What do you see as the biggest risks and failure modes of EA Outreach?
Some of the most salient failure modes for EA Outreach are in the individual sub-projects:
For Will and Peter’s books, the outside-view median outcome is that they don’t make a splash in the media and don’t sell very well. Unfortunately there are just so many books published each year (~1m per year) that the outside-view chances of ours being one of few that gain considerable attention and sell well is slim. Even when you account for the fact that we have a substantial advance and top-tier publisher, the outside view says that we’ll only sell a moderate number of books and will be unlikely to make more than the advance. Inside view says that we’ll do better than this because of the amount of resources going into the book, the fact that there is a movement behind the book, and because people seem pretty interested in EA-style questions at the moment. The reason we are doing this is not for the median case though, it’s for the upper tail in which we become a best-seller and EA becomes well known enough that media hosts feel the need to include it in their discussions of charity, philanthropy, and doing good. The outside-view chances of this happening are slim, and the inside-view chances are better but not huge, though we have been told by publishers, publicists, etc. that it is a real possibility. I will be working very hard over the coming year to give these books the best possible launches I can, but unfortunately the risk is still probably the biggest one that we are taking.
My most salient worry for EA Global is that it doesn’t sell tickets, people don’t come, and it makes a major loss. We are going to be marketing it hard, and part of the reason for moving to a global model is that it makes it easier for more people to come as they won’t have to travel as far.
My biggest worry for EffectiveAltruism.org is that it doesn’t get much traffic. There are now many popular sites that discuss EA, and though EA.org is ranked at no. 5 when I do an incognito search for ‘Effective Altruism’, my main worry is that it won’t get enough traffic. My other worry is that we will never actually finish building it as other higher-priority projects will take precedent, but I don’t see that as as much of a risk.
I suppose my worries can be put into two broad categories, either we fail to get enough attention for EA, or we get the attention and fail to convert it sufficiently into growth of the movement. I think both of these are very real possibilities that we are working every day to reduce the chances of.
For EffectiveAltruism.org, do you know the search volume for “effective altruism” or related queries? Do you have any sense of how much traffic you could possibly get (e.g., what quantitatively would count as “lots of traffic” vs. “not much traffic”)?
Google estimates that there are 720 searches a month for “effective altruism.” I expect that to increase over time, but I don’t expect search to be the biggest source of traffic initially. I’m expecting that link traffic from people who want a site to explain EA will be the biggest source of traffic initially.
I don’t have a number attached to my expectations about traffic numbers. But, since CEA has a number of web properties, we can benchmark effectivelatruism.org pretty easily by comparing our traffic profile to the traffic profile of other CEA sites. We can use this to figure out whether the site is succeeding or failing from a traffic standpoint.
However, I slightly disagree with Niel that low traffic is the biggest risk. I think the biggest risk is that the site does become the de facto landing page for EA, but the site isn’t very good or doesn’t encourage people to want to get involved. You can get lots of traffic and still not have much of an impact.
I’m working to hedge against this by working with Andy Fallshaw who is an extremely talented designer and branding expert and by A/B testing a wide variety of aspects of the site.
For EA Global, I have a couple additional worries that I’d love your thoughts on:
It seems like organizing the events itself will be another large difficulty. I understand that previous EA summits required a fairly large push from Leverage; I’m guessing Tyler Alterman will be able to command at least some of the same resources, but potentially not all of them (at least, that’s what I infer from him explicitly being in charge and not Geoff Anders). Meanwhile, if costs are lower, the events could be even larger and more difficult to organize, especially for those without such extensive experience as Leverage.
If there are a number of different summits going on, some of which are less-officially organized but under the same “EA brand,” it seems like this increases the potential for damaging that brand if something weird happens. I think this was a decent risk factor even for the 2014 summit as it was run, for instance.
Do you have any plans to deal with these? Sorry to ask so many questions, but I’m hoping they’ll be useful to other people as well!
If there are a number of different summits going on, some of which are less-officially organized but under the same “EA brand,” it seems like this increases the potential for damaging that brand if something weird happens. I think this was a decent risk factor even for the 2014 summit as it was run, for instance.
I think the probability that some negative event happens increases as we add more different locations for events, but I think the total risk to the brand decreases. If something weird happened in 2014, it would mean that something went wrong at the only EA gathering of the year. If something weird happens at a single location this year, it’s more plausible to explain that it was an isolated incident than that it is representative of all of EA.
That said, I’m vetting people running the events carefully and focusing mostly on finding experienced event runners. I plan to exercise more extensive oversight for areas of managing the event that could damage the EA brand.
It seems like organizing the events itself will be another large difficulty. I understand that previous EA summits required a fairly large push from Leverage; I’m guessing Tyler Alterman will be able to command at least some of the same resources, but potentially not all of them (at least, that’s what I infer from him explicitly being in charge and not Geoff Anders). Meanwhile, if costs are lower, the events could be even larger and more difficult to organize, especially for those without such extensive experience as Leverage.
