The AdOptics service works by automatically generating a to do list of tasks to create, monitor, and end ad A/B tests in your ad account. This article covers the current ad tasks generated by the service and what they are used for.

1. Start Ad Test

The start ad test is the most simple task and easiest to run. This to do task will be generated when the service detects more than one ad of the same type (Expanded Text Ad, Responsive Ad, etc.) in a single Ad Group. This means you have two ads running against each other and that you should run an A/B test to see which ad is performing better.

When you run the Start Ad Test task, no changes are made to your Ad account. You simply create an internal test at AdOptics which will begin to monitor the performance of the two ads going forward. Once we get enough data in our daily ad performance import, we will generate a new ad task for you, an End Ad Test.

Once the Start Ad Test task has been ran, you can see the progress of the test under the "In Progress" tab of the Task Manager. Note: the label will change from Start Ad Test to Running Ad Test.

2. Create Ad Test

The Create Ad Test task is the most important and powerful part of the AdOptics service. This task is created when only one ad is running in an ad group. The task will prompt you to write a new ad for the ad group to test with.

When you run the Create Ad Test task, you will push over a new ad to your ad account that will run side by side your existing ad. Running the task opens the Ad Writer interface which is a powerful tool that allows you to craft a new ad based on the highest performing ad elements form the ad group and account history. You can select an existing or paused ad as a starting point based on its past performance.

Once the Create Ad Test task has been ran, you can see the progress of the test under the "In Progress" tab of the Task Manager. Note: the label will change from Create Ad Test to Running Ad Test.

3. End Ad Test

The End Ad Test is the task that actually improves your ad account performance. It is generated after a running ad test has accumulated enough data to declare a winning and losing ad. Normally, the End Ad Test task waits until a statistical significance has been reached and a confidence level of greater than 90% is achieved. In some cases, the End Ad Test task will be generated if an experiment has been running for 30 days with no winner, or if one ad is being shown as being heavily favored by the ad platform in impression share. This will be indicated at the top of the ad comparison tool display.

Running the End Ad Test brings up the ad comparison tool where you can see the performance of the two ads in the test side by side. The system will automatically select the winning ad for you based on the available data. When you execute the End Ad Test, the losing ad is paused on the ad platform and the winning ad will continue to run.

Once the End Ad Test task has been ran, you can see the results of the test under the "Completed" tab of the Task Manager. Note: the label will change from End Ad Test to Ad Test Results.

Did this answer your question?