How to Set Up Effective A/B Tests with Google Ads' Ad Variations

Iterative testing is crucial to the ongoing success of any PPC campaign. It allows marketers to continuously refine and optimize their ads based on performance data, ensuring that they can adapt to changing market conditions and consumer behaviors. As more of the testing process is entrusted to ad platform AI, gaining insights into what worked becomes more complicated. Therefore, having a clear plan and understanding the testing process is essential for building practical takeaways for future advertising tactics.

To set up ad variations in Google Ads, follow these steps:

1. Go to Campaigns > Experiments from the left menu.
2. Select Ad variations within the page that appears.
3. View the table containing any active and completed variations. To create a new one, click the blue “+” symbol above the table.
4. Define the parameters for the campaigns and ads you’d like to involve in the test. Note that this option is currently only available for responsive search ads.
5. Use the Filter ads option to select which ads to involve based on headlines, descriptions, paths, or URLs.
6. Choose between three types of variations: Find and replace, Update URLs, or Update text.
7. Set the details for your variation, including a name, start and end date, and experiment split.
8. Click Create variation to finalize the process. The variation will go live after Google’s review process and upon reaching the start date.

Here are some practical suggestions for using ad variations in your account:

• Test wording changes in ad copy: Swap terms like “coupon” for “offer” to see which generates a better response.
• Try location-specific ad copy: Compare performance between location-specific and more generic ad copy.
• Seasonal messaging: Test including seasonal messaging vs. generic messaging during specific sale periods.
• Product or service name changes: Test the name change in a portion of your ads to see how people respond.
• Pin headlines in different orders: Compare performance when mentioning a “Save 25%” offer in the first vs. second headline position.
• Test pinning assets: See if pinning assets performs better vs. not pinning assets.
• Remove headlines and descriptions: Test performance with fewer assets vs. including the maximum number of assets.
• A/B test landing pages: Use the “Update URLs” option to insert the URL for another page variant.
• Test display path changes: See the effect on user response or if tailoring the display path to the keyword/ad group theme performs better.
• Impact of ad descriptions: Test how changing ad descriptions impacts performance when all else is equal.

When reviewing ad variation performance data, focus on the following metrics:

• Conversion rate: Ensure you have conversion tracking set up properly in your account.
• Cost/conversion: Compare the cost per conversion between the variations.
• ROAS (Return on Ad Spend): If you have values associated with conversions, include ROAS in your analysis.
• CTR (Click-Through Rate): Consider CTR as an indicator of ad relevance to the user, while noting that ad tweaks may involve gating out unwanted users, which could decrease CTR.

Ultimately, your testing should focus on driving more sales or leads at a more efficient cost. Based on the data, decide whether to apply the changes made in the variation or manually update the ads.

Google’s ad variations feature allows you to modify ads across a campaign or multiple campaigns. This ensures that a clean subset of people receive one version and another subset the next version. The feature supports testing different offers, types of messaging, asset orders, and more. The value of using ad variations over standard Google Ads experiments lies in the ease of setup and the ability to test across multiple campaigns. Users are assigned to a variant based on cookies and will continue to see only one version of the ad, although factors such as private browsing and blocking third-party cookies may affect accuracy.