# NoEstimates



## Notes

[[Estimation]] is not much better than gambling, since we ascribe a _guess_ to an estimate.

80% of software projects fail.

Estimates focus on the wrong things. If a project comes out on time and in budget but produces the wrong software, the focus is wrong.

A [[user story]] is not a functional requirement, it’s a narrative.

Story points were meant to obscure time, and [[agile development]] is not time-based.

Even measuring a team’s output by velocity requires we return to estimation.

The developer team should be in control of dictating how long things take. The business should then decide _what_ it wants to build.

Estimates are waste, if waste is something that doesn’t deliver value to the customer.

We do have to do accurate projections. Projections are for answering:

1.  Is this project worth our time?
2.  Do we need more people?

Presenter plugs https://noestimates.com

A cumulative-flow diagram can be an alternative. After a certain amount of sprints you should be able to somewhat accurately predict feature completion.


## Refs

-   https://youtu.be/QVBlnCTu9Ms