> ## Documentation Index
> Fetch the complete documentation index at: https://docs.oration.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Tracker groups

> Group many customer tickets under one tracker ticket when they are all the same incident.

When fifty customers report the same outage, do not update fifty refund tickets with the same root cause. Create one **tracker** ticket and **group** the customer tickets under it.

## Rules

| Rule                                       | Why                              |
| ------------------------------------------ | -------------------------------- |
| Only a **tracker** ticket can be the group | Trackers are the incident object |
| Only a **customer** ticket can be a member | Those are the individual reports |
| A ticket cannot join itself                | Prevents loops                   |
| A customer ticket cannot join twice        | Duplicate membership is rejected |

Back-office tickets do not join tracker groups. Keep specialist work on back-office types linked by process, not by this grouping.

## Add a member

Open the **tracker** ticket. In grouped tickets, add a customer ticket from the same workspace.

The tracker stores the membership. History on the tracker records the set of member ticket numbers (`#1042`, `#1088`, …).

## Remove a member

Detach a membership when that customer’s issue was not actually part of the incident. The customer ticket remains; it is just no longer grouped.

## How to work the incident

1. Create a tracker type ticket (for example *Payments outage 24 Aug*).
2. Group inbound customer tickets as they arrive.
3. Update status and notes **on the tracker**.
4. Resolve or refund the customer tickets individually when each person is made whole.

<Tip>
  The tracker is never shared with customers. Grouping does not notify them. Use each customer ticket (or its conversation) for customer-facing updates.
</Tip>

## Clone and types

Cloning a tracker **type** copies the form, not the live group memberships. Each incident is still a new ticket you group by hand (or via automation you add later).
