Tip:
Highlight text to annotate it
X
Google AdWords Campaign Experiments
AdWords Campaign Experiments makes it easy to test your ads.
For example, it’s easy to add experimental text ads that have variations on messaging, headlines, display URLs, and calls to action.
For display ads, you can similarly test experimental ads, which include variations on your images, calls to action, or display URLs.
And now for the first time you can test the performance of text ads against the performance of your display ads on the Google Display Network.
To begin testing your ads, first you’ll navigate to the 'Ads' tab of your campaign.
Then you’ll click the status icon to the left of the ad.
You’ll click the empty beaker entitled 'Control only' to make your ad active in the control split only.
Click the half-full beaker entitled 'Experiment only' to make your ad active in the experiment split only.
If your ads are set to run in the experiment or control split only, you can get a holistic view of your experiment performance by looking at your stats at the campaign level.
If you have ads set to run in both the control and experiment splits, you can view statistics broken out by split for each ad.
To do this, just click on the 'Segment' button and select 'Experiment' from the 'Ads' tab in your campaign.
I mentioned before that on the Display Network, ad experiments allow you to test the performance of including only text ads in an ad group versus only display ads.
Let's look at how to run this type of experiment.
First you'll need some display ads in your campaign.
If you don't already have display ads, one easy way to create them is with the Display Ad Builder.
Once your display ads are ready, you can follow this example to set up an experiment:
First, click the check boxes next to each of your display ads.
Then, click the 'Change status' drop down and select 'Experiment only' so that your display ads will run only when the experiment split is active.
You can try just one display ad to start, or you can include many different display ad variations.
You’ll follow a similar process to change your text ads to 'control only'.
A variation of this experiment would be to see leave your text ads running in both the control and experiment splits
with display ads still running only in the experiment split.
This allows you to measure the incremental value of having both display and text ads in an ad group.
One thing to note is that your ad rotation settings, found on the campaign settings tab, will still affect how your ads are served if you have more than one ad in each of the control and experiment splits.
To learn more about AdWords Campaign Experiments, watch the other videos in this series or visit the AdWords Help Center.
Happy experimenting!