Quadratic is an AI spreadsheet that connects to your data sources and helps you analyse them in a familiar grid. You ask questions in plain English, and the AI writes the formulas, Python or JavaScript code and charts needed to answer them. The company says it is used by more than 250,000 people at companies and universities.

Connecting data

Quadratic can bring in CSV files, Excel workbooks and PDFs, as well as live databases, so that data from different places sits in one sheet. The site describes AI agents that help connect to data sources. This addresses a common problem where databases, spreadsheets, files and APIs are scattered and hard to combine.

Code instead of nested formulas

A central idea is replacing long, fragile formulas with readable code. The homepage contrasts a deeply nested VLOOKUP with a few lines of AI-generated Python using pandas that do the same lookup across several tables. Code cells can be read, checked and reused, which makes analysis easier to verify. Traditional formulas are still supported, and prompts such as forecasting quarterly revenue or explaining a dip in a product line can be used as starting points.

Sharing

Results stay in a spreadsheet that colleagues can open, and the product supports shared workspaces and repeatable analyses, so a report can be refreshed rather than rebuilt.

Pricing

The Free plan includes limited AI usage, sharing, files and connections. Pro costs $18 per user per month billed annually and includes $20 of AI credits per month, more models, extra privacy controls and unlimited sharing, files and connections. Business costs $36 per user per month annually, with $40 of AI credits and on-demand credit top-ups. Enterprise adds SSO, a self-hosting option and dedicated support.

Who it suits

Quadratic suits analysts, finance teams, researchers and developers who are comfortable in spreadsheets but want the power of Python and direct database access.

Things to consider

Review AI-generated code and results before sharing them, especially for financial figures. Check how database credentials and data are handled, and consider the self-hosting option if your data is sensitive.

Because analyses live in code cells, a teammate can trace exactly how a number was produced, which helps when reports are audited or handed over.

Quadratic Multiplayer

Overview