Website Optimiser, Part 5: Your experiment
January 29th, 2009 | Published in Google Grants
Now that we have installed and validated our Google Website Optimiser code, we can begin creating our experiment.
You should be able to see your experiment overview page by logging into your AdWords account, clicking the Campaign Management Tab, then Tools, then Website Optimiser, then your first test setup in Part 4.
You should be able to see your experiment overview page by logging into your AdWords account, clicking the Campaign Management Tab, then Tools, then Website Optimiser, then your first test setup in Part 4.
Step 1.
You will want to add a couple of variations to each section. The sections are defined by the small bit of code you installed in part 4 under Page Sections. You will see that each section name you placed in the code appears as the heading in the left hand column.
You will be able to view the original and add new variation. You should also see a preview link which opens in a new window. This shows you the changes you have made and how they look.
If you're unsure of what you should test you might find this article from Conversion University helpful.
After you click Add new variation you will be asked to enter a name for this variation. You're limited to 25 characters, so keep it simple and short. This information is only used to help you identify which parts of which pages are working the best for you.
We called ours "Give Today!", because our new headline is "Give Thousands Hope Today!"
This however is just one variation of heading. You can literally try as many as you want. However since this is one of your first experiments we recommend to sticking with 2-3 variations.
Step 2.
Once you are happy with the variations you have created, click on the preview link to check the various combinations of the page that you have created. You can see in the example below, that the headline is changed for what we placed in our experiment.
Step 3.
So you are happy with the experiment and are now ready to continue. Click on the continue button. Be warned however that once an experiment is launched you won't be able to make any more changes. Therefore before launching double check you are happy with it one last time. If you find that you made a mistake you have a few options. The easiest is to simply pause the experiment so only your original page shows. You can also delete individual variations from your experiment. This is called pruning.
Step 4.
The final step before launching your experiment is to let Google Website Optimiser know how much traffic you wish to devote to your experiment. If you only wanted to show your experiment to 20% of your traffic, you could select 20% here.
In this instance we are going to let all traffic see the original and our variations. So we selected 100%. The more traffic you send to your experiment the sooner you will get clear results on which variations are working better.
If you are nervous about making changes to your site, you can set a lower amount of traffic say 10-20% to view your variations.
When you are ready to launch your experiment, click Launch Now
Step 5.
We are just about done. Once your experiment has launched you will be presented with an overview page where you can see the status of your experiment. Keep in mind that Website Optimiser won't show reports until a few hours have passed.
You will want to add a couple of variations to each section. The sections are defined by the small bit of code you installed in part 4 under Page Sections. You will see that each section name you placed in the code appears as the heading in the left hand column.
You will be able to view the original and add new variation. You should also see a preview link which opens in a new window. This shows you the changes you have made and how they look.
If you're unsure of what you should test you might find this article from Conversion University helpful.
After you click Add new variation you will be asked to enter a name for this variation. You're limited to 25 characters, so keep it simple and short. This information is only used to help you identify which parts of which pages are working the best for you.
We called ours "Give Today!", because our new headline is "Give Thousands Hope Today!"
Once you are happy with the variations you have created, click on the preview link to check the various combinations of the page that you have created. You can see in the example below, that the headline is changed for what we placed in our experiment.
So you are happy with the experiment and are now ready to continue. Click on the continue button. Be warned however that once an experiment is launched you won't be able to make any more changes. Therefore before launching double check you are happy with it one last time. If you find that you made a mistake you have a few options. The easiest is to simply pause the experiment so only your original page shows. You can also delete individual variations from your experiment. This is called pruning.
The final step before launching your experiment is to let Google Website Optimiser know how much traffic you wish to devote to your experiment. If you only wanted to show your experiment to 20% of your traffic, you could select 20% here.
In this instance we are going to let all traffic see the original and our variations. So we selected 100%. The more traffic you send to your experiment the sooner you will get clear results on which variations are working better.
If you are nervous about making changes to your site, you can set a lower amount of traffic say 10-20% to view your variations.
When you are ready to launch your experiment, click Launch Now
We are just about done. Once your experiment has launched you will be presented with an overview page where you can see the status of your experiment. Keep in mind that Website Optimiser won't show reports until a few hours have passed.
Congratulations, you've launched your first experiment. We hope Google Website Optimiser brings you and your organization more success! Our last part in the series will cover interpreting your experiment results and how you can continue to use testing to improve your site.