AI Effort Estimator
Describe a story, rate the things that actually make work hard, and get a t-shirt size, Fibonacci points, and a confidence read - with the reasoning shown. Runs entirely in your browser.
Story description
0 words - used only as a rough base size.
Complexity factors
How much is undefined or unresearched? Vague requirements, new domains, "we'll figure it out".
Other teams, third-party APIs, upstream work, or shared services this relies on.
Coverage expectations, edge cases, manual QA, and regression surface area.
Data migrations, backwards compatibility, production blast radius, rollback difficulty.
Extra risks (0)
Optional - add any wildcard risks that don't fit the four factors above.
Estimate
Why - what pushed it up
Total weighted score: 2.3. Numbers are illustrative estimates - tune the ratings to match how your team actually delivers.
Estimate the hard parts, not the typing
AI coding tools have collapsed the cost of writing code, but they have not touched the things that actually govern delivery time: undefined requirements, cross-team dependencies, the testing surface, and the risk of migrating live systems. This estimator makes those factors explicit. You describe the work, rate each source of friction, and get a t-shirt size, a Fibonacci point value, and a confidence read - plus the reasoning that shows which factors moved the number.
The point isn't the number. It's the conversation. A large estimate driven mostly by unknowns is a prompt to spike or prototype before committing. Low confidence is a signal to split the story. Treat every weight here as an editable starting point - the illustrative defaults exist so you can reason about them, then adapt them to how your team really ships.
Estimates are a plan, and plans only earn their keep when you measure against them. Turn a prompt into a scoped product plan, see how your predictions hold up with the sprint velocity tracker, learn to measure delivery, and read our guide on measuring AI coding tools.
Frequently asked questions
Measure what you ship
An estimate is a hypothesis. FeatureFactory helps engineering teams plan, build, and measure AI web development - so you can close the loop between what you predicted and what you actually shipped, with outcomes you own.
Get early access