This article provides an overview of the performance reports generated for individual experiments across email, push and SMS tests that have used a dynamic optimization testing methodology.
Please refer to this article for more information about performance reporting on the individual experiments that use a standard optimization methodology.
For more information about performance reporting at the account level, please check out this article.
What is an experiment-level performance report?
Experiment-level performance reports summarize the performance of your Phrasee experiments based on the results received from the test.
Where can I find the report?
You can find the performance report for individual experiments under the Results tab for any experiment in Phrasee with result data.
Experiment-level performance reporting for experiments using dynamic optimization.
How to run an experiment-level performance report
Phrasee will automatically start generating the report once it starts to receive the results of an experiment.
What does the report show?
There are two tabs in the results step to look at:
1) Mature data: This is where you can view the true performance of an experiment as it only shows data where Phrasee has received the expected number of opens or clicks for the sends that have taken place.
Mature data will not populate for at least two days after the experiment starts, as Phrasee needs to wait to receive open and click data for an accurate picture of how your experiment is performing. Depending on your message type, this means:
Fo broadcast messages using dynamic optimization: Mature data will not populate for at least two days following the completion of an experiment.
For trigger messages using dynamic optimization: Mature data will lag the raw data by at least two days for the duration of the experiment.
2) Raw data: This shows all the data used in the experiment. It will include send numbers where Phrasee is still waiting to receive open or click data. This data tab is not representative of your experiment's performance. Instead, its purpose is so that you can confirm that the experiment is sending and receiving data correctly.
