(I'm posting this on behalf of the EAGxBoston 2016 core organizing team, which consisted of Taymon Beal, Dillon Bowen, Randy Carlton, Mi Shan Wong, and myself.)
We organized EAGxBoston, one of the first EAGx conferences. (Along with EAGxBerkeley, our event took place on April 30, 2016.) This is a post about some of our organizing experiences and what we learned from them, which we hope other EAGx organizers (and people organizing similar events more generally) will find useful.
We held our first organizing meeting on February 7th and almost 200 people came to EAGxBoston on April 30th. This was a very tight timeline and we would not recommend trying something similar with a shorter lead time; a longer lead time would have been better in several ways. In the first few weeks we each spent 2-4 hours per week on conference-related work. During the last month, our organizing meetings alone took two or more hours per week and we spent considerably more time on other tasks. Randy put in full time hours in the couple weeks leading up to the conference.
Choosing a venue and a date early was critical. Our venue was available for only one of our three preferred dates, so we reserved it immediately. Choosing and contacting speakers was the other thing we spent most time on in the early planning. This paid off, in that we got a great lineup of speakers overall, but it would have been good to start even earlier. There were some specific slots in our lineup that we were still trying to fill very late in conference planning, and if we’d started earlier some of the people we contacted about them might have been able to attend our conference.
We had a lot of uncertainty about how much funding we would have available for the conference until less than two weeks before it happened. This meant we spent a lot of time making duplicate plans for potentially expensive things like video recording, audiovisual services, catering, and general supplies. There wasn’t much more we could have done about slow responses to grant applications. But it would have helped us to know what to expect from our ticket sales - although they often leveled off for one or two days at a time, on a larger scale we sold tickets at almost a constant rate for the month preceding the conference. Ultimately, because we were still uncertain about funding when decisions had to be made, we went with the cheaper plan in all cases, including no recording of the conference. We’ll be donating surplus revenue.
The cheap plans involved a lot of work on our part and that of our volunteers, such as picking up food from grocery stores and a variety of restaurants, generally taking care of food logistics that catering staff would have been able to handle, and setting up and running a sound system for amplification in one of the rooms we held panels in. We had a great team and our CEA representative was able to help with some day-of logistics, so for the most part things went smoothly and conference attendees seemed happy with the event.
Although the conference went well, we think others could have better experiences by doing some things differently than we did:
Start planning farther in advance of the conference date! It’s easier to find a venue and confirm speakers with more lead time, and not scrambling on these tasks will give you more time to spend on thing that really need to happen near the time of the conference, like promoting it to possible attendees.
Confirm funding earlier! We think that CEA will be able to confirm grant amounts farther in advance with other events, which should help.
Have a stronger plan for day-of logistics! We mostly hadn’t assigned organizers or volunteers particular areas of responsibility for during the conference until we all arrived at the venue. While everything did get done, it probably would have gone more smoothly if a specific person had been in charge of each major area like food and registration, and had been able to plan for that in advance.
The rest of this post goes into details about various aspects of our event.
In Boston most small conferences happen at universities. Mi Shan is a student at MIT, which has a few suitable venues for conferences the size of ours. She took the lead on all communications with MIT. When we started looking, conference venues at MIT were very full, and the Wong Auditorium was only available for one of the dates we were considering, so Mi Shan reserved it immediately. Over time we changed some parts of our reservation, but we ended up reserving the Wong Auditorium, the lobby space outside it, and four classrooms upstairs in the same building for 8am-9pm on April 30. Because Mi Shan was a student, there was no cost to use the space. Mi Shan reserved the space as a student club leader of MIT Sloan EA, so it is important to have a university liaison who has the capability to reserve conference space.
