We’d like to request a native Slack integration within Fivetran to support granular, configurable alerting for connector failures. While we understand email-based notifications can be routed into Slack, this approach is limited and doesn’t meet the needs when expanding our Fivetran usage to many engineering teams at the company and managing data pipelines at scale. Without native Slack support, we’re stuck building brittle workarounds in DataDog or email forwarding which have limitations based on the logs available and their customizability. 

Email-to-Slack shortcomings:

• Poor formatting: Emails in Slack are noisy and hard to scan; native Slack alerts are cleaner, interactive, and take up less space.

• Lack of control: Fivetran alerts today are all-or-nothing. We can’t target specific Slack channels or filter by connector, error type, or retry threshold.

• Modern expectations: Most other tools in a modern data stack support native alerting via Slack, Webhook, and PagerDuty.

What We Need

Native Slack integration with support for:

• Per-connector or per-destination configuration

• Alert thresholds (e.g. only after 3 retries or X minutes) or some way to filter to only meaningful failures that won't self-heal and require user action (maybe something already exists in Fivetran that can be reused here)

• Channel-level routing (so we can route alerts to the specific working/alerting channel for each engineering team)

Nice-to-have:

• Webhook support at the connector level

• Control over which event types to notify on (failures, schema drift, sync delays, etc.)

• Alert templating with quick links to logs and connector metadata