# Data dictionary

**All records are fictional.** Keep the raw files unchanged and build your analysis separately. `rules.json` defines the population and assumptions. Blank, zero and unknown are not interchangeable.

## tickets.csv

One ticket in the reporting cohort.

| Field | Meaning |
|---|---|
| `ticket_id` | Stable fictional ticket identity. |
| `created_at` | Timezone-aware creation timestamp. |
| `team` | Descriptive queue assignment, not a randomized treatment or employee-quality label. |

## events.csv

One event, with one deliberately repeated exact event ID.

| Field | Meaning |
|---|---|
| `event_id` | Unique event identity; exact repeated payload is a retry and conflicting payload is an error. |
| `ticket_id` | Parent ticket; unmatched IDs are exceptions. |
| `at` | Timezone-aware event timestamp. Future events do not belong in the report. |
| `actor` | agent or bot. Only human agent events qualify in this exercise. |
| `visibility` | public or private. A private note does not constitute the required customer-facing response. |