Historically, two kinds of organizations put in the work to design their meetings.
The first are led by enlightened executives with a deep appreciation for how jangled things get when humans must work with each other's wackiness. They've given us methodologies like Open Book management, agile rituals, and Consent decision making.
The second operate in environments that are highly ritualized, highly regulated, populated by unpredictable characters, and where the stakes are high. Think parliamentary procedure, formal consensus, and emergency response protocols.
But most organizations never got around to it. Despite the complaints and calculators showing all the money wasted, bad meetings have never been bad enough to prioritize. After all, things still got done. Also, meeting design at scale can be a gnarly project.
Maybe next year, they say.
Only now, that year is here!
Not because people suddenly care about well-designed meetings.
Leaders are warming to these projects because the conditions that let them ignore wasteful meetings no longer exist.
Several forces are pushing these projects up the priority list.
1. Unstable teams AND rapid change.
When you've known everyone on your team for years, and the work you're doing today looks a lot like what you did last quarter, you can just talk to each other and get on with it. You have more conversations than meetings, and the few meetings you do have feel uncomfortably formal, like wearing a tie and heels at home.
But today's teams aren't stable.
If the work stayed unchanged, perhaps they could absorb some generational turnover, layoffs, and re-orgs. Throw widespread economic turmoil into the mix, though, and nothing's holding still.
Instability isn't news. The news flash is this: while you can't control what comes at you, you can design how your teams will deal with it.
Today's teams need a predictable way to flow through change. They need a way to quickly make sense of the situation together, identify their options, and replan as a group when the next obstacle appears. You could plop everyone in a room to talk it out, but this is wildly inefficient in practice. Which room? Who exactly? When? Because calendars are already full! Once you find a time, who leads? Who decides? What happened anyway?
Teams that assume an unpredictable future design ways of rapidly working through all of that. They design their meetings, they train people how those meetings work, and they practice. Surprises? No problem.
2. Flatter organizations.
Fewer managers + scale demands more process clarity.
The organizational structures of the industrial era create job and performance clarity by narrowly defining job roles and establishing a chain of command. This is super simple. Your job is to drive the truck. You coordinate with people who load and unload the truck according to the schedule set by your supervisor. Need something? Ask your supervisor. Is your supervisor a terrible person? Go to HR.
But now, companies think they don't need so many managers after all. Some think that maybe they don't need dedicated teams or so many employees either, because wow, there are sure a lot of contractors out there!
By jettisoning the industrial-era pyramid, organizations trade their rickety org-chart skeleton for more fluidity. In theory, this makes them nimble and ready to flow with change.
In practice, fluids need a container and a way to direct their form. Think of an octopus, which has no skeleton, but does sport skin, muscles, and "mini-brains" distributed through each tentacle. In successful flatter organizations (see Corporate Rebel's list for examples), the role of deciding what to do and how to do it moves from managers out to the arms of the network, where those closest to the action make decisions. And while every arm can act autonomously, the octopus is never torn apart because each one decides to take off in a different direction.
The octopus coordinates through its nervous system. Organizations have to build theirs.
Each team's collaborative ecosystem determines how people share information, make decisions, and alert folks when they spot a better path. To function well, that ecosystem must be designed. Then, people need the skills to navigate it.
Historically, these systems leak insights like juice through cheesecloth. Diana Hu explained it like this:
In the old world, companies basically ran as open loops. (...) Open loops are inherently lossy. (...) You needed middle managers and coordinators to route information inefficiently up and down an organization.
Diana Hu, The Playbook For Building An AI Native Company, Y Combinator
In the old world.
Which, if your company ever celebrated a birthday, is very likely your current world.
3. AI moves the work.
AI-native companies are flatter, wildly unstable (see #1 & 2 above), AND faced with a whole new set of problems.
Ann Miura-Ko recently visited several AI-native software companies. She found engineers and designers working with clients directly. She notes:
With this role collapse, a natural question arises: when you can build anything a customer asks for in a day, what stops you from building everything? Multiple founders named this as their biggest strategic risk. We call it the feature factory problem: speed of execution becomes its own trap, and the product slowly becomes a Frankenstein assembled from customer requests rather than a coherent vision.
