Machine Learning

Send Time Optimization

The Send Time Optimization model leverages the past email behavior to predict the best send time to optimize email open behavior. The model predicts the best send time for weekdays and weekends separately as well as collectively.

Use cases

Email engagement optimization

Identify when customers are likely to open an email and tailor the send time at the individual level accordingly. This helps avoid email fatigue for customers who receive multiple emails. In addition, marketers can prioritize the quality of emails over quantity.

Target audience

The Send Time Optimization model predicts the behavior of all customers, including buyers and non-buyers.

How does the model work?

For each contact, the model:

  • Calculates the number of successful email sends for each hour slot in a day, across the past year. A successful send is defined as an email that was opened within 24 hours of being sent.

  • Leverages this information to learn about the user behavior to enhance the model’s predictions.

  • Sends random send slots to a small percentage of users to enhance the data diversity.

  • Uses population-level best send time for users without sufficient data.

  • Predicts the final outcome, which is a combination of all the previous points.

Using the Send Time Optimization model

CDP displays the output from the Send Time Optimization model in:

  • 360 > 360 Profiles

  • Actions > Campaigns and Actions > Campaigns+

  • Analytics > Metrics

To learn more about how the Send Time Optimization model can enhance your marketing workflow, contact us.