Skip to main content

Outline

Welcome​

In Optimizely Analytics, cohorts are powerful tools for segmenting users or other entities based on shared behaviors or properties over time. Whether you're identifying high-value users, targeting marketing efforts, or measuring product impact, cohorts let you define dynamic groups that update with your data.​

​This chapter will walk you through how cohorts work within Analytics, how they are created and managed using flexible templates and logic blocks, and how they can be applied to real-world analysis. You’ll learn how to define cohorts directly within a dataset, use them in explorations and dashboards, and publish them to your warehouse as reusable audiences foractivation across your data stack.​

​After completing this module, you will be able to:​

  • Define and configure cohorts using the block editor​

  • Differentiate between saved and inline cohorts​

  • Publish cohorts as reusable audiences in your warehouse​

  • Use cohort templates to build tailored segments​

  • Apply cohorts in analysis via the catalog and exploration templates

What is a Cohort?​

​A cohort is a computed, boolean column added to a dataset based on logical or behavioral rules. Once defined, it behaves like any other column and can be used throughout your analysis environment.​

​Every cohort is configured with the following:​

  • Dataset – Cohorts are tied to a single dataset and cannot be reassigned later.​

  • Name – Must be unique among all columns in the dataset.​

  • Definition – Defined using the Block Editor, where you set the rules that determine cohort membership.​

​You can follow a step-by-step process to create a cohort using the block editor. We recommend viewing the demo on how to create a cohort for a guided experience: https://optimizely.navattic.com/e8b0ok4

​How to define an Inline cohort​

​Inline cohorts are temporary segments defined within the context of an exploration. These are not saved in the catalog and are useful for one-off or ad hoc analysis.​

​To see how an inline cohort is defined within an exploration, refer to the demo walkthrough that shows the inline cohort creation process: https://optimizely.navattic.com/4na08u1

Publishing Cohorts as Audiences​

​When a cohort is published to your warehouse, it becomes an audience. Audiences allow you to use cohort definitions outside of analytics and within other tools or downstream workflows.​

There are two publishing formats:​

  • Table – Publishes a static snapshot of the current cohort membership.​

  • View – Publishes the logic as a live view, automatically reflecting updates.​

​To learn how to publish a cohort to your data warehouse as a table or view, you can explore the demo on audience publishing: https://optimizely.navattic.com/0ta0t1r

Cohort Templates​

Optimizely provides several cohort templates to help you create cohorts more efficiently. Each template offers a structured starting point and can be edited as needed.​

​Behavioral Cohort​

The Behavioral template allows you to build a cohort based on user activity, such as sign-ins, page views, or transactions. This template is ideal for behavior-based segmentation tailored to your business.​

​You can follow along with the demo that walks through creating a cohort using Behavioral template from scratch: https://optimizely.navattic.com/qmk07jh

​Formula Cohort​

This template enables you to define cohorts using formulas and logical expressions.
For example, you might segment users with email addresses ending in .com or those whose age is above a certain threshold.​

​To see this in action, check out the demo on building a cohort using Formula template: https://optimizely.navattic.com/agu0pzs

Conditional on Property Cohort​

With this template, you can define a cohort based on logical (AND/OR) conditions applied to dataset properties—such as users who are both active and verified.​

​A demo is available that shows how to create cohorts using conditional on property template: https://optimizely.navattic.com/h6a0rd0

Custom Cohort​

The Custom template provides full flexibility by letting you combine multiple logic blocks to create advanced segments.
For example, you can create a cohort of Regular Subscriptions, where at least half the users are regular watchers.​

​You can view a detailed demo on how to use the custom block editor to create a cohort like this: https://optimizely.navattic.com/gl130c5l

Conditional on Aggregate Cohort​

This template lets you define cohorts based on aggregate values—such as users who completed more than five purchases in a given period.​

​A step-by-step demo demonstrates how to set up a cohort using Conditional on Aggregate template: https://optimizely.navattic.com/iea01cy

Conditional on Intervals Engaged Cohort​

This cohort type is based on engagement across time intervals.
For example, you might define Regular Watchers as users who streamed content on five different days in the last 30 days.​

​You can learn how to configure this by watching the demo on creating a cohort using Conditional on Intervals Engaged template: https://optimizely.navattic.com/t8a0ytd

​Additional Resources​

To explore this topic in more detail, visit the official documentation: Optimizely Cohorts Documentation