To give you greater control over the way your rankings are generated, Convert offers a suite of customization tools. By experimenting with custom data and different ranking methods, you may be able to unlock higher quality and better performing results. Below is an overview of each model customization feature.
Custom Data Imports
Knowledgebase article: Managing Custom Data
The past shopping behavior data collected by Google Analytics is just one lens we can use to predict the likelihood of a product being sold in the future. Take advantage of the full breadth of the information you have at your disposal by importing additional product data from third-party systems such as your PIM, CMS, ERP, or others.
For best results, custom data should be imported in a format that will allow it to be digested by machine learning models. If you don't have a data science team that can perform this action for you, we can help!
Custom Model Features
Knowledgebase article: Managing Custom Model Features
Once you've imported a custom data set, you'll need to define the fields that you want to make available to your ranking models. These product attributes or metrics are called model features.
Custom Ranking Models
Knowledgebase article: Managing Ranking Models
Ranking models are the machine learning algorithms that Convert uses to generate its rankings. For experimentation and testing, you can create and edit your own models. This gives you control over the goal of the model, as well as the model features used.