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

# Report scheduling

> How to schedule Ryze reports and alerts, the default cadences, and how to route them to Slack and email per workspace.

Ryze delivers reports on a schedule so you do not have to check a dashboard. Alerts fire on significant change rather than on a timetable.

## Default cadences

| output                | cadence                      |
| --------------------- | ---------------------------- |
| AI visibility summary | Weekly, Wednesdays           |
| Rank tracking data    | Daily                        |
| Agent Activity        | Continuous, as actions occur |
| Alerts                | On significant change        |

## Delivery channels

* **Slack** — connect a channel per workspace
* **Email** — to individuals or a distribution list
* **Dashboards** — live, always current

Agencies should route per workspace so each client's data goes to the right place. Mixing clients into one channel is the usual cause of a report reaching the wrong recipient. See [agency and multi-workspace](/plans/agency-multi-workspace).

## Choosing a cadence

Match the reporting interval to how fast the thing actually moves:

* **Paid ads** — daily or weekly. Changes land within 24 hours.
* **Technical fixes** — as they ship. Approved fixes go live within hours.
* **Organic rankings** — weekly at most. Daily rank data exists, but daily *reporting* on it invites reacting to noise.
* **AI visibility** — weekly, read monthly. Volatility makes anything faster misleading.
* **CRO tests** — at conclusion, never mid-flight. Peeking at running tests is how false positives get acted on.

<Note>
  Reporting more often than a metric meaningfully changes does not give you more information — it gives you more noise to react to.
</Note>

## Alerts

Alerts are for things that need a decision: significant traffic drops, spend anomalies, indexation failures, connection losses.

Keep them few. An alert channel that fires constantly gets muted, and then the one that mattered is missed too.

## Scheduled agent work

Report scheduling is separate from when the agent *works*. Unattended agent runs are configured in [Schedules](/agent/schedules), and what requires approval in [Permissions & approvals](/claude/permissions).

## Related

* [Exporting reports](/reports/exporting)
* [Agent schedules](/agent/schedules)
* [Reports overview](/reports/overview)