Improving Email Marketing by Making a User Interests Prediction
Overview & Challenge
Our client in the retail industry has a marketing department which is dealing with sending email marketing campaigns. The department created marketing campaigns twice a week, but the reach was rather low – a lot of recipients received the mail but there was very low conversion. After thorough research, our client found out that the reason for this is not having information about their users’ interests, meaning that there was no user segmentation and targeting.
The marketing campaigns should include messages that resonate with the right audience. To achieve this the client needed to create email messages relevant, and useful for the customers that they have, sent at the right time.
The solution that we provided is a model that uses machine learning to collect and analyze data of the consumers based on their previous behavior, and present that in reports. After processing the data, the model determines the appropriate time for sending emails, as well as the content that the email should contain.
On the whole, the solution provided enabled the client to make better-targeted campaigns, which resulted in conversion. This, in turn, is enhancing the experience of the customers and increases their confidence.
Improved performance for 100% compared to all automatic and manual, and analytics tools used so far.
Improved business decision making based on the approximation retrieved from the machine learning model.
Increased usability and value for the customers by providing efficient methods for informing them.
Ability to automatically and approximately see the customer’s interests and effectively market the company.