An AdWords Display URL Experiment

Note – If you find this article interesting or helpful in any way, please feel free to ‘tip’ the author with a share on your preferred social network. Thanks!

Over the past month I have been conducting an experiment with insertion of keywords in display URLs. The typical display URL includes the keyword at the end as seen below:

Whilst this is great and makes the viewer think the ad is extremely relevant to them, there doesn’t seem to be much experimentation in the display URL realm. This got me thinking about other ways to insert keywords. Which lead me to inserting keywords as sub domains, which can be seen below:

Hypothesis

My first hypothesis was that by using the sub domain keyword it would be more eye catching to users because it was different to the majority of competitor ads.

My second hypothesis was that having the keyword shown a second time directly below the head line would attract more attention due to a larger area of bolding.

I would consider either of these hypotheses true if I saw an increase in click-through-rate.

Method

In order to attain reliable results I ran the experiment over a 30 day period. The control ads had the keyword inserted at the end of the display URL. The experiment ads had the keyword inserted as a sub domain in the display URL. The rest of the ad copy was exactly the same for both the control and experiment ads. Each group of ads were split roughly 50/50 between 100% of traffic.

Results

Conclusion

From this initial experiment it’s quite clear that sub domain keywords are a much better use of the display URL region. Click-through-rate was 6.52% higher than the standard display URL, cost-per-click was 4.88% lower, and average ad position was 1.41% lower (meaning in a better position on the SERP). These results aligned quite closely to my hypothesis.

The Unexpected Results

The best thing about running experiments is when you get results you didn’t factor in! What I didn’t factor in was cost-per-click decreasing, and average ad position improving.

My reasoning for the decrease in cost-per-click is that the quality score was improved due to the increased click-through-rate, which in turn placed our ads in better positions because our ad rank was higher due to an improved quality score.

This goes to show how important click-through-rate is in determining quality score and also how important it is to split test ad copy.

Reminder – If you’ve found this article interesting or helpful in any way, please feel free to ‘tip’ the author with a share on your preferred social network. Thanks!