Avoid Sprints


Avoid Sprints.

Sprints create an arbitrary deadline every x days. This creates estimation overhead, and fosters short-term thinking. 👎

Instead:
1. Break work down into small tickets.
2. Estimate future flow by tracking the number tickets completed per week.

Simple.
Oh, and breaking tickets down is naturally useful - even if you don't care about estimating future flow.

Small tickets reduce risk.
Small tickets increase flow.
Small tickets enable developers to work in parallel.
Small tickets make code reviews easier and more approachable.
Some people fixated on the word "week" here: "Estimate future flow by tracking the number tickets completed per week."

The week is just an example. The real point: Estimate future velocity by counting the number of tickets closed historically over a given time-period.

View original on X