Archive for the ‘metaphors’ Category

Explaining the value of estimation to the team

March 5, 2009

Before the next release planning of Brasigo and BlogBlogs, I called each team to show them the hour-glass metaphor. The message I wanted them to understand was: estimation gives us predictability, and this is crucial for every business, including ours.

To achieve this, I played the hour-glass estimation game exactly as I did with the product owners. After the third sprint, they correctly estimated the total amount of time needed to pass all the sand from one side to the other. And then I said: “Do you know what we did? Before the end, we know how much time it will be needed to pass all this sand to the other side. This is very similar to release backlog estimation: after a few sprints, we have a very good estimate of when we will finish it! This gives us and our stake holders predictability, and this is crucial to our business!”

Both teams liked this metaphor very much, and they truly understood why we need to estimate, and why we need to take it seriously. Also, I felt that they liked to be with me in a meeting explaining this, and also that they liked the gaming thing. Based on that, and on this recently post that I read, I’m now researching some other kinds of games to play with them. But this is the subject of a future post.


Advertisements

Explaining empirical process control to a product owner

March 4, 2009

The day after, I called each of the product owners of Brasigo and BlogBlogs, separately, and showed them the hour-glass metaphor. The message I wanted them to understand was: as the current release evolves, we will have a good estimate of when it will be finished, with its current scope, and that is empirical process control.

To achieve this, first I played the hour-glass estimation game: 3 sprints of 30 seconds each, so that them could make four guesses for the total time needed to pass all the sand from one side to the other. At the third sprint, both of them made the right guess, and I stopped the game.

Then, I started it again saying: “Imagine that this sand is the product backlog slice corresponding to a release. At the beggining, the developers have no idea of the time needed to implement it. But after some iterations, usually the team will have a very good estimate of how much time they still needs to finish it. And that is empirical process control.”

After that, both of them had an instant insight!

Hour-glasses and empirical process control: a good metaphor

March 3, 2009

hour-glassOne night I was paying my bills, and suddenly I saw a tiny hour-glass on my desk. My wife probably left it there for some reason… But after I saw it, I wanted to know how much time it would take to pass all the sand from one side to the other. Curious as I am, I turned it and started to count. After a few seconds, like a flash, I stopped it, putting it with the two sides on the desk, because I saw Scrum happening right in front of my eyes, clearly as water, and in a sufficiently easy way to explain it to anyone else! Actually, I saw not only Scrum, but any empirical process control method.

What happened? I didn’t know how much time it would take to pass all the sand from one side to the other. This situation is like an unestimated product backlog: you just don’t know how much time it will take to build it. So, I made a guess: 5 minutes. Then, after 30 seconds, I stopped the hour-glass, looked how much sand passed, and made a new guess, given the amount of sand and time passed. This was my first sprint! After 3 sprints, that is, one and a half minute, half of the sand had already passed. So, my guess to the total time needed was certainly 3 minutes, and that was right! With only half of the sand in the other side, I had the exact amount of time that the hour-glass needs to pass all the sand from one side to the other! WOW!

That’s it! Empirical process control for dummies! =)