Last weekend, I went to visit the city of Paris.
With so many things to see and so little time, we decided to make a prioritized list of places to visit.
Since this was my very first visit to Paris, my list included many of the big tourist attractions:
2. Notre Dame
3. Montmartre & Sacré Coeur
5. Centre Pompidou
6. Tour eiffel
My wife and I prioritized the list according to the personal value, ease of access, available time, …
This includes transportation, walking distance, waiting queues,… every possible factor that can influence the time needed.
To get a little more detailed information, we used a tourist guide of Paris.
Our list was transformed into the following:
Priority – Place – Time Needed
100 Louvre 1 day
90 Notre Dame 3 hrs
80 Panthéon 3 hrs
70 Montmartre & Sacré Coeur 4 hrs
60 Centre Pompidou 1 hr
50 Tour Eiffel 2 hrs
If you haven’t already made the analogy with the creation of a product backlog, it works pretty much the same way.
Together with a client, we make up a list of features and prioritize them according to business value, risk, etc.
We do a first rough estimation on these features by assigning story points. At that point we don’t have all the detailed
information about the features, but we try to get as much info as possible from the product owner and discuss the complexity as a team.
Pretty much the same as we estimated the places to visit. We used the knowledge and tools available to come up with an estimate.
As you know, these rough estimates (story points) are used to make up a release planning.
How many features can we deliver if we know we can tackle x story points per sprint.
That’s exactly what we did for our Paris trip. How many places can we visit in one day (which is our sprint length). Our release length was 2 days.
So our release plan shows as follows:
first day in the afternoon: half a day at the Louvre (we agreed to minimize the scope and not visit all the corners of the museum)
second day: Centre Pompidou, Notre Dame, Panthéon and Tour eiffel
third day in the morning: Montmartre & Sacré Coeur
When we looked back at these three days, our estimates seemed quite accurate. This is surprising since we ‘ve never visited these places before.
I believe the reason is, just as in agile software development, we were able to reduce scope on some items, keeping only the stuff that was most valuable to us.
Off course we included some margin in our estimates, but the places we overestimated were compensated by the ones we underestimated.
How did I came up with this analogy? I remember reading a blog post that explained how we estimate when to leave to the airport to catch a plane.
When making up the list of places to visit, that blog post popped into my mind. So I decided to try and do the exercise on our situation.
I don’t remember which blog it was, but I’ll post the link as soon as I find it.
Finally, a picture of the Panthéon in Paris.