Effective Altruism Outreach is a sub-project within the Centre for Effective Altruism, led by Niel Bowerman, Kerry Vaughan, Tyler Alterman and William MacAskill. Its primary purpose is to grow and strengthen the effective altruism community. For more information, see here.
A core part of EA Outreach’s strategy for 2015 involves gaining attention to EA ideas from mainstream media. It’s highly unclear to us how valuable this is: it’s perfectly plausible to us that this will involve generating a lot of noise, but that this won’t ultimately generate much value; it’s also plausible to us that it will lead to significant connections that wouldn’t have otherwise happened. It may also be that some sorts of media attention are much more valuable than others. For these reasons, and because we feel we don’t currently have a good understanding of these issues, we’re conceiving of this aspect of EA Outreach as an experiment. We’ll therefore periodically take the opportunity to reflect on aspects of our media initiatives, so that we can better assess whether we should continue with this aspect of our strategy, and if so in what way we should approach it. It seems almost impossible to get high-quality evidence on the issues that we care about in this area, so in our ‘experimentation’ strategy we are forced to rely on a small number of data points and a lot of reflection.
In 2014, getting media attention for EA was not one of our aims. However, the popularity of my ALS ice bucket challenge article provided us with an opportunity to do some media outreach and start some discussion of effective altruism in the media. In this post, I describe what happened, what short-term and long-term impact I think it had, and what lessons I took away from it.
Background
On August 14, 2014, I wrote an article on the Ice Bucket Challenge for Quartz, which the magazine entitled “The Cold Hard Truth about the Ice Bucket Challenge.” It became extremely popular. We can’t disclose the viewing figures, but it received about nine times as many views as my second most popular article, and 30 times as many views as my median article. The article received 141 Google+ shares, 467 LinkedIn shares, 1490 tweets, 16,008 Facebook shares and 41,967 Facebook likes. It also led to a significant amount of attention in other venues, with coverage in The Guardian, The Telegraph, Los Angeles Times, The New Yorker, BBC News, CBS News, Vox, Slate, and dozens of other news sources, and appearances from Niel Bowerman and I in international radio and television.
The article had a mixed reception. Of those we sampled, most twitter shares were positive; most facebook comments were critical.
On August 18, I wrote a follow-up article, which Quartz entitled “This week, let’s dump a few ice buckets to wipe out malaria too” (again, I wasn’t responsible for the title). This one received the typical number of views.
Positive impact
The short-run impact of the article was as follows:
Referrals to GiveWell: ~10,000 unique visitors. Elie told me that, on average, 1 in 200 unique visitors donate, that the average donation is ~$1000, and that he estimates that 50% of people are repeat donors. However, we both agree that these numbers would give too high an estimate of the value of a unique visitor, because much of GiveWell’s money moved comes from very large donors, who have heard about GiveWell from multiple sources.
From their 2013 metrics data, $3mn was moved via small donors (less than $10k) in 2013. On average, 14% of donors came via online referral; assuming that this is representative for small donors that would give $420,000 via online referral. There were 151,000 visitors via referral links, giving a money moved per visitor figure of $2.78.
Using this number, the Quartz article would have moved $27,800 (not including regular giving beyond a year). This was quickly estimated, and there may well be errors in this assessment, and still seems somewhat too high to me. We could therefore give cautious bounds of between $5,000 and $50,000 moved as a result of the article.*
Referrals to Giving What We Can: ~9,000 unique visitors. Soon after the article, we received 3 pledges and 8 try out givers who cited ‘on-line article’. We’ll follow up with these people, but it’s likely that not all were influenced by the Quartz article, rather than some other online article. If we assume 3 pledges only, the article raised approximately $1 million in pledged donations. Converting this into the value of present donations is a tricky issue, but my guess is that the long-run money moved via new pledges is somewhat greater than that via increased traffic to GiveWell’s site.
