Advanced Engine

Using the Advanced template in Bloomreach Engagement, you can combine several RS. This option allows you to create a customized model when you combine more models and set additional options.

The following engines are available in Bloomreach Engagement for you to combine.

Name	Which model is it
Manual selection	Manual selection technique
New items	New items template
Chosen by metric	Chosen by metric template
More like this	More like this template
Textual similarity	Text model
Customer recent interactions	Customer recent interactions template
Personalized recommendations for you	Personalized recommendations for you template This model lets you add as many events as you want, each with its own custom weight. The weight is a positive number that shows the event's importance. The bigger the number, the stronger the event signals user similarity.

Setup guide

1. Choose a catalog

Choose a product catalog that contains all your products and set a catalog filter to filter our only relevant products. Typically, it's used to filter only available products.

2. Choose and set up the models

You can add any model from the list by navigating to Choose and set up the recommendation models and hitting the + Add engine.

This example combines 2 models: More like this and Personalized recommendations for you, with priorities 1 and 2 respectively.

You will first get all results from the More like this model (since it's priority 1) and then results from the Personalized recommendations for you model will follow (as it's priority 2).

864 — Example of Advanced template with two models

3. Select the combination strategy

After you pick preferred engines, you can also choose a combination strategy for building the recommendations:

Combination name	What does it mean
Only the best one	We will pick the best model according to its performance. Performance is measured after the live traffic reaches at least 1,000 views (`recommendations.action` = `view`) of the results of all the engines. The best model is the one with the highest number of clicks (`recommendations.action` = `click`).
Combine multiple	Recommendations will be mixed from the results of all selected engines.
Order by priority of models (You can pick the priorities in the previous step while designing the engines.)	Recommendations are made using engines in order of their priority (1 is the highest). This means we'll first use the engine with priority 1. If this engine doesn't provide enough results, we'll use the engine with priority 2, and so on, until we've returned the needed number of items. This method helps define custom backups in case a recommendation model can't personalize.

715 — Combination strategies explained visually

Model combinations

It's recommended to mix different engines in a model, especially for email recommendations so you get the most relevant suggestions. As the lowest priority (the biggest number), we suggest using a manual selection engine as your backup if the previous engines don't give any or enough results.

Remember that the more models you combine, the longer the API response time becomes.