How to kick-start 10 university groups in 6 months

Okay, so the title is a bit clickbait-y. But over the past months, EA Netherlands really has been trying to seed groups at all 12 universities in the country. Many people have asked us how we’re going about this, so we did a write-up of the steps we took and why we think this is worth our time. We hope this inspires other national community builders to start incubating groups as well!

A big thank you to Rob Gledhill, Jessica McCurdy, Jesse Rothman, James Herbert, Robert Praas and Joris Pijpers for providing feedback on this post!

Why do we want to do this in the first place?

The first Dutch university group (PISE) was quite successful. Since its founding two years ago, more than a hundred students have completed fellowships and more than twenty attended EAGx conferences. In addition, over a dozen members are currently truly enthusiastic about EA-careers. This showed us that Dutch universities are fertile ground for EA groups, and we wanted to replicate that success eleven more times.

Now, if you’re already familiar with why university groups are great, please skip over the list that follows. But if you’re not convinced of this idea yet, here are some reasons why launching university groups might be particularly impactful:

Universities are dense clusters of talented, flexible people. Historically, most EAs also first heard about EA during their early twenties. Furthermore, local groups in general are an important way for people to first hear about EA and to stay engaged thereafter.
Recent research suggests that most students who might like EA ideas haven’t heard of EA yet (spoiler: only 14.5% of EA-sympathetic students knew what EA is). Starting university groups that do outreach to their peers seems like a great way to fix this!
Over the past year, some amazing resources for running university groups have been created (e.g. GCP’s Handbook, Resource Center, OpenPhil). However, one of the main bottlenecks seems to be that many students who find EA interesting don’t know of any other EAs in their area. National and regional community builders are uniquely positioned to find and connect multiple of those EA-interested students, so they can start a group together—thereby complementing the existing services to launch and run university groups.

To this end, EA Netherlands set out to seed groups at all universities and prepare them for CEA’s University Group Accelerator Program (UGAP). Just to refresh your memory: UGAP is a program that provides new groups with a mentor, a stipend and game plan for their first semester. Handing the newly seeded groups over to UGAP after a few months ensures that they get specialized support during the crucial first year, while freeing up our time as national community builders to do other things!

So what’s the situation in the Netherlands?

Almost all Dutch universities are in the top 200 worldwide and most of them are even in the top 100. In addition, there is no university that is clearly more prestigious than the others (as their relative ranking often changes and is subject-dependent). As such, we believe that there is significant value in establishing EA groups at all Dutch universities, rather than at just a few.

In addition, student-run organizations are an important part of student life in the Netherlands. Being part of a board or subcommittee is very common, and is often viewed as an important part of your personal development. This means that there is a very good chance that groups will be able to continue to thrive after their founders graduate.

So this is where we are at now: at the end of 2021, there was only one university group in the Netherlands (PISE). Less than half a year later, there are ten budding new EA groups. Five of those have already finished running their first fellowships and multiple events (in Utrecht, Delft, Maastricht, The Hague and Amsterdam), two are running a fellowship this month (Eindhoven & Tilburg) and three are still in the planning phase (Groningen, Leeuwarden & Nijmegen).

Let’s give you an example of a timeline:

The process of setting up a group can go quite fast. For example, this is the timeline of the new group in Eindhoven:

May 4th: 1-1 with one of the founders
May 12th: first strategy meeting with 3 founders
May 20th: tabling at the campus
June 3rd: intro event (pizza and a presentation)
June 6th: start of the intro fellowship^[1]

So how did we get people to start groups in the first place?

So far, we’ve tried to kickstart each group by first finding one or two founders that were already EA-knowledgeable^[2]. We found these students through various channels.

