In Sentry, go to Monitors, then Alerts, and click Create Alert. Pick a source and an environment, build the alert with When triggers, If filters and Then actions, set throttling, name it and click Create Alert. Actions can notify people through Slack, Discord, MS Teams, PagerDuty or Opsgenie, or create tickets.
An alert decides who hears about a problem, how fast, and how often, and it's worth a few minutes of thought. Get it right and the team fixes errors before customers notice; get it wrong and everyone mutes the channel. Older guides send you to Alerts, then Create Alert Rule; the current docs, as of Nov 2026, use Monitors, then Alerts.
Before you start: who's allowed to build one
They're an organisation-level feature. Owners, Managers, Admins and Team Admins can always create and edit them. Members and Team Contributors can too by default, under the Let Members Create and Edit Monitors and Alerts setting, unless someone has switched that off. Alerts scoped to all projects can only be managed by Owners and Managers.
Building the alert
1. Open the form
Go to Monitors, then Alerts, and click Create Alert to open the New Alert form.
2. Choose the source and environment
The source can be all issues in selected projects, specific monitors, or all issues in all projects, which includes projects created after you save. The environment defaults to All Environments; set it to one, such as production, if you want production alerts going somewhere different from staging.
3. Add When triggers
These start the evaluation and use ANY logic, and you can't add the same one twice. Options include a new issue being created, an issue escalating, a resolved issue regressing, an issue being resolved, and an event or issue activity being captured.
4. Add If filters
Each If block uses one qualifier: all, any or none. Filters include issue age, assignment, frequency, category, type and priority. Categories cover error, feedback, outage, metric, db_query, http_client, frontend and mobile. With no filters at all, the alert fires on every issue that matches the trigger.
5. Add Then actions and test
Actions can notify suggested assignees, a team or a member, or send to Slack, Discord, MS Teams, PagerDuty or Opsgenie. They can also create a ticket in Jira, Jira Server, GitHub, GitHub Enterprise or Azure DevOps. Use Send Test Notification on the If/Then block to check the wiring; event-based triggers may behave differently in real use.
6. Set throttling, name it and save
Throttling runs from every trigger, the default, through 5, 10, 30 and 60 minutes, 3 and 12 hours, 24 hours, 1 week and 30 days. Give the alert a name that says what it watches, then click Create Alert. You'll find it in the Alerts list straight away.
You can add several If/Then blocks to one alert, and each is evaluated separately against the same triggers. That's how one alert's able to page the on-call person for high-priority issues and post everything else to a quieter channel.
Filters that quietly stop working
Sentry Docs flag a limit that's easy to miss. Frequency and event attribute filters only work with two triggers: 'A new issue is created' and 'An event or issue activity is captured'. Pick an escalation or resolution trigger and those filters are greyed out with a warning, so the alert won't filter the way you'd expect. If you need a frequency rule, build it on one of those two triggers.
Want AI doing the heavy lifting in your marketing?
I build the systems that handle the boring 80 percent, so you get your week back. Done properly, with the human kept in.
Other error trackers with alerting
Datadog says its Error Tracking captures and processes errors across web, mobile and backend applications, using the Browser SDK, the Mobile SDK, or errors ingested from Logs, Traces and Real User Monitoring. Honeybadger lists integrations with Slack, Jira, PagerDuty, Microsoft Teams, GitHub and others, and officially supports Ruby, JavaScript, Elixir, Go, NodeJS, Java, Python, PHP and more.
If alerts create Jira tickets for you, this list of issue trackers to compare is worth a look, and my guide to API monitoring tools covers the uptime side that error tracking doesn't.
My advice for your first alert
Start with one alert on production only: trigger on a new issue being created, filter to high priority, send to one Slack channel, and throttle to 30 minutes. Test the notification, watch it for a week, then add a second If/Then block for paging if something urgent slips through. Name each alert after what it watches so the team trusts it.
Want your tool on this list? Featured spots and reviews here are paid partnerships. Here's how to partner with me.
Frequently asked questions
Where do I create an alert in Sentry?
Go to Monitors, then Alerts, and click Create Alert.
Who can create alerts?
Owners, Managers, Admins and Team Admins always can; Members and Team Contributors can by default unless the organisation setting is turned off.
Can one alert send to different places?
Yes. Add several If/Then blocks; each is evaluated separately against the same triggers.
How do I stop alert spam?
Use throttling, from 5 minutes up to 30 days, so actions don't repeat for the same issue within the window.
Can Sentry create tickets from an alert?
Yes, in Jira, Jira Server, GitHub, GitHub Enterprise and Azure DevOps.
