How to run an Idea Stampede to make a group gallop

We make the best decisions when we take into account a number of diverse perspectives. And people engage and execute better when they're part of the decision-making process. But it’s hard to have discussions and make decisions in a large group. It’s hard to make sure that everyone feels heard, that the conversation moves at the right pace and in the right direction, and that the best ideas bubble to the top. Most meetings with a large group end up feeling like a waste of time to many of the participants.

I remember all these thoughts running through my mind as I sweated quietly to myself, watching the tenth face join the Zoom room. I was about to run a tricky session as a consultant. I needed to somehow generate a prioritised list of potential ideas, flag any potential roadblocks, and get the group excited and gelled. In 90 minutes, over Zoom.

I knew this would be tricky, for a few reasons. To do a good job of generating and prioritising ideas requires bringing everyone’s unique perspectives together, even though those different perspectives were a potential source of conflict. Previous conversations had got bogged down in detail and tangential discussions. And I was conscious that - as a consultant - I was an outsider and lacked a lot of the context that they all had as employees.

I decided to take a risk, and introduce them to the Idea Stampede. It’s a technique that I’ve been evolving for a little while, which ends up being surprisingly fun and effective for generating and prioritising ideas^[1]:

1. Start by introducing the problem, discussing why it’s important, and clarifying the goals and constraints.
2. Share a blank Google Doc ^[2], and write down the problem description and goal statement in a couple of sentences at the top. Check that everyone’s happy with it, especially the key stakeholder.
3. Set a timer for 15 minutes or so, then get the group to follow these instructions:

• Scribble your ideas in the doc, all at the same time. Even stupid ideas. Don't worry about prioritisation.
• Do this in silence (though jokes or anything that raises the energy are always allowed).
• Express each idea clearly enough that someone else could definitely understand what you mean just by reading it. Maybe explain why it’s a good idea. Be concrete and descriptive (examples are good!).
• As the moderator, keep an eye on what’s happening, and refine the instructions as needed. Highlight anything unclear so that the author (or someone else) can improve it. End early or add more time as you see fit, but don’t wait until everyone has completely run out of ideas - you want to finish when there’s still a little frantic energy in the room.

4.  Now it’s time for some reflection. Instructions for the group:

• Read over all the suggestions, still in silence.
• Move the best ones towards the top.
• If you see two or more that are saying more or less the same thing, then merge them together into a group.
• If you don't understand an idea, ask for clarification in a Google Docs Comment (or just in bold next to it), but silently.
• Feel free to modify to extend or improve, or add to anything you see.
• But you can't delete anything. And you can't change someone else's idea. And you can't move it down.
• Stop when it looks like people are starting to slow down.

5.  Now, we’re ready to talk a little. Starting from the top, ask someone to speak up on behalf of each idea in 30 seconds or less.

• Let others chime in, but keep things moving along quickly.
• Perhaps allow listeners to improve the text as it’s being discussed in small ways as a form of active listening.

This first stage maximises the diversity and inclusivity of ideas. Because everyone’s working in parallel, you maximise the output. Because it’s in silence, you’re freed from the tyranny of the loudest voices. Because each idea gets a proponent, everyone gets a chance to speak. Because of the anonymity, there’s no bias towards the HIPPO (the highest-paid person’s opinion) in the room.

By this point you’ll have a long, creative, and roughly-prioritised list of suggestions. Great!

Now, stop and take a break. Give everyone five minutes (with a specific time, and a be a strict time-keeper) to make a cup of tea, so that they can come back refreshed for the next phase.

There are a number of ways you can go from here:

1. Pick the top few ideas, and divide into small groups to think in depth about each one, and then present back to the full group.2.
2. Give the key stakeholder(s) a chance to guide or focus the discussion towards a few ideas.
3. Have a group discussion, explicitly encouraging debate. (A 2003 experiment at the University of California–Berkeley, conducted by Charlan Nemeth, set out to examine the effects of different kinds of group idea generation on idea creativity. One group was asked to answer a problem with no other instructions; one group was asked to brainstorm with no criticism of others' ideas allowed; and one group was asked to debate one another. The last group outperformed the other two by a significant margin when it came to coming up with diverse solutions.)

4.  Do another round of silent-parallel-brainstorming^[3]. Perhaps ask people to drill down into refinements of one or two of the best ideas. Or a pros/cons list. Or to flag potential concerns.

Finally, it’s time to harness the energy that has gathered in the room. Ask people to put their name next to ideas or actions they’re excited about, and define the next step. For example, perhaps each team is charged with investigating their idea, and presenting back to the stakeholders to decide about allocating budget. Or nominate someone to chase everyone up. The key thing is to have names next to actions and a process for making sure that progress gets made!

I’ve tried variants of this approach with various groups, and I’m continually surprised by how well it works. You’d think everyone will trample over each other when typing, but after some initial confusion and giggling, people make space for one another. You’d think that allowing anyone to move ideas upwards would result in competitive self-promotion, but it doesn’t. And as a nice bonus, it works equally well in-person or over remote video.

The Idea Stampede worked beautifully with my group of a dozen disparate stakeholders. In 90 minutes, we co-produced a pretty comprehensive and prioritised list, formed a collective sense of what was most valuable and feasible, and discovered alliances of shared excitement that planted the seeds for future cross-functional teams.

At the end, we left with a rich document, and a sense of collective ownership and energy that we could easily turn into action. A 90-minute Idea Stampede is usually more fun and more effective than 4 hours of out-loud discussion would have been.