However, there were other costs associated with the venue. The Wong Auditorium has a capacity of 291 people, and for a large event, it’s necessary to use the built-in sound system. This meant charges for equipment rental, and for a mandatory technician who operated the room’s AV system throughout the event. The technician was professional and helpful, and freed us to take care of other things. For the second room in which we held panels, which was smaller but still large enough that we wanted some backup amplification, we chose not to use MIT’s AV services. Jeff Kaufman is a contra dance musician and owns a lot of related sound equipment, and he set up a system for us using his equipment and some he’d rented cheaply from a friend. We had a volunteer learn the system and run sound during the panels, so Jeff was free during the conference itself. We were able to use the built-in projection system.
We also needed to pay for table rental (for panels, catering, and registration) and for MIT’s custodial staff to clean the venue after we left.
An additional upside to holding the conference at a university where one of us was a student was that there was funding available through the university. Mi Shan sought funding from several sources at MIT, and these accounted for almost all of our success in getting outside donations and sponsorships for the conference.
We chose to hold EAGxBoston on a single afternoon because of our short lead time. The longer the conference, the more logistical issues we would have had to deal with, including additional meals and possibly even lodgings for people who had traveled to attend the event. This meant we had a short time available for the programming we wanted to offer and for networking opportunities.
Originally we planned to have two panel time slots with three panels each, but switched to three time slots with two panels each because we thought many people would want to attend more than two of the panels (especially as were were able to secure some very good speakers). To accommodate this, we pushed the start time of the conference a little earlier and cut down on break time; we’d originally planned for long breaks to facilitate networking, but actually had shorter breaks that primarily allowed attendees to get snacks and move between rooms. During the planning process we also moved from simply having space reserved after the event so people could network to providing dinner food during that networking time. This worked well; a large number of people did stay for the networking and dinner, and some stayed until we had to clean up.
We chose not to have a specific theme, but to cover a broad range of topics relevant to effective altruism. The Boston area is home to a number of organizations and individuals working on global development, including some with connections to the EA movement, to the Future of Life Institute, which works on mitigating global catastrophic risk, and to a strong local EA community, so we knew we would be able to get a variety of excellent local speakers.
While many of our speakers and panelists, including both our keynote speakers, did come from Boston or New England, we also invited speakers from farther afield when we thought they would make a unique contribution to the event. Originally we conceived of these requests as big reaches; we thought that speakers wouldn’t travel for our event unless it happened to coincide with other travel plans or they could put together a series of talks at area universities in the weeks before and after. We were surprised by how many people accepted our invitations. We would encourage other organizers to write to their first-choice panelists and speakers even if they expect them to be too busy to participate. Getting better speakers than we had counted on really made us excited about our event and forced us to do our best at all the rest of our organizing to avoid letting them down.
While we did well at finding great speakers, we could have done better at timing our outreach to them. Early on, we left some meetings with different organizers having different beliefs about who would be contacting certain potential speakers and panelists and when, so some of our panels were filled before anyone had been contacted about others. Over time we got better at this. We also got better at asking for help from each other when we were too busy to send cold emails to potential speakers or were uncomfortable doing so. But our early delays and short timeline did combine with some delays in response from potential speakers to lead to a few problems. Ultimately there was only one panelist slot that we were simply unable to fill - we wanted someone to talk about climate change on the panel about global catastrophic risks, but weren’t able to find anyone in time who was available and interested. Most other panels were much easier to fill, because we weren’t looking for panelists to fill such specific roles.
While we had originally contacted panelists based on which of us knew them best or was available to write emails, closer to the date of the conference we transferred contact to a single point person for each panel, who also connected panelists with each other and with their panel’s moderator. We sent speakers a guide covering conference logistics and asked them to coordinate their presentation order (if applicable; some panels were question-only). Each panel was also moderated by someone with connections to the Boston EA community, and we asked them to prepare questions in advance to start the discussion and in case the audience was quiet or got off-track. It was nice to have this plan, even though our audiences actually had lots of relevant questions!
We had projection systems available, which many panelists and speakers chose to use. We didn’t try to coordinate a single slide deck for each room, but we did encourage people to use the computers we provided in each room and to coordinate so that all slides for each panel were on a single computer or slide deck. This didn’t provide us with perfectly seamless transitions, but it worked reasonably well. One thing we didn’t think about that would have made us look more professional was having a single background slide to use whenever there wasn’t something else on the projector (and as the desktop on the computers we were using, for transitions between decks). Julia Wise, who gave the CEA intro to our conference, had such a slide that we used at times in the main auditorium.