Here’s a rough breakdown of the channels that we used to find (co-)founders:

3 from a national fellowship (with follow-up 1-1s)
5 from our personal EA-networks
3 through active outreach to social media followers
- 1 from a whatsapp group
- 1 from linkedin
- 1 from a facebook group
5 were connected to us by the international community
- 1 through 80,000 Hours
- 1 through CEA
- 2 through other community builders
- 1 through slack
4 contacted us about organizing
- 1 after a linkedin post
- 1 after involvement in PISE
- 2 through a contact form

We then worked with these founders to do basic marketing, organize introduction events and run a 4-week fellowship. The latter mostly served as a way to quickly get people up to speed with EA to become contributors or early members.

How did the groups recruit fellowship participants?

There are dozens of creative ways to find members, and our advice is to just try a combination of several ideas. Having said that, the strategies below seem to have been particularly successful in the Netherlands:

Hanging up posters all over campus (EA Utrecht found dozens of interested students this way)
Going to events and handing out QR codes for a WhatsApp announcement group chat there (EA The Hague filled a group chat with a 100+ students this way)
Perhaps the most successful strategy for every group was to promote their fellowship among friends and fellow students. However, this does have some obvious drawbacks (e.g. founder effect).
We recently experimented with tabling and the results look promising! In our case, tabling meant just putting up a table on campus and giving away hundreds of flyers and dozens of books. If you want to try this, I’d recommend to order a large variety of different EA books (e.g. Scout Mindset, How to Create a Vegan World, The Alignment Problem). This has a few advantages:
1. students immediately see how broad EA is in terms of cause areas
2. they take some time to flick through various books before deciding which book they want to take home, which gives you enough time to interact with them
3. it diminishes the ‘handing out bibles’ vibe and increases the ‘bookshop’ look.

Tabling in Eindhoven

What support does EA Netherlands offer?

We tried to make the process of starting a group as easy and motivating as possible. Here are some of the things we did:

We organized several points of contact:
- Semi-regular 1-1s with founders (for guidance)
- Biweekly group calls (for accountability & collaboration)
- A group organizer retreat (for team spirit)
- Encouraging organizers to apply for EAG (for international connections)
We also provided operational support:
- Small €750^[3] seed grants (because applying for funding can be daunting)
- We performed career-oriented intro talks at various universities
- We provided a list with other Dutch EA-aligned speakers for events
- We made the organizers aware of important resources (here’s a quick list^[4]) so that they don’t have to reinvent the wheel (e.g. the editable graphics on Canva) and are in the loop.

The university group organizer retreat

Lessons learned:

Incubating several groups taught us a thing or two about common failure modes. Here are some of the most important lessons we’ve learned:

As always, work in an ⁸⁰⁄₂₀ way! We can’t stretch this enough: don’t let perfection get in the way of action. There really is no need to have a polished strategy and several social media channels before starting your group. The bulk of the work comes from having a WhatsApp group, an intro event and a fellowship.
- Similarly, national community builders don’t need an elaborate plan to start seeding groups. Just start by doing some postering at universities or running group calls for potential founders.
Get a former university group leader to support you in this endeavor. They likely experienced the same first stumbles as the new organizers will, and now know how to address these adequately.
There’s a delicate balance between giving founders autonomy and providing support & efficiency. We initially thought that providing lots of creative freedom would be most motivating, but giving people an actionable game plan makes starting a group seem much more doable (which is also motivating). Of course it depends on the personality and knowledge of the founders, but over time we’ve shifted more towards the latter approach.
Quantity over quality is a serious concern. PISE experienced a strong push towards value drift when the new members outnumbered the EA-knowledgeable members in the first year. It’s therefore that the new board made the 4-week fellowship mandatory. Similarly, bringing enthusiastic but not-yet-EA-knowledgeable students on as co-organizers^[5] can be appealing (and may have some benefits like diversifying), but it can also seriously threaten the direction and decision-making of the group. It’s really important to discuss this trade-off with founders when they are looking for co-founders.

Of course, this leaves one burning question—how do we maintain quality of groups?

