Notion Office Hours: Formulas 101 🧮

Notion formulas turn database fields into automatically calculated values—so instead of manually typing results for every row, users define logic once and let Notion compute it item by item. The session frames formulas as a practical tool for two big jobs: calculating new values from existing properties (like full names, ages, taxes, totals) and reformatting values into more useful, readable forms (like labeling an age as “Age: 34” or formatting order dates for a gallery view). That matters because it reduces repetitive work while making databases easier to scan and act on.

The walkthrough begins with the core concept of a “formula property.” Unlike text, date, or number properties where values are entered directly, a formula property generates a unique output for each database item by running the same formula across the database. The presenter demonstrates creation of a formula property, starting with a simple literal expression (2 + 2) to show that literal values produce the same result for every row. From there, the session quickly moves into why formulas are worth the effort: merging first and last names, computing age from a birthdate using the now() function, calculating tax from a tax rate and price, and building totals by combining computed fields.

To make formulas less mysterious, the session breaks them into a hierarchy of parts. At the highest level, a formula answers what action to take and on what input values to take it. Actions come from two places: operators (like +, -, *, /, %), and functions (like add(), concat(), join(), dateBetween(), formatDate()). A key detail is that the same symbol can behave differently depending on value types—for example, + adds numbers but concatenates text strings. Functions are treated as packaged operations: a keyword followed by parentheses, with arguments separated by commas.

The session then drills into input value types, because they determine whether an expression works and what it returns. Notion formulas primarily deal with numbers (including decimals), strings (text), dates (including now()), and booleans (true/false, which render as checked/unchecked checkboxes). Conversions become essential when mixing types—such as turning a number into text so it can be concatenated into a label. The presenter uses format() to convert numbers to strings and toNumber() to convert numeric text back into a number.

Practical examples tie everything together using two sample databases: a basketball roster and an e-commerce orders table. In the roster, formulas generate full names via concat() or join(), compute age using dateBetween(now(), birthday, "years"), and create a labeled age by converting the age number into text before concatenation. In the orders database, formulas generate a unique line-item ID by joining order number and product SKU with a dash (formatting the order number as text first), compute tax with multiply(price, taxRate), compute total cost by adding price and tax, and calculate fulfillment timing using dateBetween(now(), orderDate, "days"). Date formatting is highlighted as especially powerful: formatDate() plus pattern strings can produce custom month/day/year and time-zone-aware outputs for display.

The Q&A adds important boundaries. Formulas can only control the output of the formula property itself (including true/false checkboxes), not other property types like Select or Status, and they can’t directly automate changes in related properties. Questions also point toward future sessions on conditional logic (Formulas 201), decimal rounding, and replicating spreadsheet functions like VLOOKUP using relations and rollups.