My 2nd mobile campaign on Decisive (34)
Hi All!
My first campaign was a flop. Never mind, I’ve just started a new one!
Offer Selection
I chose “United Kingdom - [WAP] Win Grocery Vouchers” offer for following reasons:
• Payout 0.80$ - so it should be easy to get data without spending a fortune
• It’s UK – I live in the UK and I’m shopping in supermarkets here. I can really relate and understand this offer. Just yesterday I spent over £60 for my weekly shopping!
• Broader appeal – Many people are buying food and want to save money on it.
Angles
I struggled a bit with angles. I didn’t want to use an infamous “Congratulations, you’re 1,000,000th visitor and you’ve just won…”. The advertiser is strict and requires sending creatives for their approval.
I pushed my imagination and came up with 10 angles. Some of them are probably not very good but some of them may hopefully have some potential. But how do you choose an angle from a list of possible angels? Are you just using your intuition? I’ve chosen “Fed up with Food Prices Going up? Win a £1,000 Supermarket Voucher”. Is it a good angle?
Banners
I’ve created 8 different banners. Just static ones at the moment.
Now I’m waiting for approvals from Decisive and the advertiser.
Results after 1 day of testing
Spend: $26.72
Revenue: $0
ROI: -100 %
Clicks in Decisive: 2,288
Clicks in
Clicks in an affiliate network: 813
Se here I am again with the same problem like my first campaign. 
65% click loss and I have NO idea why. 
Any idea what to do about it? Please, help.
Im experiencing clickloss as well, could be your server? are you using a VPS based in Europe or a CDN?
also the text for your banner is too long, try to make it under 5-6 words 
Have you create your own landers? I suggest you try this to eliminate the bot traffic, specially at post #32
I'm direct linking at the moment. So this is not a problem with my server or landers. What else could it be?
Originally Posted by peterpan
Something off the top of my head...
"Fed up with Food Prices Going up? Win a £1,000 Supermarket Voucher”.
You can make it shorter to ->
Stop Overpaying on Food!
Win Voucher!
Yep, shortening things is a bit of an art - you'll get better at it with practise, and it's practise that will pay off a lot in the long run. It's important on ads, but it's absolutely vital on landers.
Whenever you're trying to shorten some copy, try breaking the sentence down into chunks.
Are there any chunks that you can remove and still make sense?
Are there any chunks you can rephrase in less words?
Often, the bits you think are absolutely vital aren't - split-test! But keep aware of core advertising principles like CTAs, specific numbers, etc.
Also, shorter is not ALWAYS better - just usually. It's worth split-testing between a long and a short version, but 80% of the time you'll find the shorter version wins.
As far as angles: this is where research into your target market comes in.
Do you know enough about your target market to at least make guesses as to what angles will or won't work?
For example, do you know who in the UK controls the household shopping finances? How much information do you have on the lifestyle of your most common demographic group?
This stuff might sound intimidating, but it's very easy to find in the age of the Internet. A couple of hours' work reading forums, Googling, and checking out some UK newspapers will give you tons of information.
For example, check what UK magazines and newspapers offer cash-saving household tips. In those articles, who are the people mentioned and pictured? That's a quick way into understanding the likely most profitable demo.
I've decided to test this campaign with a landing page.
Stats after 1 day:
Clicks to a lander: 4,218
Clicks to an offer: 3
LP CTR: 0.07%
Conversions: 0
Over 4 thousand clicks and just 3 clicks to the offer? And I have no idea why. Can you look at the lander I've created, please, and give me some feedback? See: http://www.phselection.com/vouchers/01.html What's wrong with it?
I was not able to answer all questions correctly 
Maybe send them to the offer even when they didn't get all the answers correct.
Confession:
I tried solving the quiz ... answered 4/5 correctly couple of times, NO way to win the voucher.
Turns out the capital of California is Sacramento instead of L.A.
Other people may make the same mistake, resulting in low CTR%.
Most people probably couldn't answer all the questions.
Also, have you tested your lander on an actual mobile device?
Because the 31 KB picture you are using looks like this:
I'm no expert but:
- Make the Landing page actually look better. The image has a white background and isn't transparent. The button doesn't look high quality at all as well.
- Even if ppl fail to answer everything correctly, redirect them to the offer!!! I actually can't believe that your quiz doesn't let retards pass!
