Report from Progress Conference 2025 In Berkeley
And thoughts how to make conferences even better
I was in Berkeley at Lighthaven for the Progress Conference, a two-day gathering on how to accelerate progress in society. I got to take part and help out a bit. It’s a really great conference, with lots of attendees saying it’s the best conference they’ve been to. I had an amazing time and was genuinely honored at the chance to work with these guys.
A big part of what I do for a living is advise people on how to make their conferences better. I wanted to use the Progress Conference as a case study for ways to do a conference right and also to help clarify a vision of how conferences can and should be even better. (For more of my thoughts on this, see my essay Everything You Did To Make Your Conference Better Actually Made It Worse.)
Some of the many things the Progress Conference did well
Here are a few things they did that made the event great, that are lessons for other conferences.
Have a clear goal: In my work, I see many conferences where the organizers can’t articulate goals at all. (“It’s our annual conference, and it’s been almost a year since the last one” does not count). Jason Crawford, the organizer, is laser-focused in what he wants the conference to do: He wants to gradually build up a movement, over several years, to advocate for more technological and economic progress. He’s incredibly clear about where the conference fits into the creation of this movement. That clarity of vision makes a huge difference.
Gather a community of people who have mutual excitement to be together: A conference is, at its root, about bringing people together. The best conferences bring together people who really want to be together. In this case: there was a simmering interest in Progress Studies (from this article, and this blog) but as far as I know, this conference (started last year) was the first time people who shared this interest had been brought together. People here were really excited to meet.
Curate your attendees: A conference is, at its heart, a gathering of people. I’ve seen many conference planners put hours of thought into the venue, into who the speakers will be into the meals but not give any thought to who they want to attend. The attendees are the conference. Failing to think about who the attendees are is like throwing a party but not giving any thought to who the guests are.
The main way people wound up at the conference is that they were individually invited to attend. Those who were not invited could apply, but only a fraction of those were admitted. This really helps ensure that the group who attend are amazing, and excited to be there. I don’t think every event should be invitation-only, but I think not enough are. One thing that made the event amazing was that everyone there was a really top-tier person.Prioritize conversation: The event took place at Lighthaven, a space that is optimized in a million ways for casual conversation: There’s lots of space and nooks and crannies to hang out in. There are snacks everywhere. The grounds are beautiful. (You can hear Lighthaven boss Oliver Habryka and Patrick McKenize talk about this on a recent Complex Systems Podcast episode) The conference organizers strongly encouraged people to not go to sessions the whole time, but to hang out in the space and talk to people. The conference felt like just a huge 12-hour-a-day cocktail party where, if you wanted, you could occasionally take a break from talking and go listen to a talk.
Lots of little details: A million little details were generally done right: The food was good and served buffet-style in a way where waits were never too long, the emails were clear, the schedule was well-presented. Nametags were so well designed that someone posted a twitter thread about them. All of these were done in ways that served the goals of the conference. (Buffet-style lunch encourages mingling, clear name tags make it easy to identify people, etc..)
The end result was a really successful event.

A three-level taxonomy of conferences
Spending time at the event, and talking to people there, I found myself dividing conferences into a three-level hierarchical taxonomy:
Level 1:
Get people together, create obstacles to connection
(This is what most conferences do)
A level one conference brings people together and then forces/encourages them to sit quietly in one-way information delivery sessions all day. This is what you would do if you wanted to stop people from meeting each other. I’ve been to lots of events where it is someone’s job to herd people who are talking in hallways into sessions. Most conferences, amazingly, do this. (See Everything You Did To Make Your Conference Better Actually Made It Worse.)
Level 2:
Get out of the way, to allow serendipitous connection
A level two conference has lots of breaks, and lots of unstructured time. There are lots of nice nooks and crannies. The idea is to let serendipity do its work. The Progress Conference did this as successfully as any event I have been to. It was amazing.
Level 3:
Intentionally design for the connections you want to create
Social interaction at conferences is so important! Amazingly, even the good conferences mostly leave interaction to chance. We don’t do this with other things we care about. A level-three conference is one that takes steps to actually solve for the sorts of connection it wants to create rather than just leaving it all to chance.
You might want to help attendees find the right people to talk to quickly, or help integrate new people into what can feel like a cliquish group of insiders. You might want to create conversations that are more likely to lead to follow-up, or balance the needs of introverts with more outgoing participants, or the needs of high-status members of your community with more junior people. A level three conference recognizes that these are all engineering problems. There is no reason to accept leaving them to chance.
A few examples: Unconference sessions or roundtables that let people quickly find people who share their interests. A physical bulletin board in the lobby where people can post asks and offers to get people to help each other solve problems. Facilitated discussions that encourage deeper talk. Matchmaking apps that pair people up by interests. Working groups to co-create artefacts. Facilitated small-group discussions on important topics. There are a million answers, but these require that you begin by asking the question “how can we make this better”
These sorts of solutions take thought and care. Badly facilitated activities can be worse than no facilitation at all. And you definitely don’t want to engineer serendipity entirely out of the equation- a well designed event has a balance of intention and chance. But thoughtful design can accomplish way more than just trusting that social interactions will sort themselves out on their own.
Level three activities at the Progress Conference
Here are some solid level three things that the Progress Conference did:
They had a directory of attendees: Most conferences publish a list of speakers, but not attendees. The progress conference had an online attendee directory, where everyone could share information about who they are and what they do. It was tagged and searchable, so it was possible to pre-select people you might want to talk to, and to find contact info for people you did talk to and wanted to follow up with.
They had space in the agenda where anyone could add a session: The conference schedule had pre-programmed sessions for their official speakers, and also space where anyone could add a session. This makes it easier for people to self-organize around shared interests, and generally unlocks a lot more agency in attendees, which in turn builds energy and community
They had an opening session (run by me!) to help participants find their people quickly: We did a session on the first morning where I got participants to talk to each other, and, most importantly, find others who shared their interests. I asked people “What’s something that if you get to talk about it, you’ll feel your conference was a valuable experience” They I had people write their answers on signs, and walk around, clustering into groups of shared interest. They got to quickly finding some of the people they most needed to talk to, instead of hoping to randomly bump into each other by the end of day two (or never).
They had a Slack for all attendees: This was an online space where people could talk, coordinate ride shares, introduce themselves, and more.
These were great! That said: even at a top-tier event like the Progress Conference: Most of the interaction was still at level two: lots of wandering around from group to group, hoping to find the people you really need to talk to. It was so, so much better than other events. But I also believe that with a bit more design, it could have fostered even more effective and meaningful connections than it did, and unlocked a lot more value.

