Recipes are end-to-end for popular predictions, combining a use case with its connection to create a fully-provisioned pipeline.

Recipes overview

Recipes are end-to-end guides for popular prediction use cases with Faraday. Browse through popular use cases in the Dashboard's recipe wizard for a guided walkthrough, or customize how-to docs in the recipe builder on our website for full documentation for the Dashboard or API. From lead content personalization in BigQuery, where you’re free to deliver persona predictions to your favorite activation platform, to customer churn scoring in Salesforce.

When building a recipe, you'll walk through the resources required to create a prediction. Starting with creating a connection, you'll pull your data into Faraday. Next, you'll describe your data in a dataset through its identity sets, events, and traits so that Faraday can resolve identities and recognize individual parts of your data.

Moving on, you'll use the events and traits you described to organize the people in your data into cohorts, the building blocks of your predictions in Faraday. Then, your cohorts will be the fuel for your propensity predictions in outcomes and persona predictions in persona sets, while your events power recommender predictions.

Last, you'll delpoy your predictions with a pipeline, where you'll choose which cohort you want to make predictions on, what predictions to make, and where to send them.

For recipe creation instructions using both the Dashboard UI and API, see our recipe builder.

👍Key takeaway: recipes

Deleting recipe components

Building a recipe will walk you through the creation of each resource within Faraday. For the order in which resources should be deleted, see object preservation.