There’s no doubt that EA Global is an ambitious project and I agree that the size and scope of the plan creates an opportunity for failure. But, I have a few tricks up my sleeve that make me think this is manageable: 1) we’re planning to hire an event planner to help with coordination and logistics, freeing up more of my time to focus on strategy and marketing; 2) I’ve been consulting with Nevin and Cathleen regularly and am benefiting from their experience; 3) We’ve scaled down some aspects of the events themselves which will reduce the number of components that might go wrong.
Some of the most salient failure modes for EA Outreach are in the individual sub-projects:
For Will and Peter’s books, the outside-view median outcome is that they don’t make a splash in the media and don’t sell very well. Unfortunately there are just so many books published each year (~1m per year) that the outside-view chances of ours being one of few that gain considerable attention and sell well is slim. Even when you account for the fact that we have a substantial advance and top-tier publisher, the outside view says that we’ll only sell a moderate number of books and will be unlikely to make more than the advance. Inside view says that we’ll do better than this because of the amount of resources going into the book, the fact that there is a movement behind the book, and because people seem pretty interested in EA-style questions at the moment. The reason we are doing this is not for the median case though, it’s for the upper tail in which we become a best-seller and EA becomes well known enough that media hosts feel the need to include it in their discussions of charity, philanthropy, and doing good. The outside-view chances of this happening are slim, and the inside-view chances are better but not huge, though we have been told by publishers, publicists, etc. that it is a real possibility. I will be working very hard over the coming year to give these books the best possible launches I can, but unfortunately the risk is still probably the biggest one that we are taking.
My most salient worry for EA Global is that it doesn’t sell tickets, people don’t come, and it makes a major loss. We are going to be marketing it hard, and part of the reason for moving to a global model is that it makes it easier for more people to come as they won’t have to travel as far.
My biggest worry for EffectiveAltruism.org is that it doesn’t get much traffic. There are now many popular sites that discuss EA, and though EA.org is ranked at no. 5 when I do an incognito search for ‘Effective Altruism’, my main worry is that it won’t get enough traffic. My other worry is that we will never actually finish building it as other higher-priority projects will take precedent, but I don’t see that as as much of a risk.
I suppose my worries can be put into two broad categories, either we fail to get enough attention for EA, or we get the attention and fail to convert it sufficiently into growth of the movement. I think both of these are very real possibilities that we are working every day to reduce the chances of.
For EffectiveAltruism.org, do you know the search volume for “effective altruism” or related queries? Do you have any sense of how much traffic you could possibly get (e.g., what quantitatively would count as “lots of traffic” vs. “not much traffic”)?
Google estimates that there are 720 searches a month for “effective altruism.” I expect that to increase over time, but I don’t expect search to be the biggest source of traffic initially. I’m expecting that link traffic from people who want a site to explain EA will be the biggest source of traffic initially.
I don’t have a number attached to my expectations about traffic numbers. But, since CEA has a number of web properties, we can benchmark effectivelatruism.org pretty easily by comparing our traffic profile to the traffic profile of other CEA sites. We can use this to figure out whether the site is succeeding or failing from a traffic standpoint.
However, I slightly disagree with Niel that low traffic is the biggest risk. I think the biggest risk is that the site does become the de facto landing page for EA, but the site isn’t very good or doesn’t encourage people to want to get involved. You can get lots of traffic and still not have much of an impact.
I’m working to hedge against this by working with Andy Fallshaw who is an extremely talented designer and branding expert and by A/B testing a wide variety of aspects of the site.
For EA Global, I have a couple additional worries that I’d love your thoughts on:
It seems like organizing the events itself will be another large difficulty. I understand that previous EA summits required a fairly large push from Leverage; I’m guessing Tyler Alterman will be able to command at least some of the same resources, but potentially not all of them (at least, that’s what I infer from him explicitly being in charge and not Geoff Anders). Meanwhile, if costs are lower, the events could be even larger and more difficult to organize, especially for those without such extensive experience as Leverage.
If there are a number of different summits going on, some of which are less-officially organized but under the same “EA brand,” it seems like this increases the potential for damaging that brand if something weird happens. I think this was a decent risk factor even for the 2014 summit as it was run, for instance.
Do you have any plans to deal with these? Sorry to ask so many questions, but I’m hoping they’ll be useful to other people as well!
I think the probability that some negative event happens increases as we add more different locations for events, but I think the total risk to the brand decreases. If something weird happened in 2014, it would mean that something went wrong at the only EA gathering of the year. If something weird happens at a single location this year, it’s more plausible to explain that it was an isolated incident than that it is representative of all of EA.
That said, I’m vetting people running the events carefully and focusing mostly on finding experienced event runners. I plan to exercise more extensive oversight for areas of managing the event that could damage the EA brand.
There’s no doubt that EA Global is an ambitious project and I agree that the size and scope of the plan creates an opportunity for failure. But, I have a few tricks up my sleeve that make me think this is manageable: 1) we’re planning to hire an event planner to help with coordination and logistics, freeing up more of my time to focus on strategy and marketing; 2) I’ve been consulting with Nevin and Cathleen regularly and am benefiting from their experience; 3) We’ve scaled down some aspects of the events themselves which will reduce the number of components that might go wrong.