Why do so many conferences operate at level one, and what can be done about it?
It’s a source of real puzzlement to me why the culture of conferences is the way it is.
The majority of conferences are at level one, where they mostly get in the way of connection. People seem to know that this is bad- it’s not a secret. Things like unconferences and Open Space meetings and Liberating Structures have been around for decades, but for some reason, people keep running level one events. I’m trying to get a better understanding of why this is- it seems a little mysterious and fascinating, and my hope is if we got better understanding, it might help fix it.
As an example: I see lots of companies that spend huge amounts of money to bring to together large groups of very expensive leaders in their organization, and then not do anything to encourage them to talk to each other. The value of these conversations in many cases should be in the tens of millions of dollars: New collaborative opportunities identified, inefficiencies recognized, the effects of silos weakened. But instead they listen to informational talks that people grudgingly endure, and all that potential value gets left on the table. This is much more the rule than the exception.
Oliver and Patrick talk about this a bit on the recent Complex Systems. (They were also both at the Progress Conference. I got to talk to each of them a bit, which was a highlight). The main explanation they give, I think, is that the person running the conference is many levels down in the org chart from the person who cares about the conference ROI. But why is this true? The CEO cares about the conference, and it’s expensive to bring people together, and the company could get *SO MUCH* value out of it that gets lost. Why does it get left to a random junior person? Or why don’t more CEO’s say “go run something where people actually learn and connect”
Ivan thinks it’s about status, which seems probably partly right but I don’t think is the whole story. I think there’s something about legibility (in the “Seeing Like A State” sense of the word)- it’s easy to know what happens in a plenary talk, less so in a hundred informal conversations. I talked to a few other folks on X, and there are some good ideas, but I’m still not sure I get it. I hope to think/write about this some more. If you have thoughts, I’d be really interested to hear them.
Conference progress
Conferences matter a lot. I’ll keep making it my job to make them better. Getting to participate in amazing events like this one keeps me inspired.


Everything you say sounds like the way effective altruism conferences have been run for years :)
In 2018, me (a volunteer facilitator) and one of the two main hosts dragged the keynote speaker into a room for an ad hoc session that eventually led to the Happier Lives Institute!
Great Article.
Here is my cynical take on why most conferences are “Level 1”: the purpose of most conferences is *not* to form connections between people. It is to get people to buy into something: a product, a corporate strategy, etc.
Most conferences I attend are industry conferences, where many of the attendees (sellers) are there to sell products to other attendees (buyers). Many of the speakers are sellers who have paid the conference for the opportunity to speak: no wonder the conference organizers want to fill the room. In theory, the buyers are there to learn and make connections. But in practice, a lot of them are motivated by taking a day off of work, traveling somewhere exciting (whether a different city or just their local downtown), getting wined and dined, etc.
Another category of conference you mention are corporate events: where everyone from a company comes together for a day or two. While part of the goal here is to build connections across the company, the more essential goal is to get people to buy into the corporate culture/strategy/mission. Presentations are coordinated by senior execs to present a cohesive message to the broader team. Applause is expected. Dissent is discouraged. It's more of a political rally than a seminar. The point of bringing people together in a room for this is to make it harder for people to misunderstand or (god forbid) disagree with the senior team’s point of view. The last thing you want is people mingling in the hallways and thinking up new ideas that conflict with what the exec committee has spent the last 6 months deciding on.
This is obviously a very cynical take. And no doubt that there are a whole host of events that would benefit from your advice. But I think a lot of the reason you see events that don’t seem designed to encourage connections is because that isn’t the goal.