This preview of pull request 1639 is meant for internal use only.

Personas Default Limits

Segment Personas enforces a set of default use and rate limits to help ensure the performance and reliability of the Personas product. These limits are designed to provide consistent, reliable performance at scale. Most customers do not exceed these limits.

Contact your dedicated Customer Success Manager or friends@segment.com to learn more about custom limits.

Default Limits

Name Limit Details
Inbound Data Throughput 1000 events per second The total stream of events from the sources connected to Personas. Once the limit is reached, Segment reserves the right to slow the processing of requests. This limit also includes historical data replays into Segment.
Outbound Downstream Destination Rate Limits Reduced retries when failures exceed 1000 events per second Most destinations have their own rate limits that Segment cannot control. In some instances, Segment can ingest and attempt to deliver data faster than the downstream destination can accept data. Outbound requests to a destination may also fail for other reasons outside of Segment’s control.

When requests to downstream destinations fail, Segment makes additional attempts to deliver the data (retries). However, when more than 1,000 requests per second to a downstream destination fail or when the failure rate for a downstream destination exceeds 50% for more than 72 hours, Segment reserves the right to reduce the number of retries until the condition is resolved.

Profile API

Name Limit Details
Profile API Throughput 100 requests per second If requests exceed 100 per second, the Profile API returns HTTP Error 429.
Events Lookback History 14 days The Profile API retrieves up to 14 days of a profile’s historical events within a collection. This applies to track events, not traits sent through identify calls.

Identity

Name Limit Details
Identity Merges 100 merges Personas supports up to 100 merges per profile in its identity graph. A merge occurs when two existing profiles are joined together by a common external_id. For example, if a user starts on mobile, and then signs in through a web application, those two user profiles are joined together by a common identifier like a user_id. Messages that attempt additional merges are usually a sign of a corrupt profile and are dropped.

Once the limit is reached, Segment rejects additional events.
Identity Mappings 1000 mappings Personas supports up to 1000 mappings per profile in its identity graph. Mappings are external identifier values like a user_id, email, mobile advertising id, or any custom identifier.

Messages that attempt to more than 1000 mappings are usually a sign of a corrupt profile and are dropped. This limit counts mappings across all merged profiles.

Audiences and Computed Traits

Name Limit Details
Compute Concurrency 5 concurrent audiences Segment computes five new audiences or computed traits at a time. Once the limit is reached, additional audience computations are queued until one of the 5 audiences finishes computing.
Compute Throughput 10000 computations per second Computations are counted as any track or identify call that triggers an audience or computed trait re-computation.

Once the limit is reached, Segment reserves the right to slow the processing of an audience.
Events Lookback History Depends on your Personas service:

- Essentials: 1 year
- Advanced: 3 years
The period of time for which events stored for audience & computed trait computations.

This limit is based on your Personas service. Contact your account team to upgrade your Personas service.

SQL Traits

Name Limit Details
SQL Traits Depends on your Personas service:

- Essentials: 5
- Advanced: 25
The number of SQL traits you can sync to your Personas space.

Contact your account team to upgrade your Personas service.
SQL Traits - Sync Frequency Depends on your Personas service:

- Essentials: Twice daily
- Advanced: Customizable, up to Hourly
The frequency with which Segment runs your SQL trait

Contact your account team to customize your schedule.
SQL Traits - Rows 10 million The number of rows each SQL trait can return.
SQL Traits - Columns 25 The number of columns each SQL trait can return.

This page was last modified: 25 May 2021



Get started with Segment

Segment is the easiest way to integrate your websites & mobile apps data to over 300 analytics and growth tools.
or
Create free account