Liberating structures in action

June 2, 02016 · Facilitation Liberating structures

The previous item introduced the idea of liberating structures—what they are, why they're interesting, and how I've grown to use and value them. If you're feeling liberating structures WTF, please read that one first.

This time: a tour of my favorite liberating structures. These are the "microstructures" I keep coming back to, with notes on use and value.1

But first, an important caveat

When facilitating with liberating structures, please, for the love of all human species present and departed, do not use the words "liberating structures". Avoid using the name of a particular "microstructure". Not much preamble is needed: share some notion of the end product or state people are moving towards, and then clearly state the immediate next step everybody must take.

#1: 1-2-4-all

This is the bedrock. 1-2-4-all is the answer to the question, "how do I get all these people talking to one another?" No matter how the chairs are organized, or how defensive people are when they come in, this is a starting point. If you can't present your best ideas to the room, you can develop them in a small group. If you can't do that, you can share your thinking with one other person. If you can't do that, you can think about something for a moment, by yourself. In 1-2-4-all, everybody has a chance to do each of these. It never doesn't work.

#6: TRIZ

TRIZ is a scavenger hunt for things a system is doing that are actively keeping it from fulfilling its purpose. It's fun and terrifying. In TRIZ, a group will:

  1. create methods for consistently ensuring an unwanted outcome;2
  2. list things they're already doing that are very similar to these; and
  3. commit to stopping doing those things.

TRIZ is the great antidote for organizations that have trained their people that everyone is too busy to change or improve, or that punish imagination.

#7: 15% solutions

My friend André asks people to think about what they can do with the resources they have and the authority they have, today. This is the core of 15% solutions. Even if 85% of a system is "somebody else's problem", or profoundly and irretrievably fucked because of management or the software or the customers or the county board or regulations or [INSERT FORMLESS CHIMERA OF ACCOUNTABILITY HERE], there's still that 15%.

What can you do with that obscure, limited amount of flexibility in order to improve the situation?

Another way to think about 15% solutions is that if you do it—and you keep at it—you have cracked the lid of continuous improvement. Smaller is better.

#8: Troika consulting

A method for practicing two very difficult and mega important activities:

  1. Asking for help.
  2. Listening for and responding to a request for help.

#12: 25/10 crowdsourcing

25/10 crowdsourcing is a machine for generating bravery. It starts with a question: "If you were ten times bolder, what big idea would you recommend?" It proceeds through a process of hanging out and sharing distributing ideas, while scoring them in a secret ballot format. And it ends with a tantalizing display of the weirdest, least intuitive, bravest ideas that were elevated by the whole group.

#14: Min specs

Min specs ("minimum specifications") is a workhorse. This is my go-to for software specifications3 or other circumstances where nobody has thought about what it would mean for this to be successful—or how we would even know if it was a success. Min specs is a way of lazer beaming a group's focus onto the very few critical constraints that will implode an effort if they're not satisfied.

I was working with a group that had hundreds of "must-haves" characteristics for its software product. Before the meeting we'd categorized them and printed them out in tiny print. I thought I'd mount them in a little area of the room so people could see; the printouts ended up stretching across two entire walls. After a min spec exercise, the group had gotten that down to less than ten. That's almost few enough to keep in mind when creating software!

#21: Design storyboards

Look, you are aware that meetings and other conversations can be designed. That you can come in swinging, not with an agenda, but with a plan; with what Dan Klyn calls choreography.

Design storyboards are how you deploy this awareness out to the other people you work with, or spend time with. With a design storyboard, you work together to identify the structure of how the meeting will go. By getting everybody involved in structuring the time they have together, you can neuter so many objections. You can transition people from status DGAF to status GAF. The risk of doing this is that a group can crawl up its own butt—but I think it's well worth that risk.

#22: Celebrity interview

Age old quandary—what to do with a subject matter expert (SME)? Especially in a continuous improvement environment where the "doers" are the real experts, and the "worders" need to stay out of the way? A celebrity interview to be a fantastic way to get a SME situated relative to the rest of a group.

Celebrity interview is a good example of chaining liberating structures: note that it uses a 1-2-4-all to generate the group's questions for the interviewee.

#25: Open space technology

Last, but certainly not least.

It's strange for me to think of open space as a liberating structure—my personal experience participating in open space technology events mostly occurred before I'd ever heard of liberating structures. But I'll be dipped if the shoe doesn't fit.

The items in this list are in given in order from simple to complex. Open space technology is last. Here's the deal: don't do this first. Maybe don't do it ever. If you do do it, don't half-ass it.

Once you get a taste of what it's like to breathe, communicate, and work in open space, you'll never want to stop. The good news is that you won't have to—but that's a story for another day entirely.


  1. Advanced dorks will note that they're given in ascending liberating structures menu order.

  2. Fake examples: all sales prospects are lost; 100% of patients are infected.
    Real examples: every release of software breaks features a key client depends on; no-one can find the information they need via search.

  3. Why folks want to design software from a specification is beyond me.

Next item
    PDCA WTF

Previous item
    Liberating structures—the idea