Forecasted spend

Forecasted spend predictions reveal how much individuals will spend, over how many transactions, and over a certain timeframe, enabling you to find and focus on your best customers.


In this tutorial, we'll show you how to:

  • Create a predictive model for forecasted spend using a forecast.

Let's dive in.

Uses prerelease features

This document refers to features which are not yet available for general release: Forecast. Contact support if you'd like to request early access. Screenshots are disabled on this document.
  1. You'll need a Faraday account — signup is free!

Confirm your data

Event stream

Unless you’ve already created it for another quickstart or purpose, you’ll need to add the following event stream to your account:

  • Transaction

What’s an event stream?

Predicting a certain customer behavior requires historical examples of customers exhibiting that behavior. Faraday works best when that data comes in the form of “events” — specific actions or occurrences that happened at specific times.

Formulating data this way helps you define cohorts more expressively.

For example, a Customers cohort could be defined as the group of people who have all experienced a Transaction event at least once.

For more, see our docs on Cohorts, Events, Traits, and Datasets (which define how events and traits emerge from your data).


To verify, use a GET /streams request. Your response should look like this:

  "name": "Transaction",
, ...}]

Make note of the IDs of the necessary streams.

If the required event stream isn't there, follow the instructions using this button, then return here to resume.

Configure your prediction

Create a forecasted spend forecast


Use a POST /forecasts request:

curl --json '{
  "name": "Forecasted spend",
  "stream_name": "transaction",
  "stream_property_name": "value"

Your forecast will start building in the background. You can proceed immediately with the next set of instructions. When your forecast is done building, you’ll get an email.