Designing your workflow around AI capabilities means you worry less about production. Instead, you need some way to maintain coherence, relevance, and quality: the water previously carried by middle management. Which was just eliminated.
Take Dan Sirk, who found that with AI's help, he could act as a Chief Marketing Officer for two companies at once. But, while the marketing production work is faster than ever, he doesn't think he could serve more than three companies, because:
As A.I. makes the production of knowledge work more and more efficient, the job of presenting, debating, lobbying, arm-twisting, reassuring or just plain selling the work appears to be rising in importance.... And, of course, there are the meetings to hash it all out — many, many meetings.
That Meeting You Hate May Keep A.I. From Stealing Your Job, Noam Scheiber, The New York Times
We can sum that up as ensuring strategic alignment, which most organizations aren't good at... because in the old world, they didn't have to be. Momentum would carry them through rough patches, giving them plenty of time to corral wayward teams.
AI-native companies move too fast for that, which leaves them with three options:
Dictatorship: One person is appointed the Decider. Like a general on the field of battle, they hold the true vision. Any questions escalate to them. Lots of well-known drawbacks here.
Swarming: Like flocks of birds, the group agrees on a few simple heuristics that they constantly monitor. Action is taken with full awareness of the current state of the swarm, with everyone course correcting together in real time. Success requires always-on vigilance and a level of constant communication that humans suck at.
Of course, AI excels here once it has full context. For now, though, people are still in the driver's seat.Collaboration Design: Norms and rhythms for deciding what to do, syncing on what's changed, reviewing the AI-gleaned insights, learning, and deciding again. Practically, this means designating a single-source-of-truth repository for sharing information and a series of well-structured meetings.
I predict that most of these groups will start by trying to swarm, realize they can't keep it up, and then appoint a dictator. This will work until a) they scale, b) they fold, c) the dictator burns out, or d) the dictator drives everyone else out.
The wiser groups will opt for the third way, both because it works better for people and because...
4. AI needs the structure.
The AI industry promotes adoption maturity curves that peak in a promised land where the AI is operating across your business and harmonizing ALL the things: front-line support, strategic input from the board, market predictions, product backlogs, and who to call when you have a question for your suppliers. It is all feeding into a central repository where the AI can comprehend the full context, then make recommendations or take actions tuned to whatever is real right now.
This is scary, for sure, and also genuinely exciting. New options appear when we can easily use all that information that, in the old world, stayed locked up in separate heads and silos.
Of course, this only happens after the AI has access to all that data.
Leaders know this, so to make sure their company climbs that adoption curve and they get the return the AI peddlers promise, they ask the IT team to centralize all that data. For meetings, this means AI transcription, note-taking, and analytics. Companies that previously enforced strict "no recording" rules can't get those policies changed fast enough. That's good news, because many of those policies also blocked tools that make meetings better.
But transcriptions and analytics can only capture what's already there.
Do you know what it looks like when an organization that hasn't designed its meetings turns on the AI transcripts and dumps all that raw... output... into a central repository? I'll tell you. It's a hot, stinky mess.
Meetings with no clear purpose that meander across topics. A few people speaking in a room supposedly full of many. Decisions made, forgotten, revisited, made again, then ignored. Strategies shared by leaders that are never mentioned by teams. Lots of two-person meetings clogging the calendar, with transcripts abruptly disabled.
I'm starting to get requests from organizations that bought pricey licenses and soon hit the limits of what's possible within their old-world structures. They're recognizing that to get the benefit they've already paid for, they need to redesign the work.
AI needs clarity on what the strategy is, where the decisions live, the criteria used to make decisions, who to ask when questions arise, how to adjust, how to sync, how to learn, and how to judge when something's awry. That is communication architecture. That is understanding how knowledge, wisdom, and opinion flow through your organization, and that's meeting design.
So why now?
Not because people suddenly care about meetings.
Because they can't get the AI benefits they were promised without fixing meetings.
If you’d like to explore how we can help your organization, please visit our website here.

