Converting Feelings to Metrics with a “Safety Check”
In the quest to understand how the delivery teams feel about changes in delivery process, a simple tool called the “Safety Check” can be used to measure how empowered the team feels. Other measures such as participation, active discussion, and happiness indicators can be subjective or influenced externally. With all the metrics there is a possibility of gaming so it is really important that feeling metrics not be tied directly to performance reward mechanisms for the team members involved.
Safety checks are often used at the beginning of retrospectives to “take the temperature” in the room. the process for using the safety check follows:
Supplies: White board, delivery team, facilitator, one color of sticky notes, sharpies
1. Write the numbers 1-5 on the whiteboard with space to the side for stickies
2. Hand out about three stickies and the sharpies to each member of the delivery team
3. Appoint someone from the team to collect the stickies
4. After reciting the “Prime Directive for Retrospectives”, ask the team to write a number between 1 and 5 on their sticky, fold it in half and pass it to the appointed collector.
- 1 means “It doesn’t matter what I say, nothing is going to change.”
- 2 means “I am just grinning and going along.”
- 3 means “I think you can hear me but you will not act.”
- 4 means “I think you can hear me and things might or might not happen.”
- 5 means “I think you can hear me and things will happen.”
5. Give the team a few minutes to write a number on the sticky note, fold it and pass it to the collector.
6. The collector then arranges the stickies on the board next to the appropriate number for all to see.
7. Discuss where the team sees themselves, and if the numbers are all low ask if they would like to continue with the retrospective or discuss what is keeping the numbers low.
For Managers, Coaches, and Scrum Masters, the metrics that can be collected from this exercise are the average of the entire team and the histogram of the stickies. As retros are conducted from sprint to sprint, if the team feels empowered the numbers are going to average in the 4 to 4.5 range. If not the average is going to slip into the 2s and low 3s.
It is important to discuss why the number is low and what the team can do as a part of this exercise. Working with a team that feels everything is beyond their control can be very frustrating for all team members. The leadership can use the averages over time and the histogram to keep a pulse on the feelings of the team. Some changes to really influence team performance can take a long time, like addition of hardware, seeing changes in the average empowerment feeling can help team leaders address issues before the team is thoroughly frustrated.
You may have better explanations for what the numbers could mean. I would love to hear feedback on what you have used in the past that has worked for you. Start your next retro with a “Safety Check” and learn the feelings of your team.
Comments (7)
George Dinwiddie
It’s a very useful practice, but be careful with those metrics. As I mention in http://blog.gdinwiddie.com/2012/02/06/safety-exercise/ you’re only collecting data on how safe people feel about revealing how safe they feel.
George,
Thank you for your comment. I am glad to see that other people are using this exercise at that beginning of their retrospectives. Have you heard of other uses for the data?
Thanks,
Jann
edster
I’m with @George Dinwiddie on this. If the team is feeling safe you’ll get 4 and 5s which don’t really tell you much. If anyone is at 3 or below you can’t even be sure that they feel safe to tell you or their team mates in an open forum how unsafe they are feeling. This is why I’ve stopped running safety checks at the start of retros and instead use an anonymous survey a couple of days before to collect the data — I then use this data to guide the structure of the upcoming retro session. Safety checks are good but use with care.
Edster,
Thank you for taking time to read my blog post. I am interested in the trends in your anonymous survey. With all tools and metrics it is important to look at how things are going and see when you need to switch things up. Thanks again for sharing your experience.
Jann
Scott Madden
Thanks, Jann. I will definitely add this exercise to my toolkit. I have worked with teams that do not feel empowered. This will be a good starting point for that type of discussion (as a coach or a Scrum Master). Any advice for the team that rates themselves very low? Maybe a topic for a follow-on blog post :-)
edster
@Jann, we don’t have enough data yet from the surveys to identify any trending, and at this stage I’m not too concerned about this as I’m using these surveys as a form of intervention rather than a broad metric. What tends to play out is an identification of a safety issue leads to a number of retrospectives focused on the team (individuals and interactions) then a check-in on safety to confirm things have improved/stabilized, we then target more of the nuts-n-bolts aspects of Sprints to give folks a break and see some tangible changes that lead to smoother execution and (hopefully) improved velocity.
Joshua Indech
Hi Jann,
Thank you so much for posting this. For teams that are making the transition and roles are muddy, it is very necessary, at times for the scrum master to check the room. addressing performance and reward and other complexities of team and organization dynamics is something that is always missed in new agile teams.
I do feel this is also applicable to other ceremonies as well.
Truly,
Joshua