Hey Guys,
So i bought LandingPageGenius a couple weeks ago and just got started looking at how it works with CPVLab....
It looks like it doesnt allow you to use the Campaign URL that CPVlab gives you to rotate LP's and determine traffic flow to a specific campaign...
Instead it looks like you can only test one LP with multivariate testing per traffic source campaign.... Am i missing something here?
ie. with
CPVlab URL -> LP1/LP2/LP3
with LPGenius -> mydomain.com/LP1.php?keyword=%%$KEYWORD%%
Which to me means i can only use it on one LP at a time during a given campaign and that i have to provide an individual page URL... Which pretty much breaks the ease of settign up campaigns in CPVLab...
Did i about get it right?
How do you guys use this SW?
-H
No. You do have to set it up on each lander though. LPGenius tests variables on (individual) landers, so you have to plug whatever you want to test on each lander in and it will give you a breakdown of how the tested variables on each lander are performing.
Though you should know that doing anything more than an A/B test split across multiple landers will dramatically increase the budget you need to get statistically significant data as to what is performing best with your initial test.
My advice: Test the three landers you want to test against each other and see what backs out best. If you can afford it, use LPGenius to test different angle headlines (or primary images) on each lander.
Ahh, so once you find a lander that works best through A/B testing THEN you would switch over to multivariate testing to see how to improve that 1 lander....
I think you get it but your language is a little off.
By definition an A/B test is testing one variable against another version of the same variable. For instance, you could A/B test lander design A against lander design B, or headline A against headline B on the same lander. What LP genius allows you to do is multivariate testing on the same lander (and of course A/B testing as well).
Let's say you have 3 lander designs (because you said you do) and two angles you want to test initially. You can test those three landers against each other, and at the same time (if you can afford it) test two versions of your angle headlines on each lander - this effectively tests all six possibilities at the same time.
It is more costly to do it this way, rather than testing 3 different angles OR LPs initially, and many pros probably wouldn't recommend it. The upshot is that you can quickly find out which initial combination will perform the best as long as you spend hard enough, eliminate all other possibilities and optimize from there.
Read Mr Green's post on angling here: http://stmforum.com/forum/showthread...ghlight=angles
Truth be told, I like to test different lander designs and angles at the same time. But you almost always lose money faster on the front end doing it this way... however you can get pointed in the right direction faster as well.
Just be honest about your budget and tolerance for losing it before you do it this way.
Hope this helps. Be aggressive out there.