From the beginning, we planned for the conference to be vegetarian with vegan options. This is how most EA events in Boston handle food. (Julia hosts a lot of them!)
We had originally contacted various restaurants and services about catering the conference snacks. As we got closer to the event, we realized that we needed to provide more food than we’d originally planned - we still wanted to provide snacks during the afternoon, but we also needed dinner (so people could stay and network instead of leaving because they were hungry), and lunch for speakers, volunteers, and others who arrived early. Because we weren’t sure how much funding we’d have, we decided that full catering would be too expensive. Instead we sourced food from a variety of places and served it ourselves.
We ordered trays of curry, chili, rice, and salad from a vegetarian caterer, Rhythm ‘n Wraps, which delivered them to the conference location. All the food was vegan, and they provided a discount.
We also picked up burritos from a local chain, Anna’s Taqueria, which had agreed to donate them to us. While we had clearly requested only vegetarian donations, even if this meant a restaurant was willing to donate less or not at all, when our volunteer picked them up, most of them were meat. We only served the vegetarian ones, because we didn’t think we could consistently and clearly explain why the meat ones were there.
We served fruit, vegetables, nuts, hummus, and baked goods from the grocery store. We picked up tea at the grocery store and coffee from a chain coffee shop, but there was a long wait for those because we didn’t put in the coffee order as soon as we should have. We also left the lunch food out until we needed to throw it away for food safety reasons.
We picked up trays of noodle salad, rice, and curry from a vegan restaurant, Red Lentil, which gave us a discount. We also ordered trays of chili, stew, and pasta from a local restaurant, Cafe Luna, which delivered them to the conference location; the pasta had cheese, but everything else was vegan.
Because we were self-catering, we also bought cheap plastic tablecloths and serving dishes, additional temporary trash and recycling bins, plates, cups, napkins, plastic flatware, and water (both in bottles and jugs).
We were contacted before the event by a few people who had dietary concerns other than veganism or vegetarianism, and we let them know what food we would have available at the conference and that we didn’t have the resources to handle further specialization. This was true! For instance, we bought avocados with the intention of helping people on paleo or similar diets - but we didn’t have the organizer and volunteer time to make sure that someone was always on hand to get them to the people who wanted them. Instead, we tried to make sure we had a wide variety of food available at all times, so people could select foods that would work for them. We made labels on index cards for foods containing dairy and eggs and for other common allergens that weren’t apparent, especially nuts. We left the processed food from the grocery store in or near its original packaging whenever possible, so that people could resolve their own questions about its ingredients.
We had plenty of food, which meant that there was food left at the end of the event for us to deal with. Two people from our local chapter of Food Not Bombs came and helped us with cleanup by taking away all the food and catering items that they could reuse, mainly food in unopened packages or which was vegan. They came while the dinner food was still safe to put away and eat later, but the lunch food would not have been safe, so we had already discarded all of the lunch food which would have been at risk of spoiling. (We didn’t have access to a refrigerator.) Even though there were some things they couldn’t use, they took away a huge volume of stuff and were happy to have it, so we were really glad that they were able to come!
We marketed the conference through the subdomain on eaglobal.org, our own website, a Facebook page and event, email, and posters. We had a registration process rather than an application process, because we thought it would be more welcoming and we were more worried about having too few attendees than too many. However, we did first promote the event to the groups which we thought would be the best attendees, such as the Boston EA mailing list.
We set up our subdomain on the EA Global site, but we didn’t think it was flexible enough to hold all the content we wanted to put up (particularly speaker and panelist bios), and it didn’t work well on mobile. Randy put up another website on our own domain, which he was able to configure to better fit the information we wanted to share, particularly as more content became available. During the lead-up to the conference, we continued to update the subdomain with any changes to information that was already there, but we put new information only on the other website (and in Facebook promotion as applicable).
