Agile estimation and how to... (part 1 of 2)
You would have seen my recent post on story mapping and bringing a project to life with UI, features and user stories. Now comes the fun part, estimates and how to use these in an agile process.
Understanding the size of a piece of work is always important to people, some more then others, especially when there is cost involved. The idea of estimates is to gain this understanding and provide guidelines to your stakeholders.
Some questions you are trying to answer here are:
Using the story map from an earlier post we use a tool called estimation poker.
Understanding the size of a piece of work is always important to people, some more then others, especially when there is cost involved. The idea of estimates is to gain this understanding and provide guidelines to your stakeholders.
Some questions you are trying to answer here are:
- "How long will this take us?"
- "How much will this cost us?"
Using the story map from an earlier post we use a tool called estimation poker.
How to play estimation poker
Prerequisites:
- Run through each story so the team understand the end-to-end
- Break any stories down if something is to large
- Agree on a Definition of Done - e.g. passed user testing and ready for production
- Get each developer and tester to write t-shirt sizes on post it notes, one for each size (XS/S/M/L/XL)
If there are any outliers for example 3 x M and 1 x XL ask the outlier to example their reasoning, if need be carry out a re-estimate based on this and take the most agreed on estimate. If the estimate is split evenly for example 3 x M and 3 x L take the larger of the two.
Once you have all your stories estimated, put them through a conversion table and this will give you a total amount of story points for your MVP.
Once you have all your stories estimated, put them through a conversion table and this will give you a total amount of story points for your MVP.
Conversion table | |
XS | 2 |
S | 3 |
M | 5 |
L | 8 |
XL | 13 |
This is just one method of agile estimation games you can play with your team, for more reading and methods visit http://tastycupcakes.org/tag/estimation/.
The question now is "How do we take these estimates and turn them into a projected timeline?"
To be continued in my next blog post...