Glossary - Predictions
Last updated
Last updated
CrossEngage's Customer Prediction Platform. The Predictions Platform has the URL: app.crossengage.io
CrossEngage's Customer Data and Engagement Platform. The Data & Engagement Platform has the URL:
A Customer is defined as any User whose data is present in the Predictions Platform. This includes Users who have not yet made a purchase, or who made a purchase but later returned it.
A transaction is generally any activity initiated by the User, which involves exchange of products. Purchases and Returns are common examples of transactions.
Activity is defined as any Customer Interaction recorded on the Predictions Platform, that does not generally include a financial transaction. Some examples might be a Customer logging into an Online Store, calling your helpline or receiving a Marketing Newsletter.
A Data Package is the collection of all Data Files that are used together to create a Model. This helps organize data effectively - every file only needs to be uploaded once, but can be used many times (by adding it to multiple Data Packages).
A Model, commonly known as a Machine Learning Model is a complex file that has been trained to recognize certain patterns. CrossEngage use Machine Learning to train these Models to understand Customer behavior, allowing them to Predict Conversion in the future.
Conversion is defined as the point at which a recipient of a marketing message makes a Purchase.
A Selection is a group of Users on the Predictions Platform that have been bundled together for the purpose of a Campaign. This is similar to a User Segment in the Data & Engagement Platform.
Customer Lifetime Value is the estimated value of a Customer during their entire relationship with a business.
Individual Purchase Interval is an estimate of the time Interval that a Customer takes from one Purchase to the next. IPT is defined Individually - each customer has their own IPT.
In Machine Learning, a Score is a value calculated by the model, that is indicative of the predictive power of the model.
In Machine Learning, a feature is an individual measurable property that serves as an input to the Model. It can be a single field, or a combined and transformed form of several fields.
The target variable is the variable that is being predicted by the model.
A backtest offers the possibility to train a Model on past data and get a forecast for a past point in time (as opposed to a live Model). As the forecast is based on a point in time in the past, the Model performance can be evaluated immediately