We linked the Facebook page and event together, and we linked both to our website. The event also linked directly to event registration, which we ran through Eventbrite. We all “liked” the page, indicated we were going to the event, and invited our local and/or potentially interested friends to the event as soon as we felt the website and other online materials were ready. We also developed a schedule for posting updates to Facebook every 2-3 days, mostly promoting our keynote speakers or specific panels. We posted each of these posts, with an appropriate image, to both the event and the page. We had some extra room in our posting schedule and a growing audience, so we repeated some early posts as we got closer to the conference date.
Some people asked us questions on Facebook; we had a policy of responding as the page if we were sure of the answer and as ourselves if we weren’t as sure, which let us respond to questions more quickly. We didn’t have any incidents where one of us gave an answer another was very unhappy about.
Some of our panelists and speakers spontaneously posted about our conference on their social media accounts, particularly when we had just mentioned them in a Facebook post. We encouraged the rest to mention us on social media during the week before the conference, so that anyone who wanted to see them speak would know about the opportunity. We provided a sample tweet to make things easy.
Posting updates to Facebook was important! While our tickets sold at a fairly steady rate throughout the month when they were available, there was more activity just after a Facebook post and less as time passed without a new post. We didn’t want to overwhelm our followers, but it was clear that each time we posted, some new people learned about the event, and some people were reminded that they wanted to buy tickets.
We sent emails promoting the conference to groups we thought would be interested in it, and to individuals representing such groups. These included EA groups on the East Coast (from a list CEA gave us), local and nearby colleges and universities (where we tried to reach an EA or similar group if possible), and local organizations and email lists with EA-friendly themes such as rationality, humanism, and non-profit work. We used an email template which we modified depending on whether we knew the person we were contacting and whether the group we were reaching out to was an EA group or less closely related.
Shortly before the conference Dillon learned about a way to easily contact professors in relevant departments to ask them to let their students know about our event. We were able to send about 400 emails this way, because there are so many large universities in Boston. We wish we had known about it sooner!
We designed both full-color posters and posters that would render well in black and white, and we attached them to our emails so that our contacts could put them up. We also put up posters at local universities which members of our organizing team attended or where we had contacts, since we expected students to be more interested in the conference than the general public would be. We tried sending teams to put up posters at local universities where we didn’t have contacts, but we found that many bulletin boards had policies that posters had to be sponsored by student groups and stamped in a campus office. Since we couldn’t arrange that, we gave up on postering schools where we didn’t have a connection who could help us.
On the day of the conference, our organizing team (5 people, plus Julia as CEA representative) and volunteers (about 12 people) arrived at the venue around 9 am to start setting up. Registration officially opened at 12:15 pm, but people started arriving a bit before that.
Before people came, we set up the food tables and snack food (and lunch food when it arrived later), set up the sound system in the second room, posted signs around the building to make sure people could find the second room (which was up two floors from registration, the food area, and the auditorium), and set up the registration area and welcome signs. We put bottled water in both rooms so that it would be available for speakers, and we set up both rooms so that they were arranged how we wanted them, with tables for the panelists to sit behind and computers connected to the projection systems. We sent some volunteers as “runners” with a car to pick up additional food from the grocery store and our coffee order. (They later went out to pick up additional food orders.) We also assigned volunteers specific responsibilities, such as helping with registration and handling mics for questions in the auditorium.
We were still printing nametags as people started arriving for the conference. We had a backup list of attendees and blank nametags with markers, so we used those until the printed nametags were finished. In some ways, this actually worked better than the printed nametags, which were originally in sheets of 8 in alphabetical order. As some nametags got handed out, the sheets fell apart, making it harder to find people their nametags. If we’d thought about this and had the time, we might have detached all the nametags and arranged them in a narrow box or file so that they’d stay in order better.
