> For the complete user guide index, see [llms.txt](https://portraitresearch.com/documentation/llms.txt).

# Track events

[Canonical page](https://portraitresearch.com/documentation/monitors/create-event-monitor/)

## Applicability

- **Task:** Configure and verify a Monitor for defined events
- **Access:** Portrait account with applicable usage and source access
- **Audience:** Portrait users maintaining ongoing coverage
- **Surfaces:** web
- **Availability:** Production
- **Content type:** how-to
- **Last reviewed:** 2026-09-07

Use this pattern for a discrete event such as M&A, a regulatory decision, a management transition, a guidance change, or an activist stake.

For events across a fixed coverage list, add the companies. For a market-wide topic with no company list, use the **Monitor topics or events** path only when the account shows it, as described in [Create a Monitor without companies](https://portraitresearch.com/documentation/monitors/scheduled-briefs/). If it is absent, keep the Monitor company-scoped or contact the team; do not leave the company set empty.

## Configure the pattern

1. Create a Monitor and add the intended companies, or choose the company-free topic path.
2. Define the event, its effective date if relevant, and what would count as a match.
3. Choose the source types and email cadence supported by that scope. Company-free Monitors do not support the **Web** or **Podcasts** source switches.
4. Review the configuration before saving. On accounts with usage enforcement, review the displayed usage estimate too. If Portrait rejects the projected commitment, it creates no Monitor; review **Settings → Usage** and reduce the scope or contact your account team.

Example signal:

> Alert me to new information published after 2 November 2025 that could materially change the assumptions in my model. Identify the affected assumption and attribute the source.

For company-scoped Monitors, choose **Real-time** when the event requires prompt review, or a digest when individual matches do not justify interruption. For company-free Monitors, choose **Morning**, **Evening**, or both; see [Create a Monitor without companies](https://portraitresearch.com/documentation/monitors/scheduled-briefs/) for the evaluation cycles and UTC timing. Email cadence does not guarantee immediate detection or delivery.

## Verify and refine

Confirm the saved event definition, date boundary, companies, source set, and cadence. Review the first matching update against its source. Narrow a noisy event definition or widen one that depends on wording the source is unlikely to use.

An initial **Monitor launch brief** provides background, not evidence of a newly captured event. See [Manage a Monitor](https://portraitresearch.com/documentation/monitors/manage-a-monitor/) for source start dates and stopping coverage, or [Export Monitor updates](https://portraitresearch.com/documentation/monitors/export-monitor-updates/) to retain a PDF of results.

## Up next

[Generate a 90-day Monitor summary](https://portraitresearch.com/documentation/monitors/generate-90-day-summary/)