What does the tool do?
With the Campaign Experiments tool, you can view and analyze the performance of your Google Ads experiments across all your Google Ads accounts, and use the comparative information to make the decision of either keeping the campaign experiment(s) or ending them.
Why should I use it?
With the Campaign Experiments tool, you can:
View the performance of multiple Campaign Experiments in just one go, making it easy to analyze results without spending extra time going back and forth.
Get recommendations to graduate, continue, or terminate an experiment based on Optmyzr's own check of statistically significant data.
Get a performance summary that will share more details on why the recommendation is being made.
How do I use it?
To get started, select an account - or all accounts - to analyze, and the desired date range (Last 7, 14, 30, 90 days).
Once you select the account and date range, you'll see all the active Campaign Experiments within the linked Google Ads account, along with the following:
Account
Campaign Experiments
Performance: This column shows if your Campaign Experiment is trending in the right direction. You can press the arrow to sort the data.
Start Date: The day Optmyzr started fetching data for these Campaign Experiments.
Traffic Split: This shows the percentage of traffic split between the Original campaign and the Campaign Experiment.
Metric columns: The system will show impressions, clicks, cost, and conversions by default.
You can select other metrics like conv value, CTR, cost/conv, etc from the metric chooser on the top right of the table, as seen in the screenshot below.
Gathering More Details
You can expand the experiment row for more details by clicking on the account name. There you'll see the original campaign on which the experiment is running, check out all the metrics, and see the confidence level regarding if the experiment is going to be successful.
Performance Details
The metrics show you how they are performing as compared to the original campaign. The value under a metric is the current performance of the experiment for the given date range, and below is the difference between the original campaign performance and the experiment’s performance in number and percentage.
Confidence Level based on Performance
The confidence levels in green suggest a positive trend, in red as negative and no color being neutral, which means there is no clear indication of the experiment being successful yet.
The confidence level is calculated by comparing experiments against their original campaigns. Based on statistically significant data, it is determined if the experiment has a better chance of being a successful campaign, based on the metrics like CTR, Conversion Rate, Conversion Impression, Revenue Impression, Revenue Cost, or CPA.
Note: If there is not enough data available for such analysis, you'll see "Not enough data" displayed.
Applying Changes to Google Ads
With the "Take Action" button you'll have three options:
Update original campaign. Selecting this would make changes to your original campaign to reflect the changes made in the campaign experiment.
Convert to a new campaign. This will add your experiment to a new Google Ads campaign with the same budget as your original campaign.
End experiment. If the experiment is not successful, you can end it using this option.
Note: New campaign experiments created in Google Ads will show up within 24 hours in the tool.
Use Cases
Acting Quickly on Experiment Outcomes
Problem: Even after identifying a winning or losing experiment, applying changes manually in Google Ads can be time-consuming and prone to delays.
Solution: With the “Take Action” feature, you can directly update the original campaign, convert the experiment into a new campaign, or end the experiment—all from within the tool.
Understanding Performance Differences at a Glance
Problem: You struggle to interpret how your experiment compares to the original campaign, especially when dealing with multiple metrics and large datasets.
Solution: The tool presents performance metrics alongside a clear comparison against the original campaign, including absolute and percentage differences. You can also customize metrics such as CTR, cost/conv, or conversion value to match your goals.
Prioritizing Experiments Based on Performance Trends
Problem: You have multiple ongoing experiments, but no clear way to prioritize which ones need immediate attention.
Solution: The performance column highlights whether experiments are trending in the right direction, and you can sort them accordingly. This helps you quickly identify top-performing or underperforming experiments.
Demo Video