DISCLAIMER: This is a Cognitive Edge methods post. If you are unfamiliar with Cognitive Edge and Cynefin, you will likely struggle to understand this post. I encourage you to probe around the Cognitive Edge site and watch the Cynefin videos on You Tube. Now, for the post…
Prompt me with a comment and I shall provide. Thanks Jabe for reminding me to get back to the topic of the Guiding Coalition off-site, also known as a large group facilitation of sense-making using the Cynefin framework.
First, I’ll outline the day. Then, I’ll reflect on the day, talking about what went well and what I would change.
The purpose of the day was to take almost 100 people representing a diverse set of organizational viewpoints through a sense-making workshop resulting in the selection by attendees of complex topics to pursue on behalf of the organization. Whew!
The metaphor we used for the day was a roller coaster ride. Some people love them; some hate them. All feel a bit bounded around and confused on their first ride, but at the end many will want to go again. The metaphor worked, with one of the attendees even commenting during open comments at the end of the day that he enjoyed the ride.
We set the structure of the day, a 7.5 hour session, by assigning people to tables according to their main identity (management, support, engineering, production) and then sub-identities. There were seven management tables with sub-identities of uniformed officers, production, engineering, projects, product lines, line support (e.g, financial, IT) and executive support (e.g., executive director, union presidents). In total there were 17 different tables, so 17 different, purposefully exaggerated perspectives.
The attendees got an initial set of slides on complexity, complete with Dave Snowden videos and then practiced working with ambiguity via the Butterfly Stamping method. I enjoyed watching the diversity of responses to the exercise. Some people didn’t know exactly why they were doing it, but went along. Others struggled so violently with the “there is no right answer” ambiguity that I worried they would mutiny against me as the lead facilitator. You know you’ve hit a nerve when you call the exercise to end and you hear a loud, “NOOOO!!” boom from one of the tables.
After a break the participants were instructed in the steps necessary to complete what we called Decision Mapping (aka 4 Points Contextualization). We could jump right in, without doing an anecdote circle because I and a group of loyal helpers had done the work ahead of time. I’d asked two probing questions via e-mail, gathered responses, and clustered the responses into 43 themes. Each theme then got a Wordle to describe it.
After the exercise, we had 17 distinct Cynefin frameworks constructed from making sense of the 43 topics. The members of each table were allowed to rotate to the other tables within their group (i.e., management, support, engineering, or production) and see:
- What was the same?
- What was different?
- What was surprising?
After they had discussed their answers among their group, we dismissed them for lunch and ordered them to leave the conference room. This step was essential because they could not “check their work” against the other tables in the other groups. We wanted to still keep them locked within their perspectives for now.
From there the participants shifted to doing what we called Group Decision Mapping. During the break my facilitation team (2 people for each group) determined which themes (each had a distinct number given to it, 1 to 43) all the tables of the group had placed in the same domain. Out of 43 themes, the most any group agreed on was eight. We posted a new piece of paper on the wall, assigning the agreed on topics into the domains and placing the remaining themes into a central disorder clump. When the participants returned they were counted off into small teams with a representative from each table. Then, those groups were given one theme at a time to adjudicate, i.e., agreeing on which domain the group agreed the theme belonged in. When all the themes had been removed from disorder, the group map was complete.
When all four group maps were complete, the participants rotated to each of the other groups and answered the three questions again.
- What was the same?
- What was different?
- What was surprising?
During the next break, the facilitation team moved the four group maps to the front wall and again determined which themes had been placed in the same domain on all four maps. This time 11 topics were unanimous. The rest were placed into disorder.
When they returned from break, we counted off the participants into 12 small teams. Each team had a member from each group. Each team was given a theme from disorder and asked to reach agreement. The teams quickly processed through the themes and we had a full-room map. From 43 undefined themes to 17 table maps to 4 group maps to 1 room map in approximately 6 hours.
Since this was an off-site for our Guiding Coalition, the goal of the day was to agree on complex themes so our networks of Guiding Coalition members could design safe-to-fail experiments to improve those topics. So, after we had the participants assess each domain and move the themes into the appropriate boundary zones, we asked the participants to look deeply at the complex domain and decide which of the themes they would like to work on during the 2013 Guiding Coalition tenure. After an exercise in team selection, the day was complete.
