How Notion uses Notion (Block × Block)

Notion’s internal operating system is built around shared, cross-team databases—then stitched together with relationships—so user feedback turns into prioritized projects, trackable tasks, and documented decisions without information getting trapped in silos. The core idea is simple but operational: keep company knowledge in a few global sources of truth (docs, tasks, meeting notes), connect them to projects, and use structured feedback tags to quantify what users want before teams commit engineering time.

At the top level, Notion organizes work in a company workspace with pages for functional teams plus three global databases shared across teams. The docs database acts as the persistent knowledge layer: it stores RFCs (requests for comment) for new product ideas and processes, company-wide announcements (including time-sensitive updates like COVID-19 and office reopening decisions), and repeatable process documentation so teams can deliver consistent outcomes. A single tasks database spans all teams, but it stays usable through filtered database views—such as a “My Tasks” view that uses the special “me” variable to show only items assigned to the currently logged-in user, and triage views that surface unassigned bugs and user issues using tag filters and empty-assignee logic. Meeting notes round out the written record: templates keep recurring meetings on track, notes can be converted from block types into pages, and those pages become actionable tasks assigned to engineers.

The next step is where the system becomes more than storage: Notion uses bidirectional relations to connect projects to the relevant docs, tasks, and meeting notes. A project isn’t just a page with a status—it’s a hub that pulls in associated work and context. In the example walkthrough tied to the Confluence import project, relations ensure that customer feedback threads, internal documentation, and the task queue all line up around the same initiative.

Feedback collection is handled through a tagging system designed to be both comprehensive and actionable. User input arrives through multiple channels—In-app chat, email, Twitter, sales meetings, and the Notion ambassador community—then customer-facing teams translate qualitative conversations into standardized feedback tags. Tags preserve meaning over time: once a requested feature ships, the tag remains so future feedback about the existing capability can still be tracked. The system also supports non-product operational categories like startup discount inquiries. Counts are aggregated into metrics (e.g., requests over the past week, month, and four months), and each tag links to related tasks and, where appropriate, public artifacts like tweets.

After feedback tags drive prioritization, project pages track who owns the work, current status (from ideation/specification through development, pause, completion, and follow-up), target ship dates, and the feedback volume by persona (personal vs. enterprise) using formulas. Linked views then surface the exact tasks and docs tied to the project via relations, with database templates making it easy to spin up new projects that automatically populate those views.

Once a feature ships, marketing announces it on the external “What’s New” page, while customer experience teams prepare help center articles and train for incoming questions. An internal release tracker supports launch readiness, and—crucially—tags enable direct follow-up with users who requested the feature, closing the loop through email or even replies to specific tweets. The overall message is that the information system is designed to augment human capability: it turns scattered conversations into structured decisions, then turns decisions back into measurable, trackable work.