Ecommerce ops · Production

Dynamic Yield delivers affinity-based personalized recommendations across web and email, driving 40% RPM increase

The problem

The marketing team lacked access to AI-based targeting and was manually handpicking products for email recommendation widgets, creating hours of manual work for the merchandising team.

Workflow diagram · grounded in source

Capture omnichannel user events

trigger

“Using omnichannel events to capture new and returning user activity”

Build single customer view

integration

“linking behavior across devices and browsers in order to create a snapshot of each user's individual preferences”

User Affinity algorithm scores preferences

ai_action

“the User Affinity algorithm, which delivered more relevant, personalized, and suitable recommendations to each user, especially as each user's affinity profile grew with time”

ML-powered email retargeting

ai_action

“machine learning-powered recommendation algorithms, they retargeted users with product recommendation widgets within emails, using a customer's entire order history and online activity to predict and display the most relevant products”

Personalized email widget served

output

“Used deep learning AI to serve personalized recommendations directly within emails”

Reported outcome

Implementing the User Affinity algorithm and machine learning-powered recommendations resulted in a 40% increase in revenue per thousand impressions (RPM) from email campaigns and eliminated hours of manual work for the merchandising team.

Reported metrics

revenue per thousand impressions (RPM) from email campaigns40%

Manual work for merchandising teameliminated hours of manual work

Reported stack

Dynamic YieldUser Affinity algorithmdeep learning AImachine learning

Source

https://www.dynamicyield.com/case-studies/email-recommendations

Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Implementing the User Affinity algorithm and machine learning-powered recommendations resulted in a 40% increase in revenue per thousand impressions (RPM) from email campaigns and eliminated hours of manual work for t…

What tools did this team use?

Dynamic Yield, User Affinity algorithm, deep learning AI, machine learning.

What results were reported?

revenue per thousand impressions (RPM) from email campaigns: 40%; Manual work for merchandising team: eliminated hours of manual work (source-reported, not independently verified).

How is this ecommerce ops AI workflow structured?

Capture omnichannel user events → Build single customer view → User Affinity algorithm scores preferences → ML-powered email retargeting → Personalized email widget served.