Discover more from Tom Kerwin
A trip into the estuary
Notes from a recent strategy workshop, sharing an emerging complex facilitation method that uses metaphor to break habitual patterns and empower teams to start journeys.
Last week, I had the chance to co-facilitate a workshop with Dave Snowden, using his new — and very exciting — Estuarine Framework.
Today, I’m sharing a blow-by-blow overview of the workshop with you. It was an absolute blast for facilitators and attendees alike, and members of the Cynefin Community (Cynefin is Dave’s original framework) enjoyed reading a rougher version of these notes.
Before we jump in, I can tell you this was quite different from most workshops.
Like a lot of complexity-wrangling methods on the cynefin.io wiki, the Estuarine Framework can seem daunting. But the principles are surprisingly simple.
It’s almost orthodoxy in much of the design and product world now that we should be working to desired outcomes rather than desired outputs. Of course you define what you want to have happen, then work backwards from there to deduce what you need to do.
But when you’re working in complexity, which is most of the time — as soon as you’ve got people in the equation, BOOM, complexity — it’s literally impossible to know in advance what’s going to work.
You can copy patterns that worked before, but they won’t work the same way again because your context is different now. Obvious solutions (“we just need to do X!”) always have unintended consequences. And some ideas that sound strange or too small can have non-linear positive effects.
This can be hard to chew on at first. Of course if we’re problem-solvers for our profession then we must logically define the problem and then deduce the correct solution, right? But look around. It mostly doesn’t work like that. Most invention, innovation and progress is the result of happy accidents and right-place-at-the-right-time. Not the result of an ordered, deterministic process.
So… does this mean we should all just go wild and hope to be lucky?
Not quite. It’s neither “just set goals/OKRs” nor “just do anything LOL”.
Instead: do the next right thing to affect the evolutionary potential of the present. Focus on getting clear about the reality of the current situation, then manage the constraints so that what you would like to have happen becomes more likely.
In this way, we allow an effect to come about. Not to aim for the effect directly, but to implicate it as a consequence of the conditions we create.
How do you wrap your noggin around this? Well, that’s where the Estuarine Framework comes in handy. We’ll go into the steps below, but it all starts with tackling the science behind how little we all perceive of our situation.
First, let’s grapple with inattentional blindness.
24 radiologists were asked to inspect a batch of scans, looking for lung nodules. This was a familiar task, something they’re impressively skilled at. Unknown to them, a picture of a gorilla, 48 times larger than the average nodule, was inserted in the last case. 83% of radiologists did not see the gorilla, even though eye-tracking showed they looked directly at it.
In summary, we don’t see what we’re not expecting to see, even if we look directly at it. In fact, the most any of us perceive is around 3% of the available data, at least in the West.
This is an evolutionary adaptation that’s generally been very good for us. When you see a large, striped animal moving swiftly towards you on the savannah, you don’t want to start with an exhaustive scan of all the available data and compare it against a best-practice database. You do what’s called a “first-fit pattern match” and you take swift action. This was effective then, but can cause us issues in the complexity of the modern world.
The bad news is that training won’t change this limitation. The good news is that adapting our methods can.
In the Estuarine Framework we use metaphors to come at the situation obliquely and increase the range of perspectives we view the situation from. The use of metaphor also increases cognitive load in the participants, which breaks us out of our usual patterns of thinking. And the process creates a very large number of data points at a fine granularity, so nobody’s able to see the big picture. This stops premature convergence.
Armed with all this, we can spot the gorillas!
OK, let’s get into the method. I’ve removed all names and specifics as everything was under NDA.
One interesting thing about this workshop method is that it works best if the facilitator isn’t an expert — or at all involved — in the details. You’ll see why in a moment. But you can also get a lot out of it by running it with your own team, as long as you resist any temptation to steer the group to what you would like the answer to be, which is a key tenet in this body of methods.
Workshop Day 1
We began with a couple of presentations of work the teams had done so far, peppered with questions from all sides. It was fascinating to observe this dynamic: four teams with different agendas politely but pointedly protecting those agendas. There were questions about various technical constraints and market situations, as well as a few rabbit hole discussions of possible issues.
