CEA strategic update, March 2017 (what we learned at YC and more)

CEA have now officially completed Y Combinator (YC) and so in this month’s update I wanted to share with you what we learned and achieved as a result in Q1 and the sorts of things CEA is planning to prioritise and why in Q2.

The TL;DR version is that in the last quarter we used YC’s expertise to focus on moving money through EA Funds to support the most effective organisations. In Q2, we will focus on building infrastructure for the EA community itself so that we can improve the ways we nurture talent and generate ideas. Overall, our main aim is to increase the total, long-run value of the EA movement. This is all in pursuit of our vision and mission to build a community focused on figuring out what an optimal world looks like, and how we get there.

An overview of our plans for the next 6 months

As I discussed last month we feel the community particularly needs:

For this reason, we’re experimenting with dividing the first three-quarters of this year by the resource we’re most heavily focusing on. We see this shift in focus each quarter as a change in tactics, rather than a change in strategy. Each quarter, our goals and metrics stay the same, we just choose different methods for improving the long-run value of the EA movement each quarter, much in the same way a for-profit company could choose to grow revenue by either gaining new customers or increasing the average spend of existing customers. This quarterly focus doesn’t mean other activities are discontinued (or even paused) during that quarter, instead we plan to maintain the base level while we focus our marginal activities on growing and trying new things in one area.

Q1: We focussed on money through EA funds. More on this below. Q2: We’re shifting our focus to talent and community (both in terms of capacity building within our team and in terms of supporting the community through new online infrastructure, events and local groups). More on this below. Q3: We currently aim to use the community infrastructure developments from Q2 to try and better incentivise and facilitate the generation of new ideas within EA. This plan might change depending on what we learn in Q2.

One of our takeaways from YC was to keep our overall mission in sight but be flexible in our approach as new opportunities come up and as we learn more about what works. As a result, we have more concrete plans and goals for the next three months and rougher plans for the next nine months. We don’t expect the basic grab-bag of activities we focus on to change all that much over 2017, but we do expect the specific projects and initiatives we focus on to change as we gain more information.

Review of Q1 goals

1. Test Effective Altruism Funds as a concept

We focused on moving money through EA Funds in Q1 because YC has a strong comparative advantage in helping organisations increase revenue. Since launching on 28th February, we’ve moved over $572,690.69 (at the time of writing) through the funds with an additional $302,536 to come in bank transfers. You can review the figures here. This puts us just short of the $1 million we wanted to show success but as the community feedback has been very positive and we have promises of very large, recurring donations this year we think it is worth continuing EA Funds. We expect the first grants to be made by Fund Managers in the next month.

As a result of the feedback and size of donations through the EA Funds, we are now recommending that Giving What We Can Members donate through EA Funds and closing down the Giving What We Can Trust. We had been investigating changing the charitable objects of the trust, to allow granting to non-poverty charities, to better reflect the cause neutral nature of the pledge, but with the development of EA Funds, it makes sense to combine the two activities. UK donors will be able to make tax-deductible donations to support GiveWell top charities, or any of the funds through the EA Funds website, thereby providing similar functionality to the Giving What We Can Trust. This reduces the administrative costs for us while still making it easy for anyone to donate to top charities. We see this not only as a concrete way of improving donation coordination and robustly doing good in the world, but also a way to get much better data on where the community is donating and an incentive for researchers to conduct research for fund managers or even consider becoming fund managers themselves.

2. Establish models of how to evaluate the impact of our activities

This goal was about helping us decide how best to spend our time and money in pursuit of our mission. We developed some internal metrics to help us compare all the different activities we pursue, from promoting GWWC to running EA Global and have been using these initial metrics to shape our overall strategy around events, student groups and online activities. We’ve made some progress on how we measure the impact of our work on events and local groups but we have fallen behind on this. It is likely that we were overly ambitious on what we could achieve during the intense, last few weeks of YC. This goal continues into Q2, particularly as we are now focusing more on talent and community, we hope to test the initial metrics we developed in the lead-up to EA Global Boston and develop the model further. We hope to share more of our thoughts on this in later updates.

3. Maintain and grow a positive relationship with the broader effective altruism community

Our relationship with the community is vitally important to the work we do. Figuring out how to create an optimal world is incredibly hard and we need as many talented people working on it as possible. Historically, we have not done a great job at communicating the why of what we do, particularly as CEA has evolved over the years. We spent Q1 focusing on improving our internal communication norms and codifying our strategic updates and core mission in order to make it easier for any member of staff to accurately communicate our goals and reasoning to other people in the community. In Q2, we plan to focus on improving our communication with the broader community in a similar fashion.

More concretely, in order to strengthen our relationship with the community we have created an advisory panel for CEA to get an outside view on our strategy, published our guiding principles, set up this blog and tried to explain more of the why behind our strategy in these updates.

We still have a lot to do but we think this has been a positive start. We’ll be continuing to work on this further in Q2 (more on this below).

