Home > Paid Traffic Sources > Mobile

Tried the Survey method for the iPad giveaway (9)

03-20-2013 11:40 AM #1 localdentist (Member)
Tried the Survey method for the iPad giveaway

So I wanted to test my luck with one of those ipad giveaways.

Traffic source: Airpush, CPC 1c
Offer: WAP Czech Republic Win an iPad, $4.10/CPL
Lander: Pinnion.com Survey

Offer LP:

Direct linking to the offer gave me:
$8 Rev, $5 exp, $3 profit, 500 clicks, 2 leads

After speaking with my AM, he kept saying all of the affiliates on the network are running LP's, and this would increase my CVR. Furthermore, 1c traffic only drips in, and at 2cents, I would be at a loss, so an LP is required to pre-sell.

So, I made a survey:

Intro, 3 questions, Congrats page
Intro:
Answer this short quiz to win!
We only have 3 left. Hurry!
Tap Begin Now

Questions:
1. who is the current prez of Czech: Havel, Zeman, Klaus (used wikipedia)
2. What country is neighbors with Czech: Germany, France, Spain (CIA fact book)
3. Who introduced the ipad in 2010: obvious

Congrats + Link
Congratulations!
You qualify to win a new ipad!
--> Click Here Now <--

Ran some more traffic, and got this from the pinnion Survey:

As you can see, the 'mobile' version of the survey converts well, while the web version dies off.

Looking at the 'clicks out' stats, I don't see a correlation in what phones have a good LP CTR.
(using Airpush's tokens: %device%, %manufacturer%, etc)

Some research into this (I don't know if airpush has %OS version tokens), reveals that CTR's happen on Android 2.3+. But I would need more stats on this.

No conversions yet, but clearly there are certain phones that work better with pinnion than others.

/food for thought.

03-20-2013 11:56 AM #2 deondup (Member)

Your lander/angle should work (check with your network if its okay - you might need to change the wording slightly)

Airpush = Android traffic

Start by optimizing carriers, bid higher and get a lot more traffic before making decisions

03-20-2013 12:03 PM #3 localdentist (Member)

Originally Posted by deondup

Thanks for the heads up. I mean what version of android the device is running.

Maybe the non-converting 'web-survey' is anything below android 2.3, but STM tracker doesn't track this, and I don't know if airpush has a token for that yet.

Upped my bid to 2c to test further.

03-20-2013 03:25 PM #4 localdentist (Member)

So I've decided to pause this survey method because the platform, www.pinnion.com, doesn't seem to support mobile too well.

As you can see, lots of the mobile traffic is being directed to the 'web' version, which doesn't work on the survey.

But the survey itself works quite well. 40% of visitors complete the survey.

Here are the stats for the Android 2.3+ only traffic, Clicks, and click-throughs

So 40% of users (who saw the actual survey) completed the survey.
Also, 60% that completed the survey, also went to the offer LP.

So, if my math isn't that terrible, we have a 24% LP CTR with this survey.

Here is some carrier data for CTR's.

For the time being, I have a small list of devices that work with this survey platform, but that doesn't mean these devices convert.

Gonna try survey monkey, or another mobile survey platform.

Things to also try: post-popup, google analytics tracking

03-20-2013 11:05 PM #5 chize (Member)

You could always make a survey style lander yourself using jQuery mobile. Then you wouldn't need to worry about users being redirect to the web version and you can make it so every answer is correct .

03-24-2013 10:52 PM #6 localdentist (Member)

Originally Posted by chize

You could always make a survey style lander yourself using jQuery mobile. Then you wouldn't need to worry about users being redirect to the web version and you can make it so every answer is correct

I purchased a 1 month membership to the Codiqa builder tool. Will test out a jQuery lander soon

So an update:

I spent a total of $95 on this lander style, and got $51 in revenue. Net loss, but not bad for a first try.

I need to improve my recording/accounting to better understand where the money was doing well, and where it failed.

Some key points tho:
$95 Expenses
$51 Revenue
$44 net loss

3 Leads were from my initial Czech ipad offer. $4CPA

But then my CPA network started redirecting traffic to other offers (didn't disable this), and I started getting better conversions on another offer.

7 Leads on a Czech Ipad mini offer. $4CPA

Unluckily, the first couple conversions didn't track properly using the STM Tracker.

Lots of 'generic', SDK's, and unorganized carrier data.

I implemented Airpush tokens after the 3rd conversion instead.
%carrier%
%manufacturer%
%device%

At the same time, I implemented 2 additional questions to the survey.

I also decided to create a spanish lander for a $10CPA... 1 Lead, terrible conversion rate.

Since then, my p202 data has magically disappeared... not sure why.

But otherwise, I need to try a new jQuery LP, and generic survey relative to each country.

03-24-2013 11:08 PM #7 maynzie (Moderator)

Keep running till you see conversions, then cut the total junk Mobile costs a fair bit in beginning testing

03-25-2013 03:06 PM #8 localdentist (Member)

Anyone know which is more effective... a survey or a quiz?

03-25-2013 09:20 PM #9 Mr Green (Administrator)

Surveys have worked better for on mobile for me. I'm guessing because they look more serious, and bring a certain level of trust in contrast to quizzes.

After looking over your lander I'd recommend restructuring your lander to make people feel more like they deserved to win the iPad. Would you be suspicious about winning an iPad after 3 piss easy questions? Make three tests, a easy quiz, medium quiz, and a hard quiz. That should give you some interesting data to play with.

Home > Paid Traffic Sources > Mobile