Sometimes a Little Serendipity Goes a Long Way: Adventures in Prioritization in an Agile Team Context
Overview
As a facilitator, it can be hard to decide whether to craft a detailed agenda in advance to help a team arrive at a particular outcome, versus taking several steps back and letting the team decide discover what direction it needs to take. During a recent team brainstorming session, where the focus was on features we might want to work on for the next several months, I chose the second of the two options (where I did not introduce much structure, and instead waited for the team to propose an approach).
Meeting Summary
The team came up with a list of features that could potentially be worked on. The Product Owner had prepared a list, but she also wisely asked the team for additional inputs.
Once we had an initial list (which consisted of 14 features), I challenged the team to think about ways that we could sort/prioritize the list. It was at this point that the true “wisdom of the crowd” (the team) came through.
Specifically, the team came up with was a set of four factors, where we chose to use a scale of zero through three to assign a score for each factor. Here are the four factors (with some minor rewording to protect the confidentiality of the work we were doing):
- Complexity (technical complexity)
- External stakeholder impact
- Internal stakeholder impact
- Glitz (think of this as a “wow factor,” analogous to a “delighter” in Kano Model Analysis)
When the scoring was complete, we had a chart that looked much like this:
__Complexity __ Internal __ External __ Glitz __ Sub-total __ Feature
===============================================================
______ 3 __________ 2 __________ 1 _______ 1 _______ 4 _______ Feature 1
______ 2 __________ 3 __________ 3 _______ 2 _______ 8 _______ Feature 2
…
Note: The Sub-total column is the sum of the Internal, External, and Glitz columns — more on why I took that approach below.
Impact and Effort Matrix
As the team was in the process of deciding on the approach described above, I was thinking about various techniques we could use to visually show the prioritized list of features. The technique that I decided to use was the Impact and Effort Matrix.
The Impact and Effort Matrix is a prioritization technique that makes it possible to visualize work based on trading off the importance of two things:
- Effort required to implement the solution (and for Effort, I used the Complexity score referenced above for each feature, where a score of zero would be the lowest possible complexity, and a score of three would be the highest possible complexity)
- Impact of doing the work (and for Impact, I used the sum of the three columns called External, Internal, and Glitz, shown in the Sub-total column above). In other words, we added those three scores together for each feature to assess the feature’s business value, in terms of the potential payoff of doing the work from several different perspectives
Setting up the Impact and Effort Matrix
Setting up the matrix was easy once we had the data:
- I drew a blank matrix on flip chart paper
- I wrote the features on individual sticky notes
- I wrote the business value score (the sum of the three columns mentioned above) in the lower right corner of each sticky note
- I wrote the complexity score in the lower left corner of each sticky note
- I placed the sticky notes in the appropriate place on the matrix.
Note: For anyone who has not used an Impact and Effort matrix, items with a low complexity (Effort) score and a high business value (Impact) score land in or near the upper left quadrant of the matrix, which is the “sweet spot,” i.e., the things which deserve consideration to be worked on the soonest (the greatest possible value for the least possible effort). Conversely, items with a high complexity (Effort) score and a low business value (Impact) score land in the lower right quadrant of the matrix, where it is much harder to make a case for doing the work, at least not right away (the lowest business value with the greatest effort).