Donations to SCI: As a result of the initial article I wrote, Tim Harford invited Niel Bowerman from CEA and Elie Hassenfeld from GiveWell onto BBC Radio 4’s ‘More or Less’ show on ‘the numbers behind the news’ to discuss effective altruism. This resulted in follow-up pieces on the popular More or Less’s podcast, in BBC magazine, and in the Financial Times. In the interview with Tim Harford, Elie and Niel discussed SCI, and Tim Hartford decided to donate there at the end of the show. After the appearances, SCI contacted us to report that they had received several £1000s of donations as a result of our media. The exact amount SCI received as a result of this media attention was difficult for them to estimate relative to the variable background rate, but they suggested it may have been as much as £10,000.
Given these figures, and given the fact that an article as successful as the ice bucket article is perhaps a one in twenty event, my impression is that the short-run benefits of writing articles are not themselves sufficient to justify the time investment, compared to other sorts of outreach.
However, my guess is that the main benefit of media attention is longer-term: leading to further media opportunities in the future; improving my position as a public advocate of effective altruism; increasing the real-world credibility of the effective altruism movement; generating connections with people who we wouldn’t otherwise have met; and, most importantly, learning via practical experience of engaging in the media.
This view is consistent with our conclusions from previous media rounds. Relative to the size of the media attention Giving What We Can and 80,000 Hours has had in previous years, the short-run benefit has been lower than one would expect. However, it has led to some of our most valuable relationships, with people who were initially in very different circles, who we would not otherwise have met.
Why was the article so popular?
My leading explanations (in order of plausibility; not mutually exclusive) are:
- Snowball effects
- The article piggy-backed on an extremely popular trend.
- The title of the article was particularly good at arousing interest.
- The article defended a minority position.
- The article discussed an issue that’s rarely covered in the media.
Mistakes made
In my view, the article is of lower quality than other Quartz articles I’ve written. It was written quickly; it half-makes several distinct points rather than having one clear focused point; given that people many find the central points surprising, it made those points too hastily; and at times the tone was unnecessarily snarky.
I do not know if I should have spent more time on the article to improve its quality. If I’d known how popular it would be, then I certainly should have done; but I didn’t know that (nor did I know how popular Quartz articles could get). Moreover, I have a distinct worry that what I call poor quality was a positive factor, in terms of the article’s popularity. People are more easily annoyed—and more likely to voice their annoyance on social media—at articles with poorly justified claims; readers will more easily read whatever they want into vague or ambiguous statements; and an article with several distinct claims has a higher chance of piquing the reader’s curiosity. This explanation, if true, would be deeply depressing, but would also explain a lot about the poor quality of many popular articles. This could be investigated further by, for example, looking at popular writers, and seeing how the quality of their most popular writing compares to the quality of their less popular writing.
The potential costs from the article were:
(i) angering people, and turning them off ideas relating to effective altruism;
(ii) damaging my brand, or the brand of effective altruism.
My impression is that, for a one-off article, these aren’t particularly great costs. Consideration (ii) is the larger concern. There is a real long-run risk of being perceived as a charity skeptic (similar to e.g. Bill Easterly), which suggests I want to strictly limit how many ‘critical of charity’ pieces I write. I’m even more concerned with associating the brand with poor quality of argument than with the sort of contrarianism that this article might have been perceived as embodying. This provides a reason in favour of writing fewer but more in-depth articles.
A separate mistake, in my view, is that we should have capitalized more on the success of the article. For example, I could have written a series of articles on the topic, using the opportunity to discuss charity cost-effectiveness in more depth. This would have also mitigated some of the brand-related worries. As noted, my follow-up article was only as popular as my typical article, which I took to be evidence against spending more time on it. However, other articles on the ice bucket challenge also did very well on Quartz, and my follow-up article had a poor title, so I think I put too much weight on that fact.
Overall Assessment
It’s very difficult to assess how valuable the article was, and whether it was positive or negative in value.
My guess is that the benefits outweigh the costs in this instance. However, I don’t think that I should keep writing similar articles: my hunch is that the marginal costs, in terms of brand damage, of articles like this are either non-diminishing or increasing, whereas the marginal benefits (learning and profile), are decreasing. I do plan on writing more articles for Quartz.com, Vox.com, and other media outlets, however I plan on making them comparatively more rigorous than the article in question.
Lessons learned
The short-term benefits of media attention are comparatively small but non-negligible.
