A video in this course demonstrated how to perform an ad variation test in Google Ads. This is one type of A/B test. Google Ads allows you to test variations for text display and responsive search ads. This reading reviews the steps to set up an ad variation test and introduces another type of A/B test in Google Ads, one that tests automated bidding strategies so you can decide which one is best.
Note: All tests are performed using the Drafts and Experiments feature in Google Ads.
Table of Contents
Steps to set up and test ad variations
Here is an overview of the steps to set up and test a text or responsive search ad variation in the Drafts and Experiments section of Google Ads:
- Decide if the variation is for all campaigns, or for certain campaigns only.
- Select that the variation is for text or responsive search ads.
Note: Starting June 30, 2022, you’ll no longer be able to create or edit expanded text ads in Google Ads. You should use responsive search ads instead. - Filter ads by the text ad component that will be tested, such as headlines or descriptions.
- Modify the text for the text ad component to be tested.
- Name the variation.
- Choose the start and end dates for the A/B test.
- Configure the split; almost always 50%.
- Save the settings. The test will automatically run on the start date.
Text ad variations
The following are descriptions of components of text ads. You can create a variation for any of these components and test it against the original version.
Final URL
Also known as the landing URL, the final URL is the URL for the page that people reach after they click an ad. If your original URL is www.website.com/members, one possible variation could be www.website.com/rewards.
Final mobile URL
The final mobile URL is the URL of the page that users reach after they click an ad from a mobile device. If your original URL is www.website.com/m/members, one possible variation could be www.website.com/m/rewards.
Headline
A headline is the text that appears at the top of an ad. Google Ads allows up to three headlines with a 30-character limit for each. If your original headline is “Act now while supplies last,” one possible variation could be “Huge savings, limited time only.”
Display path
The display path is the URL that appears under a headline in an expanded text ad. Google Ads allows up to two paths with a limit of 15 characters displayed for each path. If your original display path is www.website.com/mens/shoes, one possible variation could be www.website.com/shoes/men.
Description
A description appears in an expanded text ad. Google Ads allows up to two descriptions with a 90-character limit for each description.
Suppose your original description is “Top athletic, outdoor, casual, and dress shoes. Free shipping on purchases of $75 or more.” One variation could be “Free shipping with any $75 purchase. Top shoe brands: athletic, outdoor, casual, and dress.”
Responsive search ad variations
Responsive search ads display best-fitting headlines and descriptions based on user search queries. The more headlines and descriptions that are configured, the more opportunities Google Ads has to serve ads that more closely match your potential customers’ search queries, which can improve search ad performance. Testing variations may enable you to select the headlines and descriptions that attract more customers.
Steps to set up and test alternate bidding strategies
You can use campaign drafts and experiments to compare your current bidding strategy to an alternate strategy. For example, you can run an experiment to compare Target CPA automated bidding to manual bid changes. This will help you determine if automated bidding will improve the performance of a campaign.
Here is an overview of the steps to set up and test an alternate bidding strategy in the Drafts and Experiments section of Google Ads; for example, you can test a Target CPA bidding strategy (an experiment) against a maximum CPC bidding strategy (a current campaign).
- Name the experiment.
- Choose the start and end dates for the A/B test. You should allow a bidding strategy test to run for four to six weeks.
- Configure the split; almost always 50%.
- Decide if the test will be a search-based or cookie-based split.
- In a search-based split, users are evenly split between the existing and experimental campaigns.
- In a cookie-based split, users who perform repeated searches are directed back to the same campaign. A cookie-based split may give more accurate test results, especially for remarketing campaigns that are directed towards returning users.
- Save the settings. The test will automatically run on the start date.
Best practices to test alternate bidding strategies
A/B testing on bidding strategies must use an existing, live campaign for comparison. Choose a campaign that has an adequate budget to split between the variations and a large enough audience for a statistical comparison of the results. For more information about statistically significant data for A/B tests, refer to the reading about monitoring A/B test results in this course. Only choose a campaign that you’re willing to experiment with because the experiment may impact the performance of the existing campaign. And finally, focus the test on one change or variable only. Testing multiple changes at the same time prevents you from knowing for sure which variable caused a change in performance.
Resources for more information
You can refer to the following links for more information about testing ad variations and bidding strategies in Google Ads:
- About campaign drafts and experiments: This article provides an overview of draft campaigns and experiments.
- Set up an ad variation: This article provides instructions to set up an ad variation.
- Test your automated bid strategies: This article describes how to test bidding strategies.