Though I had to work with four times as many slides as were in the typical deck from Cognitive Edge, I needed the further process details to keep my audience of very process oriented people from mutinying over lack of instruction. That said, the day was still not structured-enough or obvious-enough for some. At first this bothered me, but now I see it as a difference with people’s comfort level with ambiguity. If you don’t like ambiguity, you won’t like the methods. Now, your not liking ambiguity doesn’t change the fact that reality is full of ambiguity, so I guess life is going to be rough unless you can dictate that everyone either forcefully hold back ambiguity or lie to you. Rough.
I used the Cognitive Edge exemplar names (New Age Fluffy Bunny, Tyranny of Experts, Mind Numbing Bureaucracy and True Chaos) once we started the decision mapping instead of using four themes from within our 43. I did that assuming it was one less step of mediation I would be putting between the participants and their data. I didn’t imagine that when I said New Age Fluffy Bunny that more than a few people could only picture a fluffy bunny and were a bit confused. I can’t blame them the error since they would with concrete objects (some figuratively and some literally) each day. New Age Fluffy Bunny was too new age fluffy bunny for them. It meant nothing. I’ll have to come up with a different exemplar title next time. On a side note though, I received some push back from participants who said they struggled to do a one-for-one translation from Complex to New Age Fluffy Bunny and Complicated to Tyranny of Experts. They saw this as a bad thing. I actually think it is a great thing because it kept them sense-making instead of categorizing.
Speaking of categorizing. I changed the rules of the butterfly stamping a bit from how I’d practiced it a few weeks before. When I practiced it I made a point to tell the participants that they could not under any circumstances draw on their paper. They had to make sense of the topics. They didn’t really though. I saw plenty of tables thinking up creative ways to draw in the lines with their eyes, or using paper to split the categories (I say categories because when they did this they missed disorder and just drew a 2×2). Rather than force their categorizing preference underground, I thought it would be interesting to give no direction on marking the paper or not and watch to see how many would automatically draw in the lines before they even put the first butterfly stamping item onto the page. The results at the off-site were stark. Almost 75% of the tables immediately or quite quickly drew in the lines, making the paper into a 2×2 matrix. This information helped me tailor how I presented the steps of decision mapping, where they were directed to not write on the paper and refrain from using the yarn (provided in their mapping packets) until I gave them permission. Giving them the yarn before they needed it was also an intentional poke. I wanted to see who could resist using it and who found the urge the whole way through. I only noticed one table that tried to place the yarn early and they were giving the exercise so little of their attention I think they had bored themselves into causing trouble.
The four points contextualization method calls for making sense of the themes, drawing in the boundaries, splitting those on the boundary, then determining the boundary placements. I chose not to do the boundary placement for the table maps, but did make the participants go through the exercise on the group maps and the room map. Next time I may just skip the boundary placements if I need more time for other conversations because what we are using the data for likely won’t require that level of placement detail. If you were going to take the conversation further you would probably want to keep the boundary step since more information is good information (if you can spare the time).
The biggest win from the day, besides getting to the selection of topics for the 2013 Guiding Coalition, was watching the participants discover other people’s perspectives. The conversations, at the tables, in the groups, and in the teams representing all, were fabulous. Lots of “How can that be?” and “Is that how it really looks to you?” mixed in with “What is this supposed to mean?” asked as though someone was waiting with the answer but just wouldn’t give it. I love chances to push people beyond their expectations. Not everyone liked it, but not everyone will. The goal isn’t comfort. The goal is to achieve a shared understanding, and reaching that often makes people uncomfortable. It was terribly fun to watch.
As the 2013 Guiding Coalition is only just beginning, I can’t offer an outcome reflection stating whether or not this method saved the three months of team building time that I predicted when I sold my leadership on trying it out. I can’t say for sure, but their are good indications. And, when I modify the select/off-site/train pattern next year to select/train/off-site, we’ll save a few months more. Then, we’ll really be flying at light speed! [ Have I mentioned lately how much fun it is to drive change?]
Whew! That’s a lot of reflections. I’m sure I have more that I could remember if prodded. For now, I’ll settle with this length of post and send this out as an extra weekend post in honor of Jabe’s poke that I get this written. This is Labor Day weekend after all. Why should I rest?