Keyword Insertion: Is Relevancy Really In The Eyes Of The Beholder?

An Experiment into Improving Ad Performance Using Keyword Insertion

Keyword insertion, or dynamic keyword insertion as it used to be called, is a feature of AdWords that can be used in ad headlines to increase relevance to the users search query. One of the features that we’re most excited about here at upriseUP is the fact that using keyword insertion increases relevancy and dramatically increases your ads performance.

How Keyword Insertion Works

With keyword insertion in place in a headline, whenever a user performs a search, the keyword that triggered your ad to show is placed into the headline. The example Google use is if you sell chocolate you might set up a headline that reads Buy {KeyWord:Chocolate}. If Dark Chocolate is one of your keywords, the headline would read Buy Dark Chocolate. For more information on keyword insertion and how to set it up see the Google support.

The Question

By using keyword insertion in your headlines, you can create a seemingly personalised ad, resulting in increased ad relevance, increased quality scores, lower actual CPC, improved ad rank… you get the idea. They are great.

However, equally relevant to the above example is an ad without keyword insertion where the headline Buy Dark Chocolate and Dark Chocolate as a keyword. What the user sees is the same ad, the question is would Google preferentially choose an ad that it deems “more relevant” because it could insert the users query into the title? What better way to test that than by using Google’s experiment feature?

The Experiment

What I created is a single keyword ad group (SKAG) with an exact match keyword and that keyword featuring in the headline. I kept everything but the headline the same between the experiment and the original to ensure a fair test. In the original I left the headline as plain text and in the experiment replaced the keyword in the headline with a keyword insertion. Given that the only keyword in the ad group is an exact match keyword, any time it could be triggered it would insert the keyword into the headline.

The end results to the user is the exact same text, the only difference is how it got there. With the traffic split 50:50 between the experiment and the original, Google will enter each ad into the auction the same amount of times since they are essentially the same ad, right?

The Results

What actually happened is that despite the ad being the ‘same’ in the eyes of the user, the Google algorithm preferentially entered the ad using keyword insertion rather than plain text. The ad using keyword insertion generated a statistically significant increase in number of impressions and clicks, as well as an increased click through rate (although not a statistically significant one!).

Key Takeaways

However, there are situations to be aware of where keyword insertion into a title would not make sense and might make the ad read strangely, so each situation needs to be looked at critically. But in terms of getting your ad seen and clicked on by more people, keyword insertion is a valuable tool.

The key takeaway here is that by implementing keyword insertion you can increase your ads relevancy in the eyes of the algorithm without it changing in the eyes of the user, which can only help improve your ads performance.