Get certified - Transform your world of work today

ScrumMaster Maturity Model

A Case Study

07/30/2013 by Jerry Rajamoney

This article illustrates some of the key things I learned after trying out the ScrumMaster Maturity Model (SM^3™) by the Braintrust Consulting Group. As an Agile coach, I work with multiple Scrum teams, ScrumMasters, and product owners. I have tried this model with few of the Scrum teams, as pilots, and I gained a lot of insights. The data shown are real and presented as collected.

We are all aware that Scrum is the most widely used Agile development framework today in the software industry. Scrum is a simple framework that helps a team inspect and adapt on an ongoing basis. One of the interesting roles Scrum introduced is the ScrumMaster role.

This ScrumMaster role is unique, and some of the key expectations are:

One of the key challenges I have faced while working with multiple Scrum teams is how the make the ScrumMasters more effective. Is there a way for me to see where the ScrumMasters stand in terms of their roles and responsibilities? Do I get some measurement criteria that are easy to use and at the same time give good insights, so that the ScrumMaster can use them and improve his or her way of working?

To get answers to these questions, I started exploring and came to know about the ScrumMaster Maturity Model (SM^3), developed by Brian Rabon, CST, PMP, president, the Braintrust Consulting Group. This model was recently presented at the 2013 Scrum Alliance Global Gathering in Las Vegas.

ScrumMaster Maturity Model
Here is a quick introduction to Rabon's model. It is based on a simple questionnaire that contains ten questions. Each question is multiple choice, and each team member is expected to pick the option that he or she believes resembles the behavior of the ScrumMaster.

Each option selected has a numeric value attached to it. After answering the ten questions, you can sum up the values associated with the selected options. Based on that final value, the ScrumMaster is placed within a scale of 1 to 5. Each level has a title, such as: 1 = Wolf in sheep's clothing, 2 = Scrum scribe, 3 = Facilitator, 4 = Coach, 5 = Servant leader. For a complete presentation and details about the ScrumMaster Maturity Model, please visit the website

Modifications to the original model
When I tried to implement this model in one of the Scrum teams I'm coaching, I added the following dimension:
  • The ScrumMaster and the team members participate in this process.
  • The options selected for each question by the ScrumMaster and the team members will be plotted in a graph that shows how the ScrumMaster thought he or she was performing in the expected role, as well as and how the team actually perceived that performance. It is similar to a 360-degree feedback process.
Data collection
I proposed this model as a pilot activity to one of my Scrum teams, and the team agreed to try it. We  had a 30-minute Wednesday afternoon meeting that included the full-time ScrumMaster and the five team members. Each person, including the ScrumMaster, received a copy of the questionnaire, filled it out (without discussion), and returned it to me.

When I plotted the data, I found some unique patterns. Here I'll share two charts and the insights they provided to all of us.

Data and results
First scenario

Question Shown / performed behavior
Two team members don't get along and are constantly fighting. It's disturbing the other team members and they have asked you to deal with it. What should you do? a) Bring the conflict up with your human resources representatives and ask them to solve it.
b) Hold an all-hands team meeting and refuse to end it until the issue is resolved.
c) Meet with the two team members and do a root-cause analysis of why they're fighting.
d) Ask management to move one of the fighting team members to another team.
When I plotted the data collected from both the ScrumMaster and the team, here is what appeared:
The ScrumMaster selected option 3, which is C (meet with the two team members and do a root-cause analysis of why they're fighting). The entire team selected the same option. Hence all answers were the same.

In a detailed discussion with the ScrumMaster and the team later on, it became clear that the way of working exhibited by the ScrumMaster was clearly visible, and the team acknowledged his involvement in removing impediments and helping them self-organize.

Second Scenario
Question #2 from Rabon's model was as follows:

Question Shown / performed behavior
A team member is late every day for the daily Scrum. What should I, the  ScrumMaster, do? a) Ignore the issue; the team is self-organizing.
b) Fine the team member $10 each time it happens.
c) Bring the lateness up at the next retrospective.
d) Publically chastise the team member for showing up late.
When I plotted these data, this is what I found:
The ScrumMaster selected option 3, which is C (Bring the lateness up at the next retrospective), but the complete team selected option 1, which is A (Ignore the issue; the team is self-organizing).

In the detailed discussion with the ScrumMaster and the team later on, it was clear that the ScrumMaster made the wrong assumption, from the team's point of view, in this scenario. And he felt it was the team's responsibility, not his, to discuss the problem in the retrospective meetings.

I have tried the ScrumMaster Maturity Model with two business units in which we have eight Scrum teams and six ScrumMasters. I've reviewed the results with all the six ScrumMasters, along with their teams. I personally have seen this model help the ScrumMasters measure their work and also validate their way of working against their own as well as their teams' perspectives.