Open Space Technology: Letting the Room Set Its Own Agenda for 'What Should the LMS Become'

I was asked to plan an agenda for a full-day session on “what should our LMS become over the next two years,” and I turned the request down — not because the question wasn’t worth a day, but because any agenda I wrote in advance would already be wrong. I couldn’t know, sitting alone at a desk the week before, which five or six topics actually mattered enough to the twenty-two people attending to be worth a room’s collective time. Only the room, on the day, standing in front of each other, could know that. So instead of an agenda, I ran Open Space Technology.

What it is

Open Space Technology (OST) is Harrison Owen’s answer to a specific, recurring failure of conference and workshop planning: organizers spend weeks building an agenda of sessions they guess people want, and then discover on the day that the best conversations happened in the hallway between sessions, not in the sessions themselves. Owen’s insight, after noticing this pattern repeatedly, was to stop guessing and let the agenda be built live, by the actual participants, on the actual morning, based on what they actually care about right then — with a light, well-understood set of rules that keep the resulting chaos productive instead of aimless.

How it runs

1. Open with the theme, not an agenda. State the broad question the day is organized around — here, “what should our LMS become” — and nothing more specific. No pre-built session list.
2. The marketplace. Anyone with a topic they want to convene a discussion on stands up, announces it in one sentence, writes it on a card, and posts it to a wall with a time and location slot. This happens for as long as people keep proposing — often ten to twenty topics from a room of twenty, with no gatekeeping on who’s allowed to propose.
3. The Law of Two Feet. The one rule that makes everything else work: if you’re in a session that isn’t giving you value or learning, you have not just permission but an obligation to leave and go find one that is. No session convener can expect a captive audience, which means sessions that don’t deliver simply empty out, and that’s treated as useful information, not an insult.
4. Self-organizing sessions run in parallel, typically 45–60 minutes each, across as many rounds as the day allows. Whoever proposed a topic takes light responsibility for it — mostly just making sure someone captures notes — but nobody “runs” it in a top-down sense; it’s a conversation among whoever showed up and stayed.
5. Convene, harvest, close. At day’s end, the whole group reconvenes and each session’s notes get shared, often through a rapid report-back or a posted document everyone can scan. Themes across sessions frequently converge even though nobody planned for them to.

A full OST day for twenty-plus people typically runs three to five session rounds, with breaks and a closing circle, and the entire structure rests on trusting that “whoever comes are the right people, whatever happens is the only thing that could have, whenever it starts is the right time, and when it’s over it’s over” — Owen’s four principles, which sound glib until you watch a room actually operate by them and notice how much anxiety they remove from people who’d otherwise be worried about being in the “wrong” session.

Applying it: the LMS’s future, set by whoever showed up caring

The room included instructional designers, engineers, a couple of merchant-facing support people, and two people from the business side thinking about pricing and packaging — a genuinely cross-functional group with no obvious single owner of “what should the LMS become,” which made it a good fit for a self-organizing agenda rather than one person’s pre-planned session list.

The topics that got proposed in the marketplace were revealing precisely because nobody had asked for them in advance. An instructional designer proposed a session on whether the platform should support cohort-based, time-boxed courses alongside its existing self-paced model — a genuinely open strategic question nobody had put on any roadmap, but clearly something she’d been thinking about for months and had never had a forum to raise. An engineer proposed a session specifically on whether the platform’s content-authoring format could support interactive code exercises, not just video and quizzes — a technical curiosity that turned out to connect directly to a merchant request the support team had been fielding for months but had never routed to engineering as a strategic question rather than a one-off feature ask. A support person proposed a blunt session titled “why do merchants keep asking us if this is ‘a real LMS’ — what does that even mean to them,” which turned into one of the best-attended sessions of the day once people realized it was really a session about competitive positioning and market perception, dressed as a customer-support anecdote.

None of these three topics would have appeared on an agenda I wrote in advance, because I didn’t know any of them existed as live concerns until the room told me by standing up and proposing them. That’s the specific value OST delivers that a planned agenda structurally cannot: it surfaces the actual, current, felt priorities of the specific people in the room, rather than the planner’s best guess at what those priorities probably are.

The Law of Two Feet in practice

The most instructive moment of the day was watching the cohort-based-courses session start with eight people and end with three. That’s not a failure of the session — it’s the Law of Two Feet doing exactly what it’s supposed to: five people realized within the first fifteen minutes that this particular question, while interesting, wasn’t where their energy or expertise was needed, and they left for other sessions without anyone treating it as rude. The three who stayed were the three people who actually had something to contribute to that specific question, and the conversation that followed was sharper for their being a genuinely engaged three instead of an obligated eight.

Compare that to a traditionally agenda’d meeting, where those same five people would have sat through the full session because leaving felt socially costly, contributed little, and left the meeting having wasted an hour they could have spent in a conversation that actually mattered to them. OST’s rule structure makes disengagement visible and acceptable instead of hidden and resentful.

What came out of the day

The closing harvest surfaced a genuine, unplanned convergence: three separate sessions — cohort-based courses, interactive code exercises, and the “is this a real LMS” positioning conversation — had independently arrived at a shared underlying theme: the platform’s self-paced, video-and-quiz-only model was hitting a ceiling with a specific segment of more serious, skills-focused course creators, and none of the three groups had coordinated to say so; it just showed up three times from three different angles. That kind of convergent signal, appearing spontaneously across independently-run conversations, carries more weight in a strategy discussion afterward than any single person’s opinion would have, and it’s exactly the kind of pattern a planned agenda — which would have kept these three topics in separate, non-overlapping time slots by design — might never have revealed as connected.

Failure modes and when to skip it

OST fails hard in a room that doesn’t trust the Law of Two Feet — if leadership is present and people feel their attendance in a given session is being watched or judged, nobody actually exercises the freedom to leave, sessions become captive audiences again, and you’ve recreated a traditional agenda with extra steps and less structure. This needs either genuine psychological safety already present or an explicit, credible statement from whoever holds power in the room that movement is expected and fine.

It’s also the wrong tool when the day needs to produce a specific decision rather than surface priorities — OST is built for divergent exploration of a broad theme, not for converging on “which of these three roadmap items do we build first.” Running it when what leadership actually needs is a decision just produces an energized, well-documented day that still ends without the decision made, which reads afterward as a missed opportunity rather than a successful one. And it needs a genuinely broad theme with no single right topic — a narrow, specific question (“how do we fix the checkout bug”) doesn’t have enough breadth to generate a real marketplace of divergent topics, and forcing OST onto it just produces three sessions all restating the same narrow question from slightly different angles.