad offer to start?

---

Maybe.

Yes man... 2/3 of the population (probably more like 95% of population) DO NOT speak english.

Blame it on our horrible education system.

Being from there I can tell you that english banners are not going to work.

Also the offer doesn't seem to be very interesting. I doubt people are dying
to download a mobile browser. Logically you would have to talk to your AM
or test it out.

---

I don't think my AM likes me. He said the offer was worth a shot and told me to figure out traffic on my own.

---

@fishinseo - AMs are not always right. I like and trust several of my AMs quite a bit, and they've still recommended me plenty of offers that I ended up shooting in the head because I couldn't get them to work (They've also recommended a fair number of gems.)

Try another offer. He'll like you plenty more when you're running $x,xxx a day

---

From what I hear this offer actually does quit well. (Do not take my word on it however)

**just some brainstorming here***

The angles you use here seem a bit to soft and I would always recommend a very strong reason why they should get XX.

- You probably lose out on 90% of the people that don't know what 9gag is or simply don't care.
- You probably lose out on a ton of people that are fine with the speed of FB.

Is 9gaga or FB a high enough incentive?

It's all about finding that right click/flow. (You are on the right track here).

With this offer I would ask the following questions.

- What would make people download a random browser when they are most likely fine with the one they have now.
- How do I create interest for a broad audience. (Broad angles that could work for a bigger audience and not just 9gag, FB etc)
- Could I scare the visitor into thinking they really need this browser? (Lol, scare tactics work if you provide them with the solution)

From there I would create much more aggressive angles then the ones you have now.

For example:

Instead of: Facebook loading to low? blabla get this browser.

Go with something like this:

Your Smartphone is running 10 times under the speed it SHOULD be running! (Some nice slow/speed compare graphics)
Your favorite websites could be loading 10 times faster then they do now! Fix this in seconds!
Speed up your INTERNET Times 10 with this simple trick.

The landing page:
A fake scanner that shows them exactly how slow their phone is browsing the web and how much increase they can make by downloading OFFER XX.

or even more aggressive:

You are running a outdated browser version and SHOULD upgrade to protect yourself from *scare tactics*. (Speed, Hackers, Virus)
You have been infected with (1) virus, please upgrade your browser now!

The landing page:
A simple download button, timer to rush the fuck out of them and possibly a nice javascript alert script to let them know once more they are in risk with their not updated browser version.

- Everyone wants to browse the Internet Faster.
- Everyone SHOULD be scared for a virus, hacker on their phone.

You see that these angles would target a much broader audience with a much stronger reason why they should get it.

I def agree with keepthecar on running this offer (and most app installs) direct first with a couple of good banners to test it.

Very nice setup tho mate! Tons of banners, angles and nice LPs* You are on the right track here and if the more aggressive angle does not work get going on a new offer and you will certainly succeed.

---

Premium advice!
The whole post went straight to my evernote

Thank you mvaneijk

---

Nice write up, while I don't disagree -- when it comes to small ad sizes like 320x50 you cannot cram in too much info. As people are lazy to read, and unless your first few words catch their attention they'll just stop looking so these:

Your Smartphone is running 10 times under the speed it SHOULD be running! (Some nice slow/speed compare graphics)
Your favorite websites could be loading 10 times faster then they do now! Fix this in seconds!
Speed up your INTERNET Times 10 with this simple trick.

can be written shorter.. ie:

Speed up Phone by 324%!
Make phone 10x faster!
View websites 10x faster!
Turbo Charge Your Internet 10x!

---

There ya go! That's already much better.

Just providing some examples which can obviously be done ten times better as you just showed us! Plus 1+

---

This looks like very dirty marketing.
Not pretty sure if the sources and the offer owner will allow that

But I'll give it a shot!

---

Fixed it for you.

---

Maybe we are not have any magic relations,Because we know you so late.
UC Browser is really a good offer , and I ran it in Brazil before, I believe that it can bring you good returns.
I won’t show you my CR, ROI , because I ran it on Face book .
According to your data, I think,
1, the quality of traffic on Decisive can not be compared with Face book, but you can analyze which location will be relatively good by tracking system.
2, a lot of friends said that, BR speak Portuguese, not English, not Spanish. Most of Brazilians don’t understand English at all. So your banner and LP is a failure.
3, UC browser are emphasizing that it is free around the world, actually it’s not important. You can try to highlight other creativity.
4, I do not know whether you are using a CDN to speed properly or not , in a view of our case on August, Brazil is very special, even the direct link from the banner to play google, CDN also can not solve the problem of the traffic loss in Brazil, I solved this problem by adding a Brazil server, Perhaps you can have a try.
Finally, share my experience to you. As the final download page is play google, you can also add the element of play google on the banner. After all, there are three jump times , you can eliminate the psychological preparedness of customers.

---

@keepthecar thanks! that's a good strategy, I'll try that. Do you create 4 campaigns for each of the 5 banners like wifisite, wifiapp, mobilesite,mobileapp? how do you benchmark the winning banners?

---

Normally you'd put all 5 banners in a single campaign, but create different campaigns for wifi / app, wifi / site, mobile / app, mobile /site. So you should end up with 4 campaigns each with 5 banners in them.

---

Thanks caurman. I actually created a more detailed post about it in the newbie q & a section. I have 23 well researched angles that id like to test out.

---

Home > Mobile > Follow-along Campaigns