A leading bank’s marketing department wanted to target customers at their preferred time to increase engagement. Profusion proposed building an algorithm that predicts the best time to target individual customers.


Our data science team ran a ‘time and day analysis’ (TDA) on historic customer data. The team designed a model trained on email campaign data to predict which day of the week and hour of the day would drive higher customer engagement. The model is designed to clean the noise introduced by a number of external factors and provide accurate predictions.

To provide an actionable and intuitive outcome, we created a score based on quantiles with the following classes:

  • Very good
  • Good
  • Neutral
  • Bad
  • Very bad

The output involved creating a reusable solution to dynamically send emails at the correct time, based on segments, using the ‘best time to mail’ information. Allocated send times were incorporated into the campaign workflow. Where there was not enough data for a customer, they were flagged with the time identified as the best performing.


Time and day analysis was employed to determine the best time to send:

  • We looked at email campaign performance over time to determine when customers are more inclined to engage
  • Provided TDA heat maps to show optimal times to send

Sign up to our newsletter

Want to make life easier by staying on top of market trends?