Representing the minority view in an ongoing debate seems to be a promising strategy of gaining media attention. In an attempt to insure “balance”, the media give roughly equal weight to all viewpoints, regardless of popularity. As a consequence, articles defending minoritarian positions will tend to receive disproportionate media coverage.
Brand-management is going to be very difficult, because it’s so difficult to predict what will go viral, difficult to predict how people will react, and very difficult to judge what the appropriate level of controversiality is.
Future experiments
In the coming months I will experiment by with writing longer and comparatively more rigorous articles on carefully chosen topics.
This will greatly increase the time cost of writing articles. How to balance increased quantity with increased quality is difficult to know. Moreover, though developing a personal brand as “the really reasonable and well-informed guy” would be optimal, it’s highly unclear to me whether this is achievable. There are examples of people who have been successful in this regard (like Nate Silver and Tim Harford), but it’s difficult to generalize from cherry-picked examples. However, it’s worth trying it out.
These articles will be stored up and released around the time of my book launch to help raise the profile of the book and to help boost book sales.
Plans for 2015
EA Outreach has hired Goldberg McDuffie Communications to raise the profile of Peter Singer’s book launch in April 2015. We have already distributed advanced copies of Peter’s book to over 100 journalists.
In May 2015 Peter Singer will be in the UK for the UK launch of his book and Yale University Press will be working with us to promote Peter’s book here. Text Publishing will manage Peter’s Australian book launch and associated media push.
We’re still deciding whether to hire a publicist or marketing firm to help with the launch of my book in August 2015; it’s unclear what the value added will be over my publishers’ own efforts. Penguin will be promoting my book in the US, and we have Guardian Faber and CEA publicizing my book in the UK. We will be negotiating rights with other countries in the next few months.
* Given that the money moved per visitor figure may be as high as $2.78, and may be even higher, it seems worthwhile to experiment with paying for ads to GiveWell’s site, to see if one can achieve a better than 1:1 return on expenditure.
Reflections like this unravel truisms like "all press is good press", and "no news is good news", to "the quality of news depends on one's goals, and has many correlations which must be separated into noisy consequences and signals of genuine value". I'm a bit nonplussed by how "a small number of data points" in this case equaled only one data point, but I'm impressed with how much you produced via reflection. I enjoyed this post, and I'm pleased with the level of learning and humility Dr. MacAskill and co. show. Thanks for writing this.
I've got some concerns about how "brand management" might be a shiny veneer to cover "centralization" from Oxford or the Bay Area, and what the consequences of it may be. These concerns aren't about projects in Oxford, per se, but about some aspects of Effective Altruism Outreach. I'll save most of them for a later post on the subject broadly. My question now is: how much will EA Outreach focus upon centralizing and narrowing who shares information, and to what extent will you make attempts at tamping down other voices in the movement in terms of "damage control".
I'm worried EA Outreach will create an appearance of a lack of diversity and peer-review within the movement, the false conception newcomers aren't explicitly welcomed to intellectually engage existing ideas, or the Centre for Effective Altruism will become a bottleneck which cannot handle all the attention it may soon receive.
On the other hand, I think these concerns might be overblown. Also, I'm aware other social movements have been more centralized and not necessarily spontaneously "grass-roots". In leading the civil rights movement, Martin Luther King, Jr. and his peers had strategies for non-violent resistance and direct action. Representations of history in popular media seem to gloss over this. So, I understand the need for a community to reach an accord on "brand management", and public perception, with its "thought leaders", for lack of a better term.
I think it would be best if Effective Altruism Outreach prepared to begin a dialogue with others in the movement, perhaps on a forum, perhaps with bloggers or writers or vocal members of the movement they already expect to share their own perspectives on publications from within effective altruism. I think it should take place in a way in which it's made clear attempts to restrict or censor will not be made, but EAO sincerely wants to work as peers with others to think about positive or negative consequences of how communication happens.
Do you have any perspective on what or how the rest of us can do to help and not hinder the growth of effective altruism, in a way that both allows us to remain independent voices, without watering down the core strength and mission of effective altruism?
That would explain how concerned people mostly come from these places, and why they have unusually high concern for a social movement.
