End-user guides:: Predict - End-user guide – Emarsys

Introduction

Emarsys Predict is a predictive recommendation engine which uses machine learning algorithms to offer your customers the best products wherever they interact with your business.

Pre-requisites

To use Predict you must have completed the Emarsys data onboarding and implemented the data collection scripts.

Implementing Predict Solutions

Predict recommendations are delivered in different ways, depending on the channel being used:

Using Mobile Engage Predict Product Mobile Recommendations

And some advanced features to fine-tune your implementation:

Best Practices and Examples

Some simple examples of how Predict can make a difference to your business.

Web recommendations

Using the Home Widget – A fully personalized home page experience; it displays a personally relevant list of topics organized into relevant sections.
Using the Personal Widget – Personalized recommendations which match a visitor’s known characteristics and interests.
Using the Related Widget – Similar items to those currently being viewed.
Using the Also-Bought Widget – Items that have often been bought alongside those currently being viewed.
Using the Cart Widget – Complementary or related items to those left in a shopping cart.
Using the Category Widget – Personalized recommendations within a specific category, for enhanced targeting or product promotion.

Email recommendations

Abandoned Cart campaigns – The Abandoned cart widget delivers the last items left in a cart by the email recipient, as well as product recommendations related to these.
Post-Purchase campaigns – The Post-purchase widget enables cross- & up-sell recommendations based on the items purchased.
Abandoned Browse campaigns – The Mail Personal widget delivers personalized product recommendations to email recipients based on recent browse and purchase history.