Is Ad Extension Overload Hurting Performance?

white_donkey_overload_1380059_nevit

This may come as a shock, but I like things that are free and add value. That’s why I have always loved ad extensions. They allow advertisers to add more ad copy and options for searchers to interact with at no extra cost. Also, the more extensions that show with your ad the more real estate you’re taking up on the SERP (which can boost overall CTR.) This is why I was always in the practice of adding any ad extensions to a campaign that made sense for a client.

Another reason why ad extensions are important is their impact on Google’s ad rank. “Your Ad Rank is a score that’s based on your bid, auction-time measurements of expected CTR, ad relevance, landing page experience, and the expected impact of extensions and other ad formats.” So it’s commonplace for advertisers to throw all of the applicable ad extensions in their campaigns.

It’s important to know that just because you have ad extensions in your campaign doesn’t mean Google will always show extensions every time your ad appears. Google shows extensions with your ad when it calculates that the extension will improve your performance. If you have multiple ad extensions, it’s up to Google to decide which combination of available extensions will show.

Extensions.. A Negative Impact?

What got me thinking about the potential negative impact of running too many extensions was a conversion decrease we were seeing for one of my lead gen clients. They generated leads from both on-site forms and calls, so of course we were running call extensions. We were also running sitelinks, callouts, and structured snippets. I noticed over time that as more extensions were being added into our campaigns, we were receiving fewer click-to-calls on our ads. This got me thinking about whether having so many ad extensions was actually hurting our performance.

Here’s a breakdown of the initial impression share for each active extension:

AdWords Ad Extension Impression Share

Remember, it’s up to Google how often and in what combination to show each extension and they always seem to favor sitelinks – one of the OG ad extensions. Call extensions showed about 70% of the time when our ads were displayed, but that’s still 30% of the time when a call option was not available to people seeing our ad.

Was Google choosing to show our call extensions less often based on it’s own “performance improvement calculation” because we were giving it more options? If so, could removing other extensions increase the impression share for call extensions?

To test out this theory, I decided to remove all extensions from our ads except for Call Extensions (obviously, because we want to generate calls) and Callout Extensions (because we call out key competitive differences).

Call Extension Impression Share

By removing Sitelinks and Structured Snippets, impression share for Call and Callout Extensions increased. Phone numbers were now showing with ads 84% of the time and Callouts 77% of the time.

Now, you might be thinking, “Wait a minute. You removed sitelinks and structured snippets from your campaigns. Why are they still showing impressions?” Ah, excellent question. That is because of AdWords’ automated extensions. For sitelinks and structured snippets, AdWords can choose to create them for you dynamically. You can opt-out of automated extensions under the Ad Extensions tab at the Account level. (See screenshot below)

AdWords Automate Extension Opt Out

Impression Share & Call Extensions

So clearly, removing less valuable ad extensions from your campaigns will allow your top priority extensions to show more often with your ads. But, did increasing impression share of Call Extensions have any impact on performance?

Here’s a look at the performance changes before and after limiting extensions:

adwords ad extension clicks

As you can see, after my changes, a larger percentage of overall ad clicks went to call extensions.

Before limiting extensions, only 12% of all ad clicks were on the call extension, 13% of clicks were on sitelinks and 75% were clicking on the headline. After my changes, a full 41% of total ad clicks were happening on the call extension. Not only that, but having fewer extensions actually led to people clicking on extensions in general at a higher rate (25% vs 41%). So not only did call extension impression share increase, but so did its share of clicks.

Call Extension CTR and Ad CTR Improvements

As you can see in the table below, both call extension CTR and overall ad CTR improved.*
Call Extension Performance

*90% confidence level

This was interesting, because it wasn’t as though CTR for the ads as a whole was remaining steady with just a higher portion of clicks going to call extensions. We were actually generating more clicks on the ads overall while running fewer extensions. Limiting the number of ways a searcher could interact with our ad and reducing the information in the copy was making our ads more appealing.

An increase in CTR is great, but again, clicks are Google’s KPI – not ours. The affect on conversion was what I was most interested in.

Conversion Rate

My findings showed that more conversions were generated at a higher conversion rate. For this client, we define a converted call as one that lasts longer than 60 seconds. The form and call conversion rates remained steady before and after the change. However, because we were driving more clicks to higher converting call extension overall conversion rate for the ads increased. (See table below)

AdWords Conversion Increase

Ad Rank

Despite these changes, ad rank was not impacted.

Average position remained the same before and after removing extensions, only shifting from 1.4 to 1.5 comparing the 20 days before and after (Note: no bid changes were made during this time.) Impression share also remained steady at 94%.  AdWords did not seem to be limiting our ad exposure after removing extensions.

Of course, this is just one example from one account. But if you’ve been in the mindset of “the more ad extensions, the better,” I encourage you to review your campaigns. Distinguish between the must-have vs nice-to-have extensions, and test running your campaigns with only the must-haves.

Just because you have all of these extensions at your fingertips, doesn’t mean you should use them all.

What are your must-have ad extensions? We’d love to hear your thoughts in the comments!

 

Comments (15)

  1. Hi Bethany,
    This is a very interesting article. When you tested this, did you create an experiment campaign (the relatively new feature Google released a few months ago), or did you just removed them from the campaign? What was your reasoning?

    Also, how do you easily view the impression share data by extension type?

  2. Hi Ben,

    Thanks for commenting!

    I didn’t use an experimental campaign only because we needed to make performance changes quickly. I do plan on testing this out in other accounts with drafts to hold as many variables constant as possible. I’m excited to see the results.

    For the extension impression share, that was my own calculation. I took the impressions for ads when each extension was showing (from the ad extensions tab) and divided that by the total number of ad impressions for the same period.

  3. At first nice blog to help out about website loading time decreasing that when the ad extension increases the speed of loading the site from the server and it leaves a negative impression on your site and about the ctr and ad ctr improvement .

  4. Hi Bethany,

    Thanks for the response! When you say “drafts” you mean you will plan to create an experiment campaign with the one changed variable to see if you see the same results?

    Regarding the impression share calculation…You just summed up the total number of impressions for each of the ad extensions (on the ad extensions) tab and divided by the total number of impressions on the campaign view for the same period, is that correct?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.