Two Truths and A Lie About Agile - Mike Cohn

Let’s play a quick round of Two Truths and a Lie.
Here are three statements about agile. Two are true. One is false.
Read them over and see if you can spot the lie before I reveal it.

1. Agile teams should be willing to change their plan for the sprint if they discover a better way to meet the sprint goal.
2. Estimation in agile is most useful for helping teams forecast and make trade-offs, not for holding individuals accountable.
3. A team that consistently finishes every planned story in every sprint is demonstrating a healthy, predictable agile process.
   The lie is #3.
   That statement sounds responsible, disciplined, and maybe even a little impressive. Which is exactly why it fools people.
   It reflects a common misunderstanding of agile: the idea that a good team is one that is always comfortable, always certain, and always exactly on plan.
   But healthy agile teams are not defined by perfect adherence to a prediction. They are defined by how well they pursue outcomes, adapt to what they learn, and make sensible decisions in the presence of uncertainty.
   Let’s look at each statement.
   

A sprint plan should guide the team, but it should not trap the team. At the start of a sprint, the team creates the best plan it can with the information available at that moment.
Once the sprint begins, though, the team learns more. A technical approach that seemed promising turns out to be awkward. A dependency proves easier than expected. A simpler solution emerges. Or a conversation reveals a better way to achieve the intended outcome.
When that happens, a good agile team should be willing to adjust. The important thing to preserve is the sprint goal rather than every detail of the original plan.
The goal provides focus. The selected stories, tasks, and implementation approach are simply the team’s current best thinking about how to reach that goal. If the team discovers a better path, it should take it.
Changing the plan during a sprint is not a sign of weak discipline. In many cases, it is evidence that the team is paying attention and responding intelligently to what it learns.
Teams get into trouble when they stick to the initial plan even after new information shows a better way forward. Agile works best when teams stay committed to the goal while remaining flexible about how to achieve it.
Estimation is most helpful when it supports planning and decision-making. Teams estimate so they can answer practical questions like these:

• How much work can we likely take on?
• When might a larger effort be completed?
• If we add this item, what will need to move?
• Are we taking on too much uncertainty at once?

Those are valuable questions, and estimation can help teams answer them.
Where estimation becomes far less useful is when it is turned into a tool for judging individual performance.
Once estimates are used to hold individuals accountable, people naturally become more defensive with them. Estimates get padded. Uncertainty gets hidden. Conversations become less honest. The numbers may still exist, but they stop helping the team make good decisions.
That is why I prefer to keep estimation focused on decision-making. Estimates do not need to be exact to be useful. They only need to be good enough to help a team forecast, weigh options, and recognize when it may be taking on too much.

A team that finishes every planned story in every sprint may look predictable. But if that happens all the time, I would not automatically consider it a sign of health. In fact, I would probably wonder whether the team is planning too conservatively.
How to connect with AgileDad:

- [website] ⁠https://www.agiledad.com/⁠

- [instagram] ⁠https://www.instagram.com/agile_coach/⁠

- [facebook] ⁠https://www.facebook.com/RealAgileDad/⁠

- [Linkedin] ⁠https://www.linkedin.com/in/leehenson/