A marketing department of a UK-based investment fund with a retail offering.
The marketing department sends a couple of emails to the fund’s subscribers every week. It must balance several considerations: it should send only relevant content to their readers, while providing enough communication to maintain interest in the fund. It also needs to manage the list strictly to remove dormant subscribers regularly.
Faculty was commissioned to develop methods for monitoring the appropriateness of the client’s email list.
We developed a machine learning (ML) model that predicts the likelihood of an email being opened by each individual reader, based on the content type of the email and the reader’s historic behaviour. We created the model by means of a web-based app that automatically incorporates data from the client’s customer relationship management (CRM) system. It automatically performs an analysis and allows the client to visualise the model outputs. The model allows for tweaks and refinements by the client before the emails are sent out. Finally, the app generates a custom mailing list for each new marketing campaign.
The dashboard allows a granular view that shows each reader’s consumption pattern. With a number of filters across categories of emails, this tool offers a high level of customer segmentation in real time at the push of a button.
The predictive dynamic mailing list for each email is automatically generated, meaning that each new marketing email can be provided with a custom mailing list that now results in a far higher level of engagement from customers than before.
The app we developed offers seamless reporting and tracking of the efficiency of marketing campaigns.