Thanks for all your feedback, guys!
I've made few changes to my lander:
- Improved the design
- Shortened the headline
- Used just 3questions instead of 5
- Showing the offer to everyone even if the failed the quiz!
Now I'm waiting an approval from the advertiser again. That sucks but hopefully could resume the campaign soon.
If you have the technical chops for it, I highly recommend putting in analytics to tell where in Javascript questionaires your visitors drop off. You can do it with Google Analytics - it's a very powerful way to save money on this sort of thing.
I just don't get it. I have done all you told me. I resumed the campaign, got 2,000 clicks and 0 clicks to the offer! 
Why? What's wrong with this lander http://www.phselection.com/winvouchers/02.html?
Have you tried to complete the quiz on your own phone?
-Your page is inaccessible without the "www" at the front of the domain.
-The lander is still not mobile optimized as the title takes the whole screen to display, which just does not look professional and makes users scroll all the way down to even see the survey.
-Each time you click on "Next" the page reloads and makes you scroll down again, which is unnecessary extra steps for users.
-You can click on "Next" without choosing an answer and it will still go to the next question.
-The CTA at the end of the quiz is just weak. Make it a big beautiful button that's clearly visible and makes people want to tap on it.
@ralphy Thanks for your feedback, mate! I've tried it on my phone and it seemed ok to me. I outsourced this lander and told that I need a responsive landing page. I guess I need to fix the issues you mentioned.
Originally Posted by ralphy
Right: if you had 2,000 clicks with ZERO clicks through to the offer, something is almost certainly wrong.
I know I keep banging on about this on the forum, but top thing: are those real clicks? The stats you describe sound like classic bot clicks to me. Are they mostly from a few placements, or spread across a lot of placements?
There are definitely lots of things you could improve about the lander (top suggestions: try a single question. Make the CTA at the end waaaaaaaaay bigger and more obvious. Much smaller title text. More appealing image. Test a background that isn't green.) but still, a CTR of below 0.05% on a lander is pretty unlikely.
We WILL fix this!
Originally Posted by caurmen
Hmm - are you getting that from the live stats, or a report? Reports only show a few placements, wheras the live stats show more.
What country is this?
Originally Posted by caurmen
It's the UK.
Originally Posted by ralphy
I bet you'll get more clicks with a simpler LP -> Headline - Congrats you won - CTA (button). Try it.
Originally Posted by tim roth
Originally Posted by phlong
I've made yours suggested improvements, resumed campaign, waited for 2 000 clicks and here are the results:
Clicks to a lander: 2,143
Clicks to an offer: 1
LP CTR: 0.00046%
Conversions: 0
I'm getting really frustrated. Can sweepstakes really work on Decisive?
I had the same confusion with Undefined placements, but if I remember correctly it's just some faulty way of describing the placements. Let me explain...
If you have a campaign and you get traffic from sites and apps as well you will get Undefined placements when you create a report. The Undefined placement under apps tab is actually the sites traffic. The Undefined placement from the sites' tab is actually the traffic from all the apps.
I'd focus on apps first. I had a harder time gathering data and finding good placements for sweeps on sites.
Originally Posted by beckslash
If you did that from the beginning and didn't have mixed traffic on the campaign, then it shouldn't. You could email decisive and ask them about that or to block all the undefined placements for you.
I also had a terrible experience with Decisive promoting sweepstakes.
I have tried to promote 3 different Grocery voucher sweepstakes offers (3 different GEOs). Each one has the same LP but in different language. THe lp is simple (quiz) style: short headline and two cta buttons. I have an event tracking setup on these buttons and out of 3,408 visitors all I got is only 3 buttons clicks.
Additionally I made an experiment to send traffic from Decisive to one of my own websites which I am currently advertising with AdWords (Display), FB, and Bing
The landing page is similar. It has a headline "Are you male or female?" and two buttons: "Yes" and "No". I usually get 78% - 84% LP CTR, but with Decisive it's almost 0. From 2,174 visitors from Decisive I have 7 buttons clicks: 0.3% LP CTR.
Originally Posted by beckslash
I plan to use them as a test to filter out bots. From what I've read on here a lot of the traffic could possibly be that. I believe there is a guide cartman may have made on the forum..
Home > Mobile > Follow-along Campaigns