Evan:
Impala:
Thanks for raising this concern. I think that a post on the topic would be really valuable - we certainly don't want to lose the benefits of diversity, to shut down debate, or to become a bottleneck. I don't think we have unusually high concern regarding branding within a relevant reference class - almost all businesses are highly concerned by brand; free market promoters are highly brand-concerned. Even PETA is highly concerned by brand management in its own way - e.g. using models and celebrities to push v*ganism as 'sexy'. For movements with unfortunate brands (I'd put animal rights, feminism, and to some extent environmentalism), that seems like a big problem and one we want to avoid. It's also a co-ordination problem (see 'the unilateralist's curse'), and if you've got a diverse range of groups within the movement who differ pretty fundamentally then co-ordination is going to be difficult - and I think that may explain why they suffer these branding issues. This is something we really should try to avoid. (Also, you mention the Bay as well - my impression, which might be wrong (based on conversations with Geoff Anders), is that they agree that it's a bad problem, but think that there's not much we can do about it and so don't actually spend much time on it. I think it's worth trying, but I agree it's a judgment call).
In terms of concern for this - this is something I've changed my view on a lot over the last 5 years. Because there's so much bullshit-speak surrounding brand management, I was initially very skeptical of its value. But experience with the different orgs has made me switch my view a lot. The most important example of this was back when 80k was 'High Impact Careers' and was quite aggressive in its marketing - focusing on earning to give and banking/doctors comparisons. This was a disaster. People thought we were being deliberately contrarian (and they were right), and we're even still trying to overcome the impression that 80k is only about promoting earning-to-give. Also people lost the message - e.g. some people thought we were promoting finance in its own right; many people didn't realise that charity effectiveness is a core part of the argument.
Interesting - what metrics are you thinking of when you say it was a disaster?
Thanks for the response. I didn't notice this until just now. I believe it's harder to screw up with brand management than without, so we should try. I think part of the problem could be solved with the EA Advocates project, or something like it, which I wasn't aware of when I made my original comment. Having advocates which are geographically distributed, and don't all come from the same culture, e.g., academia, or high-tech industry, may send the signal effective altruism is about whichever individuals are passionate about it, and isn't just a fancy label for those from one group. I think indicating how effective altruism is a distributed network could help solve coordination problems, because various advocates can act as sounding boards for newcomers, whether on the basis of location or cause of interest.
One reason I distinguished the Bay Area from Oxford is because what reference class for effective altruism promotes is somewhat different. Even more than just "non-profit" in general, the Centre for Effective Altruism is located among what people typically think of "philanthropy". It's orientation and association with giving pledges, academia and policy recommendation, all seem to make the leap from default to explicitly effective altruism more intuitive. The CEA, with e.g., The Life You Can Save and Giving What We Can, seems able to create a coherent culture.
The Bay Area is more all over the place. First of all, it doesn't have "center". In the United States, effective altruism is primarily associated with Givewell, but its culture is primarily programmed by Leverage Research, and its grass-roots advocates hail from a wider diversity of causes across the Internet, who don't communicate as much with each other. On one hand, for-profit high-tech social entrepreneurship, or whatever, is seen by some effective altruists in the Bay Area as the way the most good can be achieved. This lean startup culture bleeds into the promise of emerging technologies, and the high-minded vision of e.g., Geoff Anders, or Eliezer Yudkowsky. Then, I can turn around, and the next effective altruist I talk to is in favor of not only a radical cause, but pushes it with radical action, such as Direct Action Everywhere.
If I talk to a student from Harvard, Yale, or Oxbridge, I can trace their effective altruism back to your work, or Peter Singer's. There isn't an intuitive prototype for effective altruism in California. It's more chaotic. I don't think this is a bad thing in terms of action. I think it would make brand management more difficult. However, I think this eclectic culture could be swamped from the influx of whoever's exposed to effective altruism by your and Singer's new books. Overall, I now think brand management will work better if effective altruism is seen as a coherent set of ideas distinct from its related cause areas which are not intuitively related to altruism, like high-tech business, or which are more volatile and controversial, such as radical veganism.
Thank you for the thoughtful response. I think a big issue here is whether businesses or social movements are the right reference class. I think it's the second, and social movement don't usually try to do the sort of brand management of what other activists or groups do that I see in effective altruism.