--- title: Alerting Integrations: Slack, PagerDuty & More description: Send confirmed downtime alerts to email, Slack, Teams, Discord, Telegram, PagerDuty, Opsgenie, push apps and signed webhooks. 16 channels on every plan. canonical: https://watchfor.io/alerting-integrations --- # Alerting Integrations for Slack, PagerDuty, Webhooks and 13 More An alert that says only 'check failed' wastes the first ten minutes of an outage. WatchFor sends confirmed incidents to 16 channels, with what failed and where. [Get started for free](/auth/sign-up)[View pricing](/pricing) [One page per channel](/integrations)[Webhook payload and signature](/docs/notifications/webhooks) - Email - Slack - Microsoft Teams - Google Chat - Discord - Mattermost - Rocket.Chat - Telegram - ntfy - Gotify - Pushover - Pushbullet - PagerDuty - Opsgenie - Zapier - Webhooks Slack, Microsoft Teams, Discord, Telegram, Google Chat, Mattermost, Rocket.Chat, PagerDuty, Opsgenie, ntfy, Gotify, Pushover, Pushbullet and Zapier names and logos are trademarks of their respective owners, shown only to indicate that WatchFor can send alerts to these services. WatchFor is not affiliated with, sponsored or endorsed by them. Gotify logo by the Gotify project, CC BY 4.0. ## What Happens Between an Incident and Your Phone Where alerts go, what they contain, and what keeps the count down ### 16 Channels, Every Plan Email, Slack, Microsoft Teams, Google Chat, Discord, Mattermost, Rocket.Chat, Telegram (official bot), ntfy, Gotify, Pushover, Pushbullet, PagerDuty, Opsgenie, Zapier and webhooks. Every channel type is available on every plan, Free included. SMS is not offered yet. ### Pagers That Open and Close Themselves PagerDuty (Events API v2) and Opsgenie (US or EU region) get one event per incident. It opens when the incident fires and resolves when WatchFor sees the recovery, so nobody closes the page by hand. ### Signed Webhooks One JSON POST per event, incident.firing or incident.resolved, with the diagnosis, expected and detected values and a link to the incident. Set a secret and each request carries X-WatchFor-Signature, an HMAC-SHA256 of the raw body. Zapier gets a flat payload shaped for its field mapper. ### Alerts That Say What Broke The title says DOWN, Degraded, UP or Resolved. The body has a plain-language diagnosis such as Unexpected HTTP Status, Expected vs Detected, the target, start time and duration, the locations that confirmed the failure, certificate details for SSL incidents and a link to the incident. ### Routing From Contact Groups A monitor targets one or more contact groups, and a channel is attached to groups. An incident goes to every enabled channel on the monitor's groups, filtered by each channel's Critical and Warning setting. Changes apply to the next alert, with no restart. ### One Summary During a Storm When several monitors fail together, each chat and email channel gets one summary within 30 seconds instead of a message per incident. Recoveries are grouped the same way. PagerDuty, Opsgenie and webhooks stay one event per incident. ### A Default Channel That Can't Be Deleted Every organization starts with an email channel to the owner, so the first critical alert never goes nowhere. You can edit, re-route or disable it, but not delete it until another channel is made the default. ### Duplicate Guard and Test Sends A channel with the same URL, contact groups and severity filter as an existing one is refused, because it would notify twice. The same URL with different routing shows a warning. Send test delivers a [TEST] message straight away. ## Channels, routing and the delivery record Sixteen channel types in the New channel dialog, all on every plan. Flow view: channels notified at once, then each escalation level. Every notification with its channel, status, severity and incident. ## Fewer Messages, Each One Worth Opening Alert fatigue usually starts upstream. One probe loses a route, ten messages go out, and the team learns to mute the channel. WatchFor sends nothing for a single failed check: a notification goes out only for an incident confirmed from several locations. A burst of incidents is folded into one summary per channel, and a muted monitor or one in a maintenance window sends nothing. What does arrive