Expanded Text Ads: 7 Million Clicks Say They’re Underperforming

By JD Prater | @JDPrater | Head of Social Media at Hanapin Marketing

On Tuesday, AdWords released one of the biggest updates in the history of the platform with expanded text ads. If you haven’t seen all the chatter across the web then know that you can now create new ads that have (2) 30 character headlines, a single 80 character description line and two breadcrumb fields to enter extra keywords in your display URL called Paths.


Etx Text Ad Example


AdWords has been working on releasing this update since April after the beta got delayed to May and then finally released in June. Over the past 6 weeks, we have been lucky enough to experiment with this new ad format and frankly, I’m not impressed with their performance.


A lot of advertisers lit up the Twitter and LinkedIn feeds with blogs discussing best practices or how to set them up. I’m wondering if they’re worth all the hype. Don’t get me wrong, this is a big roll out as text ads haven’t changed in a long time. Plus, according to The SEM Post, advertisers will no longer be able to create or edit standard text ads as of 10/26/16. So I get it. However, the results are lacking.


Below is the data that I’ve seen from over 7 million clicks worth of data from June 23rd through July 24th.


Here are a few things to keep in mind about this data:


  • This is a lot of data! However, it’s only from one account.
  • Every ad group has active desktop ads, mobile preferred ads, and expanded text ads
  • The campaigns are set to optimize for clicks (this is important to remember)
  • This data strictly analyzes CTR for each ad type across devices as the KPI


Performance By Ad Type


On the surface, the results don’t look too bad. Expanded text ads show a higher CTR than desktop ads, but lower than mobile preferred ads.


Overall performance


However, we don’t want to stop here. We need to analyze how they’re performing across difference devices since ETAs are capable of showing for desktop, mobile, and tablet.


Performance By Device And Ad Type


Let’s break down each ad by device to better understand what’s really happening beneath the surface.


Ad Data by Device


We begin to see some performance differences bubbling to the surface. After reviewing the data in this pivot table, it’s pretty easy to spot the winners on each device.


  • Desktop ads have a 5.15% higher CTR than expanded text ads
  • Mobile preferred ads have a 5.58% higher CTR than expanded text ads
  • Expanded text ads have a 0.31% higher CTR than tablet ads


Performance By Ad Type And Device


Another way to slice the data is by device. This view provides a high-level understanding of the volume these new ads are getting in the auction compared to other ad types.


Data by Ad Type


We can see that AdWords is giving these new ads a lot of weight in the auction, especially when optimizing for clicks. They’re getting around 39% of total impressions.


percent of impressions


Below is a breakdown of expanded text ads volume by device. Desktop is at 49% with mobile at 42% and tablet at 9%. This is line with the rest of the account so nothing out of the ordinary here.


ETA Device Volume


However, when we break down performance by week is when we start to see CTR slowly declining. In the first week, they earned a respectable 29.95% CTR and peaked at 31.50% after two full weeks. Then they start declining in week three to 28.23% and further down to 26.30% last week.


ETA By Week


I’m guessing these declining results are probably due to Google’s optimizing for clicks.




Is it too soon to say they aren’t working for this account? Are these just marginal fluctuations that we should we ignore? After doing some marketing math, it looks like our opportunity cost was 146,606 clicks over the last 30 days. Meaning, if we could have received clicks at the CTR then we would have seen a 2% bump in clicks.


It’s interesting data for sure! I would love to hear your thoughts, opinions, and your results. Are you seeing good results? Hit me up on Twitter (@jdprater) and let me know.


Related Posts