One of the most exciting parts of the two days was observing how this cautious, prickly dynamic evaporated as soon as we moved into the Estuarine Framework.
It started with an introduction to the science behind the method, which I touched on above. Then Dave introduced the metaphor of the estuary, from where the framework gets its name.
An estuary is where a river meets the sea, and it’s tidal. The metaphor emphasises the complex and multiple flows of possibility in the system – a better match for the world of business strategy than a relatively static landscape, or recipe.
In an estuary the water flows in and flows out, so there might be things you can do only at the turn of the tide. Some elements might be stable, like a granite cliff, which only need to be checked rarely, while others, like sandbanks, could shift daily. As the water flows in and out, some elements might be covered or visible.
During the ensuing process, I noticed that Dave balanced two factors:
he rarely told the group exactly what was coming next, so it was impossible to “game” the answers
but he frequently explained how the method worked after each part, often contrasting it with more standard strategic planning methods. Much strategic planning focuses attention on a set of desired outcomes or a roadmap of projects with organisational cachet. In Estuarine, you start with the current situation and consider micro-projects you can undertake that shift the situation, making a wide range of desirable futures more plausible.
Drawing on the flip chart, Dave introduced the typology of constraints, using the geographical metaphor of estuaries. He started with robust constraints (e.g. sea wall) and resilient constraints (e.g. salt marsh), then expanded each into a triad, peppered with examples from around the Welsh landscape.
These sets of metaphors prompt people to look at the situation from more angles, so they can think of a wider range of constraints. It’s about inspiration, not categorisation. Whether a constraint is more one type or another doesn’t matter at all.
The group divided into threes. First, they had an opportunity to discuss possible changes to the metaphors they would like to use, in case others were more appropriate for them. Then the brainstorming of constraints began.
Important note here! Constraints in Estuarine aren’t necessarily negative things that must be removed, like in Goldratt’s Theory of Constraints. They can also be enabling – like how a skeleton or scaffolding makes new forms of building, movement or growth easier.
The discussion was instantly lively and engaged. After a few minutes, Dave added a third typology triad for teams to start brainstorming around: Business, Market and Consumer – a relevant typology in their context.
Everyone was stuck in and highly engaged. Walking round, I heard very varied conversations across groups, and everyone produced a nice big set of constraints.
Now we combined the groups and laid out a Time / Energy Map on a large table surface. Here, energy is a shorthand for many factors: money, resources, attention, ...
The group started adding and moving all their post-its around in the map space.
Dave encouraged everyone to notice when the constraints the groups had come up with were different from one another’s, and to note anything surprising.
When most of the post-its were down on the Time / Energy Map, Dave introduced three boundaries (You would normally mark these with pink ribbons, but we didn’t have any so improvised with laptop power cables).
First, the Counterfactual Boundary in the top right of the map. Which of these constraints are effectively unchangeable – like gravity, or a granite cliff?
Then the Liminal Boundary, for constraints you weren't quite certain were counterfactuals.
Finally the Vulnerability Boundary in the bottom left, for constraints with very low time and energy to change. This is a zone where things are volatile and possibly dangerous.
Monitors, Forward Scouts and Containers
We self-organised into groups to tackle the different zones within each boundary.
For clusters of counterfactuals, what Forward Scouts could be used to give advance warning of possible catastrophic change? Like checking for fault lines developing in a granite cliff.
For the Counterfactual and Liminal Boundaries, what Monitors could be put in place to check for constraints that crossed the boundary? Such events could be highly impactful.
And for any constraints in the Vulnerable Zone that would have high impact if they occurred, what Containers could be put in place to limit the damage?
From an observer’s perspective, this was fascinating to see. Without domain knowledge, I was stumped as to what any of the above actions might be, but the group quickly came up with several ideas.
Finally for the day, the group clustered the constraints and constructors in the middle of the map, also moving the map onto some hastily taped together sheets of flip chart paper, so we could bring the output with us for the next day.
Workshop Day 2
Micro-Projects and Ritual Dissent
On the Time / Energy Map, we brought focus onto the clusters of constraints in the middle of the map, between the vulnerability and liminal boundaries.
Now we self-organised into three groups. Each group selected one constraint cluster from the map and started to think of ideas that could reduce either the time cost or the energy cost of that constraint.
