Skip to content

Instantly share code, notes, and snippets.

@theoschao
Last active July 27, 2017 23:47
Show Gist options
  • Save theoschao/0ca720fbc6c0204fab96951669070633 to your computer and use it in GitHub Desktop.
Save theoschao/0ca720fbc6c0204fab96951669070633 to your computer and use it in GitHub Desktop.
Markdown for Clearbrain <> Segment Direct Integration

How it Works

ClearBrain is a predictive analytics platform, that ingests your customer data from Segment to build high propensity audience segments. The platform uses machine learning to learn from your users’ past attributes as tracked in Segment, and determine who is most likely to perform a predicted track, page, or screen event. Every user is assigned a ClearBrain score indicating their likelihood to perform the event in the future, allowing you to target your highest propensity users at the right time and place in your ads, emails, and A/B testing destinations.

Getting Started

It takes on a couple steps to start predicting any behavior with ClearBrain:

  1. Go to app.clearbrain.com/connections to retrieve your API key
  2. Enter your API Key into the ClearBrain Segment destination
  3. Click Historical Replay to sync your past events to ClearBrain Note: the more historical data you send to ClearBrain, the more accurate your predictions will be.

Turning on the ClearBrain destination in Segment will trigger the following calls to be forwarded to ClearBrain, which in turn will result in both user properties and events to be included in future predictions:

  • identify
  • track
  • page
  • screen

Identify

When you identify a user, we'll infer users and their properties, such as the Segment userID, anonymousID, and UserAgent from the respective Identify call. These are used as the primary key to join your respective user attributes across tracks and pages, which are in turn used to train and personalize your ClearBrain predictions.

Track

A track event can be used as the target goal for any of your ClearBrain predictions. Any track call that has been instrumented for at least one week can be used as the basis for a prediction in ClearBrain, e.g. Users Who May -- Segment Web -- Clicked Login Button -- This Week.

Page & Screen

A page event contributes to attributes in ClearBrain pertaining to pageviews, referrers, UTM metadata, etc. A screen event in your mobile app contributes to attributes in ClearBrain pertaining to screenviews. Both are leveraged as additional attributes that inform and train your ClearBrain predictions.

Settings

Segment lets you change these destination settings via your Segment dashboard, without having to touch any code.

Questions?

For questions, please email support@clearbrain.com.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment