Ok, this is very noobish, but for some people even more noobish than me, it may be a help. Heck, I just figured it out after 4 months. So here goes.
You want to try two different versions of your made-for-adsense/arbi page to see which one works better. So, first make the two different versions. Don't make too many changes, better to make one simple change so it's easy to see what change is making the difference (basic scientific method, heh).
You could try Lerchmo's rotator, but that has about a dozen different pages and about a mile of code in it. Who can figure that out. Better to do something easy for now, until you get smarter.
So you have two different, very slightly different pages. Name one temp1.php and the other temp2.php. Make sure that you use two different adsense codes that have different channels, call one channel temp1 and the other temp2 (actually name the channels anything you want, it doesn't matter).
Now for the rotator. It's super complicated:
====
<?php
define ('TEMPLATE', 'temp' . rand(1, 2) . '.php');
include(TEMPLATE);
?>
====
name the rotator "index.php" and put it on your site. get rid of index.html because i think that will get used instead if it is there.
all this little php code does is define TEMPLATE randomly as either temp1.php or temp2.php, using the rand() function which can return either 1 or 2. then it pulls in either temp1.php or temp2.php to become your page. simple huh?
now after you get a decent amount of clicks go to :::SplitTester.com::: and put in your numbers and you will see if one page is doing better.
Not sure if above link will work, I may not have enough posts to put in a link. But it's a page called splittester dot com. It uses some math to figure out whether your results are statistically significant or not.
then you can make another "improvement" on the winning page and test again. make sense?
let me know if any criticisms of this method, it seems to work for me i've been using it for a few days successfully.
incidentally my first test was a great success as i tested something i thought was a no-brainer to be an improvement, yet when i tested it the original non-improved page worked better. so that's a lesson to test everything, even if you think you know the answer.
by the way this obviously won't work unless your server has php. if you need help figuring out whether your server has php, let me know.
You want to try two different versions of your made-for-adsense/arbi page to see which one works better. So, first make the two different versions. Don't make too many changes, better to make one simple change so it's easy to see what change is making the difference (basic scientific method, heh).
You could try Lerchmo's rotator, but that has about a dozen different pages and about a mile of code in it. Who can figure that out. Better to do something easy for now, until you get smarter.
So you have two different, very slightly different pages. Name one temp1.php and the other temp2.php. Make sure that you use two different adsense codes that have different channels, call one channel temp1 and the other temp2 (actually name the channels anything you want, it doesn't matter).
Now for the rotator. It's super complicated:
====
<?php
define ('TEMPLATE', 'temp' . rand(1, 2) . '.php');
include(TEMPLATE);
?>
====
name the rotator "index.php" and put it on your site. get rid of index.html because i think that will get used instead if it is there.
all this little php code does is define TEMPLATE randomly as either temp1.php or temp2.php, using the rand() function which can return either 1 or 2. then it pulls in either temp1.php or temp2.php to become your page. simple huh?
now after you get a decent amount of clicks go to :::SplitTester.com::: and put in your numbers and you will see if one page is doing better.
Not sure if above link will work, I may not have enough posts to put in a link. But it's a page called splittester dot com. It uses some math to figure out whether your results are statistically significant or not.
then you can make another "improvement" on the winning page and test again. make sense?
let me know if any criticisms of this method, it seems to work for me i've been using it for a few days successfully.
incidentally my first test was a great success as i tested something i thought was a no-brainer to be an improvement, yet when i tested it the original non-improved page worked better. so that's a lesson to test everything, even if you think you know the answer.
by the way this obviously won't work unless your server has php. if you need help figuring out whether your server has php, let me know.