We think that searching for founders who are already EA-knowledgeable^[6] helps with this significantly. In addition, my experience is that getting less-involved people to organize actually leads to less of the problems described in the recent bad omens post; Having newer people organize means that they are often still pretty sceptical themselves, rather than just parroting what the people in the EA hubs say is most important. In addition, we’re trying to mitigate the concerns around scaling up by:

Keeping an eye on diversity. Right now, the proportion of organizers who are CS/math students is lower than in EA on average and the proportion of organizers who seems to fit the ‘talent gaps’ that the author describes (e.g. people with high emotional intelligence or potential policy experts) is much higher.
Doing regular check-ins with groups and coaching weaker groups a bit more.
We try to both inform people about concepts as HEAs (so that they are in the loop) as to open up the discussion about how the metric should be used and what we’re actually trying to achieve.
We chat to people who attended events (especially to ones who didn’t attend further events) to ask for feedback.

We’d love to hear more ideas for preventing bad situations, so please let us know if you have any suggestions for diminishing these problems further! Here’s a form for if you’d like to do so anonymously :)

So all in all, what can national groups do?

Through your network and position, I believe that national/ regional community builders are particularly well-placed to do the following:

Search for founders & hook them up with co-founders.
Give people a ‘hero license’! When I first got into EA in 2016, I didn’t even realize that I was allowed to start my own group. But when I knew that I had the buy-in from EA Netherlands (thank you Chris Szulc!), I felt licensed to finally take initiative and start PISE.
Preparing career-oriented introduction talks to perform at various universities.
Create 1 big national team! Hosting regular touch-points for organizers (e.g. group calls, retreats) is very beneficial for knowledge sharing and team spirit.

You’re an important part of the puzzle!

We think that every student should have the chance to hear of EA, and we believe that we can make that happen quite soon :) While we’re not responsible for making someone EA, we can at least try to make sure that students at every university have a chance to engage with EA. There are great opportunities available for university group organizers, and national community builders can complete an important piece of this puzzle by putting the teams together in the first place!

Last but not least: a massive thank you to all the amazing university group organizers in NL for rising to the challenge of founding groups. We’re super happy to have you all on the frontiers of EA in the Netherlands!

^
We decided to go with the 4-week fellowship because it’s a smaller commitment for students. The only drawback is that we had to cut on longtermist content significantly, which may be reflected in the amount of interest in longtermism.
^
As we aimed to get groups into UGAP, we loosely used the same threshold as UGAP for founders being EA-knowledgeable enough . That is, would someone would be accepted as a facilitator for the virutal intro fellowship? In addition, we considered whether founders were sociable/representative (‘would we be interested in joining this group if it consisted mostly of people like this founder?’), proactive and willing to learn more.
^
The €750 went a long way for organizing events and creating promotion materials. However, registering as a foundation at the notary (which is often a requirement to getting access to facilities from the university) often costs between €500-800 as well. We’d therefore recommend providing seed grants that are a bit larger (e.g. €2000).
^
Here’s a quick list of the resources that I’ve found particularly useful to share: https://docs.google.com/document/d/19cziEX-LTm5GjD7EMsJ02wvpby4VIbO0JZZ7EqTLeoM/edit?usp=sharing
^
To EAs from outside of the Netherlands it may sound unusual for students to want to organize an ‘EA group’ when they don’t have much experience with EA content in itself yet. However, due to the status that being a board/ subcommittee member has in the Netherlands, it’s not uncommon for students to be interested in just organizing activities more than in doing a fellowship or reading up about EA to figure out whether it suits them. For example, about a third of the subcommittee members at PISE in year 1 had never done a fellowship or an equivalent thereof.
^
In the Dutch context, there seem to be a lot of ‘latent EAs’. Here I don’t mean the so-called ‘insta-EAs’ (people who EA immediately resonates with when they find out about it), but people who were already engaging with EA resources online, yet weren’t connected to the community. My guess is that in other countries that previously had little EA-infrastructure, there are also many people who’ve been reading about EA for years, but didn’t engage with the community side yet. Case in point: before starting PISE, I was a ‘latent EA’ myself (between 2016-2019). Similarly, several of our current group leaders told us that they were interested in EA long before getting in touch with us.