The last session I attended today was run by Adrian Howard. Going into it I thought that the session was going to be about replacing user stories, but this wasn't entirely the case. Essentially it was about starting with a user story and then conducting an experiment.
Adrian used the following example:
As a potential customer I want to be able to signup with my Twitter account so that I don't have to fill out a registration form.
The first step is to turn the story into a question.
As a potential customer I want to be able to sign up with my Twitter account so that I don't have to fill out a registration form?
The next step is to ask why the story is needed. Too many times we just assume that there are good reasons for it and don't question. After finding out the main reason why, we should design a hypothesis to validate that assumption and then conduct an experiment to prove or disprove it.
Adrian's team wanted to validate that a percentage of new users would want to signup using twitter.
To do this they put a fake Twitter link page which logged how many people clicked on it. The user was then redirected to a page saying that the Twitter signup was currently offline and redirected them to the normal registration page.
After the experiment was over they found that very few people actually clicked on the link, so they didn't build the feature. Traditionally the team would have just built it and no one would validate if it was ever used after being deployed.
This has potential to be a catalyst for turning an organization into one of experimentation.
I'm still mulling this over, but I think it has great potential.
What do you think?