Sense-making in real time

I put a call out on twitter yesterday, inviting topics for blog posts that could be helpful. I’m happy to take requests! Today my friend Trilby Smith, the brilliant Director of Evaluation at the Vancouver Foundation, replied with this:

Sense making in real time. Like what are the practices we can use to make sense of what is happening to us as it happens? And how can those of us who work in orgs support our colleagues to do this work?— Trilby Smith (@TrilbySmith) March 17, 2020

The last few days have been full of information. It comes streaming through twitter, facebook, texts, emails. And the way the COVID-19 crisis is moving and changing means that we have to look at stuff coming in, sift through it and make some decisions.

This is true in any fast-moving, data-rich situation, but COVID-19 gives us a chance to practice in real-time. So what are some simple tools? Here are a few, rooted in, and derived from, Participatory Narrative Inquiry and Human Systems Dynamics.

Observe the situation. Just watch things for a bit. Whatever the situation, see if you can gather a bunch of data points about it. If this is a meeting, have people bring in a bunch of notes about the situation. These notes should be observations, relatively free of interpretation. Fine-grained data objects, like stories, tweets, news items, reports, stats are all good. Anything that helps describe what you’re all seeing. And having everyone do this ensures that you get a diversity of perspectives. Have everyone come to a meeting with 10 data objects. Or start your meeting by having people sit around and tell some little stories and share observations about the situation, placing each data point on a post-it note. It should only take you less than 20 minutes to generate dozens of data points if you work in pairs. This, by the way, is what we call “situational awareness.”

Look for patterns. In complexity, you’re trying to work with patterns. My go to is to have the group sort through the data and find things that are similar. Cluster these together. These start to look like patterns. From there do a couple of things…

Inquire. I sometimes think that looking at data is a bit like nosing whisky, or appreciating the scent of a wine or a coffee. You begin with overall impressions and then you use specific techniques to get the most out of the experience. Same with data. When you are sifting through data with a group start by recording what people notice in general. Overall first impressions are useful. Keep it open.

After that you can drill down with a little more discipline and rigour, Royce Holliday offers these questions, from her piece on pattern spotting:

• Generalizations: “In general, I notice…
• Exceptions: “In general I notice…but…
• Contradictions: “On the one hand I notice…but on the other hand…”
• Surprises: “I am really surprised that…”
• Curiosity: “I wonder if…”

These questions help you to find differences in the patterns and differences are what give you the potential to act.

Look at what is keeping these patterns in place. If a problem is complex, you will probably start to notice patterns that are stable and hard to change. Alternatively, you might notice a lot of turbulence and wonder what you can stabilize. Looking at what is keeping patterns in place is fairly straightforward. A system or a set of problems is made of connected agents interacting with a space defined by attractors and boundaries.:

• Attractors hold things together coherently. Think of these as the things that grab your attention or the rhythms that dictate your work.
• Boundaries separate things. These can be tight or loose or permeable or hard.
• Connections in a system describe how agents are connected to one another. Think of a murder mystery where the detective is always trying to figure out how things are related in a meaningful way.
• Exchanges are what flows over connections including information, power and resources.
• Identities come into play and can skew a system with power dynamics, expertise or the diminishment of voice and ideas.

Once you can find a few of these constraints that are at play, you can list things that are in your control and make adjustments. In general, to stabilize a system, you tighten constraints. To break up a system, in order to break patterns or learn new things, you have to loosen constraints. The art is in deciding how much and in monitoring and adjusting as you go. Choose constraints that matter that you have some degree of control over, and you will be able to shift things more easily.

I reckon you could do this quickly in 1:45 or so if you had to generate data objects to start with, less time if people are collecting data objects before the meeting. A sample flow might look like this:

1. Check in and framing (15 minutes)
2. Break into small groups to generate data objects (20 minutes)
3. Randomize the data objects and cluster them into themes (15 minutes)
4. Ask each person to look at the patterns and answer the inquiry questions individually (10 minutes)
5. Small groups to compare notes and find commonalities and differences (10 minutes)
6. Finding ABCEI constraints in small groups (use this template) (10 minutes)
7. In small groups decide on a small action to shift things. (use this template) (5 minutes)
8. Compare these actions across the group and wrap up (20 minutes)

In the comments, I would be interested to hear if this is helpful and what kinds of specific situations folks are needing to make sense of.