What we learnt at YC and how CEA’s thinking has changed

CEA has gained a lot from taking part in YC not only in terms of expanding our network and financial support but through developing our ideas. Here are just a few of the ways we’ve updated our thinking. Not all of this has come solely from YC, or advisers who we were put in touch with through YC, but having three months together to really focus has certainly helped.

1. Focus on our mission

The process of applying to YC alone has helped us to clarify our vision and mission. Since August 2016, we’ve come a long way in prioritising what we work on and YC seems to have validated this approach and helped us create a more unified CEA. This includes ceasing GWWC research and combining the outreach efforts of GWWC, EAO and the EAG team. Restructuring CEA in this fashion involved changing many people’s roles and areas of focus and letting some people go. This was a difficult decision, but three months later we are excited about having a more unified team with a clearer focus.

2. A broader range of more specific metrics

The YC partners did find it harder to suggest metrics for us than for their for-profits or even other non-profits. We still don’t have one unifying metric but we did learn a lot about what makes a good metric. In particular, during YC, we switched from focusing on the number of new pledges to focusing on the counterfactual value created by adding each cohort of pledgers. We began tracking donation history of new members more closely, in an attempt to better understand how new members choose where to donate, when to donate and how to assess cause areas. We realised that many new members often had deeper expertise in one cause area, where they might feel comfortable making specific donations, but many wanted to donate in a wider range of cause areas. They often did not feel qualified to assess which organisation to support in other areas with the same level of rigour. This was another reason we wanted to set up EA Funds. Longer term, we hope that further iterations of EA Funds can be used as the primary donation platform for most members, which will give us much better information on where members donate, how much they donate and how their cause selection changes over time. The version of EA Funds that exists now cannot fulfil this purpose for many members, but we hope that by adding the ability to donate to specific organisations and adding new, more focused funds we will be able to serve as the main donation platform for new members and any EAs who donate.

We do this with a deep understanding that measuring money moved alone captures only a small portion of the total value of the EA community. We’ve been attempting to use a metric similar to 80,000 Hours significant plan changes in order to ‘convert’ between money moved, plan changes and other significant value added by the EA community, all in an attempt to measure the overall value produced by the community. We hope to share more of our thinking on these more complex, compound metrics in the future.

3. Getting “product-market fit” and strengthening the community before we grow

The YC partners offered surprisingly useful advice on community building. It turns out that many for-profit companies have grown significantly by building a community based around their product or service and many YC companies, like Reddit, are explicitly based around a community. We were pleasantly surprised by how scientifically the partners approached community building, and the extent and depth of knowledge contained both in the YC community and in experts who they could connect us with. We realised we need to do more to build the infrastructure and networks within the EA community so that it will continue to provide value for existing members, despite the addition of many new people to the community. We need to ensure that the core community gets a lot of value and that we know exactly where that value lies before we put a lot of effort into growing the number of participants in the community. This led us to pivot from focusing on promoting EA Funds to speaking to more users and improving the website and funds first.

This doesn’t mean we won’t do any promotion to new people but it means that for at least the next quarter our focus will be on improving some of the community infrastructure. This is likely to include things like improving introductory emails to new newsletter subscribers, writing up major ideas and concepts core to EA on effectivealtruism.org and getting more people involved in local groups and EAGx events.

It seems likely that we’ll also want to create spaces both for the most engaged members of the community to generate new and complex ideas and places that welcome and onboard new people so that everyone’s needs can be better met.

4. The importance of communication

Related to the above, in order to be able to provide things that the community wants we also need to improve our communication so that we better understand those needs. Realising this (and the ways in which our communication has been falling short) has influenced our Q1 and Q2 goals to communicate more about what we’re doing and why (more on our Q2 goals below).

5. Scale

We’ve started to think about what areas of our work currently don’t scale and what we can do instead. In the early days of the effective altruism community, we were small enough to know many people personally and these one to one interactions were crucial for a lot of people in making big changes to their lives. As the community grows it is impossible for us to have one to one interactions with everyone and so we need to find ways to empower others to build those relationships and share knowledge through things like local groups and events.

Will has recently published a post about why non-profits should apply to Y Combinator which I encourage those of you at other EA organisations to read and consider.

Priorities for Q2

As mentioned in the section above, we’re now turning our attention to talent as a resource and supporting the foundations of the community so that we have more great people working on some of the biggest problems. This is a huge area and hardly something we expect to solve in the next few months. We’re simply planning to set up some of the basic infrastructure and capacity for the future.

We’re currently in the process of defining our exact Q2 objectives which I will be able to share with you next month but broadly they will be around consolidation and community.

Consolidation means building infrastructure internally through hiring additional staff and streamlining operational processes. We’ll be supporting the community through events like EA Global Boston and the San Francisco Bay Area: Community Summit EA Global event, finding ways to keep track of and nurture talent from local groups, building better infrastructure for communication online and publishing core concepts of EA on effectivealtruism.org.

I hope this gives you a flavour of what we’re working on and why and how our thinking has changed over the course of YC. Of course, it’s more than I can cover in one post so if you want to discuss any of this further please get in touch.