Using relations, rollups, and formulas to organize content

Advanced properties in Notion turn a database from a simple list into an interconnected system that can automatically organize, calculate, and audit information. Relations link pages across databases, formulas derive values or trigger logic from other properties, and metadata surfaces behind-the-scenes details like creation time and last-edited attribution. That combination matters most when a workspace grows beyond a small team—hundreds of contributors working from the same database can’t rely on manual tags alone to find what matters, so automated layers of structure become the difference between “searching” and “understanding.”

Relations are the most direct way to connect content. When a relation property is configured, clicking into it reveals the related items in the other database, effectively creating a navigable web of context. Notion also supports rollups on top of relations, letting users summarize information from the related records without leaving the current database. The transcript illustrates this with a “meeting notes” database linked to a “docs” database: a meeting can reference multiple documents, and a rollup can then pull a specific field from those related docs—such as “created by”—to produce a “doc author” rollup. The result is a meeting-centric view of who contributes which documents, enabling quick answers like which authors are most active in meetings across the company.

The relation setup includes practical controls. A “limited” relation restricts each row to only one related item in the other database, which fits one-to-one scenarios (for example, a task tied to a single project). Leaving the relation unlimited supports one-to-many relationships, such as multiple docs referenced in a single meeting. There’s also an option to “show property on the related database,” which mirrors the relationship in the other direction—so the docs database can display which meeting notes discussed each doc. That bidirectional visibility reduces the need to jump between databases just to reconstruct context.

Formulas add another layer of automation. They can calculate numbers from other properties or implement interactive mechanisms like an upvote system in a Q&A database. While the transcript notes that formulas can power complex projects (including interactive games), their everyday value often comes from straightforward computations and property-driven actions.

Finally, metadata is positioned as essential for administering large team workspaces. Properties like “last edited time” and “last edited by” make it easier to identify who is updating knowledge and to spot stale content before it becomes a liability. Together, relations, rollups, formulas, and metadata provide a toolkit for building databases that stay organized as teams scale—without requiring constant manual upkeep.