carries enough detail to start fixing before you open the dashboard. [Start monitoring now](/auth/sign-up) A single failed check never notifies; only confirmed incidents do First message within 30 seconds, grouped with anything else that fired A recovery message with what the check returns now and how long it was down Per-channel severity filter keeps warnings off the pager Mute a monitor for 1, 8 or 24 hours or 7 days; suppressed sends are logged as Silenced Optional repeat reminders while down, stopped by acknowledging the incident Tokens and webhook URLs encrypted at rest, masked in the dashboard, never returned by the API Notification history shows every delivery and failure per incident ## How Alerting Integrations Work Add a channel, attach contact groups, pick severities, test 1 ### Add a Channel Alerting → Channels & contacts → New channel. Pick a type and paste what it asks for: an incoming-webhook URL for Slack, a chat ID from the Telegram bot, a routing key for PagerDuty. 2 ### Attach Contact Groups Tick the groups whose monitors should reach this channel. New monitors join your default group automatically, so they are covered from the first check. 3 ### Pick Severities Critical and Warning are both on by default. Untick Warning on the pager channel so only outages wake someone up. 4 ### Send a Test Send test on the channel delivers a [TEST] message now. Test notifications on a monitor sends through the exact channels that monitor resolves to and reports what each one did. ## Alerting Integration Use Cases However your team likes to be paged ### On-Call Paging Send critical incidents to PagerDuty or Opsgenie, or use WatchFor's own on-call: rotations and escalation policies that page each person on email, Telegram, ntfy, Pushover, Gotify or Pushbullet until someone acknowledges. ### Team Chat Alerts Post to the Slack, Teams, Discord, Google Chat, Mattermost or Rocket.Chat room the team already watches. A storm arrives as one summary, and the recovery follows in the same room. ### ITSM & Ticketing There is no native Jira, ServiceNow or Zendesk channel today. Point a signed webhook at a small handler, or a Zapier Catch Hook at a Zap, and open the ticket from the incident.firing payload, using event plus incident.id to avoid duplicates. ### Custom Automation A webhook handler can restart a service, annotate a dashboard or hand the incident to an AI agent. Verify the X-WatchFor-Signature header; WatchFor does not publish a source-IP allow-list for alert deliveries. ## Frequently asked questions Everything you need to know about alerting integrations [All questions→](/faq)[Ask us directly](/contact) Which notification channels does WatchFor support? 16 channel types: email, Slack, Microsoft Teams, Google Chat, Discord, Mattermost, Rocket.Chat, Telegram, ntfy, Gotify, Pushover, Pushbullet, PagerDuty, Opsgenie, Zapier and webhooks. All of them are available on every plan, including Free. SMS is not offered yet, and there are no native ticketing channels; use a webhook or Zapier for those. Can I use webhooks for tools that aren't on the list? What's actually in a notification? How does routing work? What are contact groups? Do I always have a working fallback? Will I get duplicate or noisy alerts? Do I get recovery notifications? Can I test a channel before a real incident? Can I page one person instead of a whole channel? ## From our blog Guides, deep-dives, and best practices from the WatchFor team [Reliability ### Alert Fatigue: why your team stopped reading alerts (and how to fix it) Jun 25, 2026·5 min readRead article](/blog/alert-fatigue)[Engineering ### Forty monitors fail at once. Slack gets one message. Oct 20, 2026·8 min readRead article](/blog/alert-grouping-last-mile)[Reliability ### On-Call Best Practices: a rotation that doesn't burn people out May 27, 2026·4 min readRead article](/blog/on-call-best-practices) ## Connect a Channel and Send a Test Create a free account, paste a Slack webhook URL or your Telegram chat ID, and press Send test. Every channel type is on the Free plan, and no credit card is needed. [Get started for free](/auth/sign-up)[View documentation](/docs/notifications) --- Canonical page: https://watchfor.io/alerting-integrations · Site guide: https://watchfor.io/llms.txt