Here is a handy method for a simple A/B testing for CB affiliates landing pages.
In this example we have an affiliate link to ClickBank product/order form, and we want to test 2 different versions of that page, to see what works better (aka converts).
This is "free" method with Google Analytcs (Additionally, you can use Optimizely, CrazyEgg e.t.c)
OK, lets say we want to test what title will convert better. Create 2 pages that promote "How To Dig Tutorial" (just a product idea).
I like to test more than 2 versions at the same time, but 2 it's OK to. Remember that it's very important to test only one thing per test.
The first page looks like this:
This is the second version:
Feeling the difference? Cool...wait! don't buy yet
follow those steps:
1. In the <head> section of the pages, insert this:
<script type="text/javascript">
function Goal() {
_gaq.push(['_trackEvent', 'Goals', 'Clicked']);
}
</script>
2. Set your "buy now" links this way:
Link 1 = add at the end of the url: /?tid=version1
Link 2 = add at the end of the url: /?tid=version2
This is how we can track "hops from page" and "sales from page" via CB analytics.
3. Add to each link: onclick="Goal()" (now we can track the clicks in Google Analytics)
Should look like this:
page1: <a href="yourCBlink/?tid=version1" onclick="Goal()">Buy Now</a>
page2: <a href="yourCBlink/?tid=version2" onclick="Goal()">Buy Now</a>
4. Set the "goal" in Analytics:
a. Administration >>+New Goal
b. Choose "Custom", next step
c. Name your goal (whatever) and select "event", next step
d. Event conditions: Category = enter goals, action = enter clicked (other fields = empty)
e. "Use the Event value as the Goal Value for the conversion?" - set YES!!
f. Create Goal
5. Create A/B testing in Analytics (aka Experiment):
Reporting >> Behavior >>Experiments>>Create Experiment
Make sure to set the "Objective for this experiment" to your new Goal (created in previous step).
Follow the steps in the screen and activate the Experiment.
Make sure to run the test for 1-2weeks or at least 1000 very targeted visits.
OK, what we have here? We have GOLD:
1. We know what title makes more "buy now" clicks
2. We know what title makes more sales (by TID) (!)
3. We can make a correlation between clicks and CB hops, to see if any "downtimes" or "shavings"
4. It's very possible, that the link that have less clicks, is actually making more sales!
Analyze, choose the winner and start a new test
Let me know if it works for you )
Cheers
In this example we have an affiliate link to ClickBank product/order form, and we want to test 2 different versions of that page, to see what works better (aka converts).
This is "free" method with Google Analytcs (Additionally, you can use Optimizely, CrazyEgg e.t.c)
OK, lets say we want to test what title will convert better. Create 2 pages that promote "How To Dig Tutorial" (just a product idea).
I like to test more than 2 versions at the same time, but 2 it's OK to. Remember that it's very important to test only one thing per test.
The first page looks like this:
What You Should Know About "How To Dig Tutorial" Before purchasing it
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
bla bla bla blabla bla bla bla.
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
Buy Now - (button/link 1)
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
bla bla bla blabla bla bla bla.
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
Buy Now - (button/link 1)
This is the second version:
How To Quit Your Job And Start Digging
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
bla bla bla blabla bla bla bla.
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
Buy Now - (button/link 2)
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
bla bla bla blabla bla bla bla.
bla bla bla blabla bla bla blabla bla bla blabla bla bla blabla bla bla bla
Buy Now - (button/link 2)
Feeling the difference? Cool...wait! don't buy yet
follow those steps:1. In the <head> section of the pages, insert this:
<script type="text/javascript">
function Goal() {
_gaq.push(['_trackEvent', 'Goals', 'Clicked']);
}
</script>
2. Set your "buy now" links this way:
Link 1 = add at the end of the url: /?tid=version1
Link 2 = add at the end of the url: /?tid=version2
This is how we can track "hops from page" and "sales from page" via CB analytics.
3. Add to each link: onclick="Goal()" (now we can track the clicks in Google Analytics)
Should look like this:
page1: <a href="yourCBlink/?tid=version1" onclick="Goal()">Buy Now</a>
page2: <a href="yourCBlink/?tid=version2" onclick="Goal()">Buy Now</a>
4. Set the "goal" in Analytics:
a. Administration >>+New Goal
b. Choose "Custom", next step
c. Name your goal (whatever) and select "event", next step
d. Event conditions: Category = enter goals, action = enter clicked (other fields = empty)
e. "Use the Event value as the Goal Value for the conversion?" - set YES!!
f. Create Goal
5. Create A/B testing in Analytics (aka Experiment):
Reporting >> Behavior >>Experiments>>Create Experiment
Make sure to set the "Objective for this experiment" to your new Goal (created in previous step).
Follow the steps in the screen and activate the Experiment.
Make sure to run the test for 1-2weeks or at least 1000 very targeted visits.
OK, what we have here? We have GOLD:
1. We know what title makes more "buy now" clicks
2. We know what title makes more sales (by TID) (!)
3. We can make a correlation between clicks and CB hops, to see if any "downtimes" or "shavings"
4. It's very possible, that the link that have less clicks, is actually making more sales!
Analyze, choose the winner and start a new test
Let me know if it works for you )
Cheers