After a few minutes, Dave introduced the Ritual Dissent method.
This is a highly structured process designed to simulate the process of delivering new ideas to management or decision-makers.
“Any idea needs to be challenged, vigorously and early, not to destroy it but to make it more resilient.”
The groups enjoyed being put through it nearly as much as Dave enjoyed putting them through it. People later commented it worked really well to start the day with this kind of energy.
While coming up with ideas, each group selected a spokesperson. “Someone with a resilient and robust personality and unlikely to hold a grudge”. The group was told what was coming next.
After a few minutes, the spokespeople were asked to stand up and move to the next group around the room in a clockwise direction and to await further instructions but to say nothing. No chit chat!
Each spokesperson then had three minutes to present their idea while the new group listened silently. No questions or comments at this point. Next, they turned their back on the group. Through this, then metaphorically left the room, enabling the coming criticism to be depersonalised.
The groups then had 3 minutes to criticise the idea while the spokespeople remained silent. The criticism was not to be fair or reasonable, but instead a vicious and unprincipled attack on the idea (rather like you would hear around the water cooler after an executive presentation). No attack on the person themselves was permitted.
When the three minutes were up, the spokespeople returned to their original groups, without any further conversation with their attackers. They now reported back, explaining what they had learned. (“Whatever you do, don’t send me out there with half-baked ideas again!”).
The process was repeated again, but this time with spokespeople moving anti-clockwise.
More detail on the method is available on the wiki. I highly recommend trying it, but you need to pay attention to the details and get the orchestration and prompts right. It’s not fun if it goes wrong.
After 2 rounds of Ritual Dissent, each group stuck their ideas for micro-projects up on the wall, arranged in a line from low to high expected impact.
Dave explained that we were deliberately holding off from premature convergence by forcing separate groups to focus on one constraint at a time in isolation.
At this point, Dave had to dash off.
I repeated Ritual Dissent for another set of clusters and micro-projects, which also went up on the wall. I asked the group to move the Monitors, Forward Scouts and Containers from the map onto the wall too, in their own section. Then lunch.
After lunch, we deviated from Estuarine. If we’d continued the method, we would have grouped micro-projects into portfolios. This would give the group a set of next right things to do – what’s becoming known as the Frozen 2 approach to strategy.
Business Model Canvases
The client had planned the workshop around constructing Business Model Canvases, ideal for sharing the ideas with their colleagues.
I divided the group into pairs now, and each pair chose one high impact micro-project from the wall to start working on.
I walked them step-by-step through the steps of making a Value Proposition Canvas, and then a Business Model Canvas, all based around their take on the micro-project.
Once they were familiar with the tool, I introduced forced mutation.
I asked each pair to pass their canvas to the next pair. After a few minutes for each to digest, I asked everyone to select another micro-project or two from the wall, and remix the inherited canvas with those new inputs, letting the implications and new ideas pop up.
After this, the pairs shared the canvases and ideas they’d made so far, and we moved into the final stretch. I encouraged them to make more quick canvases, trying to introduce more “what ifs” and mutation to generate variety, in preparation to make one giant canvas each (I’d arranged for A1 size prints).
Each pair had half an hour to crystallise their giant Value Proposition Canvas and Business Model Canvas on the wall. The group knew they would experiment with a range of Value Propositions as this was an innovative new offering. The set they created here would form the basis for their experiments. The Business Model Canvas enabled them to quickly test the coherence of the different value props.
We spent a few minutes doing a gallery walk round, everyone adding questions and risks that occurred to them when reading other groups’ canvases. We finished with a round of presenting and discussing the canvases, after which everyone made final tweaks to their work, making the final output more resilient.
And that was a wrap.
If you’re involved in strategy, I highly recommend this approach.
The change in group dynamic across the two days was striking. There was a real sense of optimism and collaboration at the end of the two days.
And a lot of the thinking that emerged by the end of the workshop was stuff that nobody had been talking about at the beginning.
If you have questions, please reply and ask – I’m happy to nerd out about this.
With thanks to for editing and writing help, @fantomascarol for feedback, and the Cynefin Community for continual encouragement, questioning and lively debate.