Thoughts on outreach

Many effective altruists want to spread EA ideas. As co-president of Harvard Effective Altruism for the last two years, I have some experience in this area that I thought might be helpful to share. My experience is specific to Harvard Effective Altruism, but the lessons generalize to creating other groups of like-minded people from scratch.

Word of mouth

When I first took over HEA, I hoped that we’d be able to spread by word-of-mouth—that I’d find a few friends who seemed like good fits and get them as excited about HEA as I was, and then they’d recruit their own friends and the cycle would repeat.

But that didn’t work very well—we didn’t have enough members for us to grow quickly using this strategy. And even if we did, most of them probably wouldn’t have been dedicated or enthusiastic enough to recruit other people to the club. In general, it seems like many people who get interested in EA through word-of-mouth show a lower level of enthusiasm than the person they heard about it from, so as you go more degrees away people’s enthusiasm tapers off quickly.

It’s quite possible that in future years or with more people we could make word-of-mouth work better, but at least to create the initial seed group it wasn’t very helpful.

Personally recruiting friends

I tried personally to get a number of my friends interested in effective altruism. Initially, this worked pretty poorly. People would either come to one meeting of HEA and then stop going, or would continuously make excuses until I stopped asking them.

However, eventually many of my friends did get more interested in EA, seemingly without too much encouragement from me. This leads me to think that being “pushy” about talking about effective altruism is actually counterproductive; I’ve had much better results from not actively talking to people about it, but not being shy about mentioning it when it comes up, and generally showing by example to how awesome and fun effective altruism is.

Passive discoverability

We also tried using more passive outreach channels: we set up a booth at the “extracurriculars fair” at the beginning of the year, asked THINK to list us as a chapter on their website, had our events listed in various directories, and so on.

I’m quite happy with these strategies. Perhaps surprisingly, our booth at the extracurriculars fair is how we found two of our most dedicated members, who both now sit on HEA’s board. THINK also sent a few enthusiastic people to us, and was fairly low-effort from our point of view. And an event we held during Harvard’s visit weekend for high school seniors, which was advertised only through a handbook given to them at the start of the weekend, attracted a surprisingly engaged and throughtful group.

The Philanthropy Fellowship

Our main project for the past year was running a speaker series called the Philanthropy Fellowship. Fellows applied to attend dinners with various speakers, some from within Harvard but many from outside. At the end of the semester we picked a project to do as a group: in the fall we fundraised for GiveWell’s three top charities, and in the spring we did a small randomized controlled trial on the best ways to get people engaged. (Data for the trial will be forthcoming shortly.)

Although we accepted almost all applicants, I think that having an application was helpful in screening for fellows who were at least somewhat interested in EA already. It also made the fellowship feel more legitimate and official to those we accepted. (And it also helped us screen out a few people who were obviously participating only to promote their favorite organization.)

The fellowship sparked some good discussions and provided a regular time and context for fellows to think about effective altruism, which seemed to help people stay engaged. And the end-of-term fundraiser and research project were definitely fun and useful.

On the other hand, both semesters saw substantial attrition in the number of fellows attending weekly meetings, despite our best efforts. I think some attrition is inevitable since people get busier over the semester, but we certainly still have work to do in that regard.

High-profile speakers

We also brought in a few very high-profile speakers. The most famous were philosopher Peter Singer and Skype cofounder Jaan Tallinn. While these speakers drew large audiences (Prof. Singer overflowed a 350-seat room), my sense was that most of the audience wasn’t very interested in the rest of HEA’s activities, so we may not have gained very many followers from it.

That’s not to say that I think the talks weren’t worth it. Having high-profile speakers lent credibility to our philanthropy fellowship, which was a good outreach tool. It also gave us more cachet as a group and provided a good task for the core group members to rally around accomplishing.

Giving Games

At many of our passive-outreach events, we ran giving games in which we presented interested passers-by with $1 to donate to a choice of three charities. By that, I mean that we literally walked up, held out a dollar bill to them, and asked if they’d like to donate it to a charity. This approach was incredibly effective at getting people to participate in our giving game and sign up for our email list—we’ll have some actual statistics up when we publish HEA’s research for the semester, mentioned above.

I’m pretty pleased with the response at all of our giving games. In situations where we were competing for attention they provided a hook that got a lot of people to engage at least briefly. Obviously, not very many people from the giving games got involved further, but that was OK, since we didn’t spend very much effort per person we reached with them, and we also got some interesting data on which charities won and how presentation affected it.

In addition to the setup described above, we tried a different format during an event for visiting high school seniors, in which the participants discussed the three options and then voted, with the winner receiving $20. This also turned out remarkably well, inspiring a lively and thoughtful discussion.

We need more ways to keep people engaged

After the fall philanthropy fellowship ended, we essentially lost touch with many of the fellows. I think a lot more of them would have stayed involved if we’d had more ways for them to do so, and if we’d worked harder to include them in the semester’s events. (What we actually did was basically limited to advertising our talks and other events to them and letting them come to the spring fellowship dinners if they wanted.)

But it’s hard to find things for college students to do that satisfy all of the following desirable criteria:

  1. can be done in the context of an extracurricular activity with a reasonable time commitment per week
  2. are plausibly an effective use of time for college students
  3. look like a reasonable use of time for college students to a casual observer

We brainstormed a lot at the end of the semester and came up with some candidate activities, so hopefully we’ll have solved this problem next semester. But it would have been helpful to have thought of them earlier, and we’re still looking for more exciting ones.

Closing thoughts

What’s the common thread among our best-performing outreach methods? I’m guessing it’s that they select for latent interest in effective altruism. The best advertising we did was passive: we put blurbs in the right places and hoped they piqued peoples interest. Although this probably resulted in a smaller response by the numbers, the people we got were more enthused about EA.

By contrast, our temptation when doing active outreach is to make EA sound as attractive as possible—for instance by bragging about high-profile endorsements, how many smart people are involved, or how interesting/fun/fulfilling it is. But actually, I’m not sure how useful it is to try to arouse interest in people when it’s not already there.


email me replies

format comments in markdown.

Your comment has been submitted! It should appear here within 30 minutes.