Increase your Adsense CTR by using this simple testing method

This article is all bout my simple and free method of Google Adsense rotation and testing in HTML without javascript or php. It uses SSI to let you rotate adverts on HTML pages with no complicated banner roation software.

You can read a lot on the Internet about what Google Adsense formats, layouts, colors and fonts will lead to the best click through rate. Often these rules are contradictory, and often screwed up by perculiarities of specific sites. It is no good taking somebody else’s word for it: The only way to get your CTR increasing is to test different ad formats on your site and find which one has the best click through rate.

Of course, this is very difficult and tedious to do in practice. I could have all my adverts in red on Monday, and blue on Tuesday etc. However, as everybody knows, there is a spike in traffic on most sites on a Monday.... and weekend traffic is so very different to weekday traffic. No, manual testing is boring, tedious and does’t work to tell you anything useful about your Google Adsense click through rates.

Therefore we need something simple to server up different adverts at random to different visitors. Whilst Google’s website optimizer can sort of be used to do this, it is a bit of a hammer to crack a walnut. It also relies on a lot of javascript and is frankly a bit messy.

It is so much neater to use server side includes and a bit of cunning. Getting SSI to generate a random number is probably not possible, however, you can get the time. All we need to do is use a function of the current time to switch between different adverts. For example, if we want to change between 10 different Adsense advert formats, then we can use the last digit of the time – which changes every second.

First of all, you need to understand server side includes . SSI is a very basic scripting language. Go and read here http://en.wikipedia.org/wiki/Server_Side_Includes

The basic plan is to make 10 text files, each including the Adsense code for 10 different variations of your advert, and use SSI to switch between them. Because it is server side, the visitors will be none the wiser.

So...

1)In your Google Adsense account, create 10 new Adsense adverts, and make sure each one has its own personal channel. This is vital, otherwise you will not be able to track the best advert format. For example, you might setup 10 variants in different colours. You create a channel for each variant (e.g. test red, test blue, test green etc).

2)Take the Adsense code and save it in a text file on your website server, and call it something sensible e.g. test_advert_red.txt, test_advert_blue.txt, test_advert_green.txt etc.

3)Now, on your HTML page where you originally had the Adsense code, you replace the Adsense code with some SSI script which selects one of the text files to include based on the current time.

All this does is rotate the adverts based on the current time. If I visit your website at 16:23:45 I will see the orange adverts. If I visit at 20:32:19 I will see the purple adverts.

4. As the visitors come and go you can watch your channels in your google Adsense account. Check it is working by visiting your site and refreshing a few times, you should see the Google Adsense adverts rotating as the site automatically tests different ad layouts for you.

It is vitally important that you create a channel for “testing red adverts” and put it in the text file for “testing red adverts” and use a different text file and channel for each variant. Of course, if you want to test less variants, for example 5, then call each variant twice e.g.

It is only be each file having its own channel that you can track the changes. As you get visitors to your site, they will be served up different adverts, seemingly at random. When you view your channel data in your Google Adsense account, you will see the impressions for each channel going up pretty much together. All you have to do is watch the CTR click through rate.

To decide on the best advert you need a minimum of 100 clicks on each variant. Any less than that, and the odd rogue visitor can click 10 times on the same advert and skew the result. 200 clicks per variant or more is safer. For most sites, you can happily run this experiment for a couple of months and get a very fair test.

Once you have decided which is the best advert, just replace the large block of code with a simple

This means all visitors get the winning advert. Of course, you can carry on testing and testing forever. If you decide that Pink Skyscrapers are the best CTR for your site, you can set up another test of 10 pink skyscrapers in different fonts and font sizes.

It is often surprising what these tests reveal. Sometimes wild and wacky Adsense layouts make for good click through rates on some sites. Do not believe everything the so-called experts speak and write, the only way to reliably increase your Adsense click through rate is to test different advert layouts automatically using this sort of script.