Customer support · Production

How Notion uses Custom Agents internally to automate repetitive work across teams

The problem

Notion teams spent significant time manually searching for and sharing knowledge-base answers in Slack—sometimes waiting hours for a reply—while product feedback arriving from multiple channels required constant manual effort to route, transcribe, and create tasks.

Workflow diagram · grounded in source

Question asked in Slack

trigger

“It lives in the CX team's #cx-ask Slack channel. The workflow is simple: Someone asks a question, the agent searches the CX Knowledge Base and enablement docs”

Agent searches knowledge base

ai_action

“the agent searches the CX Knowledge Base and enablement docs, and it delivers an answer with links to the source pages”

Answer delivered with source links

output

“it delivers an answer with links to the source pages. Instant, accurate, and it frees the team to focus on complex questions that actually need human judgement”

Feedback monitored in Slack

trigger

“It monitors that Slack channel, identifies which product area the feedback relates to, then creates tasks for the appropriate teams to handle”

Product area identified

ai_action

“identifies which product area the feedback relates to”

Tasks created and teams notified

routing

“then creates tasks for the appropriate teams to handle. It also sends out messages to other Slack channels to make sure the right teams have eyes on the feedback”

Reported outcome

Custom Agents now answer repeat questions instantly with source links, route feedback automatically to the right teams, and have spread across all departments—freeing teams from busywork so they can focus on complex, high-judgment work.

Reported metrics

Agent response timereplies in minutes

Feedback routing manual workfreed us up to actually work on getting things implemented

Security manual researchcuts down on the manual research and busywork

Reported stack

Custom AgentsNotion AIEnterprise SearchSlackMailCalendar

Source

https://www.notion.com/blog/how-notion-uses-custom-agents

Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Custom Agents now answer repeat questions instantly with source links, route feedback automatically to the right teams, and have spread across all departments—freeing teams from busywork so they can focus on complex,…

What tools did this team use?

Custom Agents, Notion AI, Enterprise Search, Slack, Mail, Calendar.

What results were reported?

Agent response time: replies in minutes; Feedback routing manual work: freed us up to actually work on getting things implemented; Security manual research: cuts down on the manual research and busywork (source-reported, not independently verified).

How is this customer support AI workflow structured?

Question asked in Slack → Agent searches knowledge base → Answer delivered with source links → Feedback monitored in Slack → Product area identified → Tasks created and teams notified.