How do I identify users in Heap?

Users in Heap should be identified as soon as their identity is known. You can do so via our identity and addUserProperties APIs. To get started with these APIs, see Track user identity.

I’m not allowed to send PII to Heap, can I still use the Identify API?

Your organization may consider some identifiers, such as email, as PII or sensitive data. However, there are other options for passing Identifiers that may not be considered PII, such as:

• Hashing the email or other identifier
• UUID
• Unique internal user identifier

See Track user identity for guidance on setting up these APIs using your preferred identifier option.

What are the downsides of not implementing the Identify API?

If you don’t identify users, you may run into the following:

• Inflated user count: If users visit your site on different devices or browsers, Heap will consider them a separate user on each device or browser. If the user clears their cookies, Heap will create a new user for them and won’t be able to match them to their previous sessions.
• Broken user journeys: If a user starts a flow on one device or browser and completes it on another, or clears their cookies in between, the full view of that user journey will be lost. Funnel charts, journeys, usage over time charts, and tracking conversion or full-feature analysis won’t be as accurate.
• Server-side events: An identity is required for server-side events to match the event to the user that performed it.
• Connecting to other systems: If you wish to bring data into Heap or send data out, an identifier will be required to merge the event with the appropriate user in Heap when sending data in or with the user in the other system or data warehouse when sending data out.
• Troubleshooting capabilities: Troubleshooting issues reported by users may be difficult as you won’t be able to look up that user by their identity.

What else should I consider when implementing the Identify API?

• If you’re planning on bringing data into Heap from other sources, or merging Heap data with other data downstream, the identifier in Heap should match the Identifier you are using in those systems.
• If your internal users have the ability to log in as someone else, make sure their identity doesn’t change when they do so.
• If the value you use to identify users changes, an entire set of new users will be created. Review the Read This First: Key Considerations section of the Track user identity guide to avoid this scenario.