In the auditorium, we had a technician from MIT, who we discussed our plans with for the initial speakers. We didn’t realize that there wouldn’t be a lot of time between events to keep checking in, so while the keynote itself went smoothly, the start of the first panel was awkward because he didn’t know which person at a microphone was the signal to turn microphones on. We also had a lot of microphones in use - several at a table for panel discussions, a wireless lapel mic the keynote speakers could use, and two handheld wireless mics for presentations and for volunteers to bring into the audience for questions, which made communication especially important. We also had an organizer making supervising the projection system and helping speakers, and organizers or volunteers bringing microphones to the audience and timekeeping (showing a series of pages to the speaker or discussion moderator to let them know how much time remained in their presentation and keep the room on schedule).
In the second panel room, we had a volunteer running the sound system, an organizer supervising the projection system and helping speakers, and an organizer or volunteer timekeeping. We had a standing mic in the audience area for Q&A, but the classroom size was small enough that people ended up raising their hands and asking directly from their seats, which also worked out well.
We didn’t leave anyone in charge in the food area consistently, which might have helped with some confusion we had when things needed to be replaced or refilled and no one was specifically in charge of knowing where things were and what to do. However, various organizers and volunteers replaced food and supplies which had run out, removed food which had been out long enough that it needed to be put away, and refilled pitchers and jugs of water.
80,000 Hours decided about a week before the conference to send representatives and offered career counseling during the conference for attendees. We largely provided them some space and notified attendees, then left them to their own devices.
After the last talks, we took down the sound equipment and our runner returned it when he took back the car he’d been using (borrowed from a friend). We waited to take down the registration and food areas until near the end of our reservation for the space, when we needed to get out of the way so that janitorial services could do their part of the clean-up. We left the venue around 10 pm.
While most of our organizing team was working (or at least looking for potential issues) most of the time, we tried to allow most volunteers to have about half of the conference free. This wasn’t fully successful; we hadn’t planned ahead and had more volunteers for the first shift than the second, so we ended up asking some to do additional tasks when they were available. (For instance, to carry microphones for audience questions in talks they chose to attend.) Another problem with not having many second shift volunteers was that by the time we started cleaning up, we didn’t have much volunteer help.
We sought funding through ticket sales, outside sponsorships and donations, and CEA. We ended up raising money through all of these methods, and also getting an offer from an individual donor to be a backstop if we had expenses we couldn’t fund. We didn’t want to have to use that offer, so we proceeded as though we weren’t going to; in fact, some of our organizing team didn’t know about it until after the conference. We ended up with access to about $6,800, including day-of ticket sales:
Since we tried to do everything so cheaply, we spent much less money than we had available. In total, putting on the conference cost about $3,900. This was about the bare minimum we could have spent, and a trade-off with our experience and probably conference attendees’ experience as well. The major places where it would have been nicer to spend more money were catering and recording. We put a lot of extra effort into obtaining relatively inexpensive food and also into setting it up and serving it, and paid caterers would have done a better job with set-up and serving, though the quality of the food itself was good. We didn’t get any recordings of our conference at all, and there were some great talks and discussions.
Breaking our spending into categories, it looked like this:
Because we had access to more money than we spent, we ended up not using any CEA potential grants, using most of the MIT grants, and we will still have some surplus to donate to effective charity. We haven’t yet reached a consensus on exactly where and how to donate it (whether to organizations directly or to finance Giving Games conducted by local EA groups).
January 28: We started expressing interest in organizing by email so that Taymon could fill out the initial application to hold an EAGx conference.
January 31: Applications were due; Taymon submitted one for our group.
February 7: We held our first organizing meeting. We started contacting potential keynote speakers.
February 9: Randy, Taymon, and Mi Shan met with CEA to discuss our application.
February 12: We selected a venue and a date.
February 18: We got a quote for audiovisual services at our venue.
February 19: We sent CEA a proposed schedule, funding plan, and outline of organizers’ responsibilities.
February 27: We started contacting potential speakers/panelists more generally.
March 3: CEA officially approved our event to be an EAGx branded conference. We started talking to caterers.
March 16: We started working on marketing materials.
March 19: We started working on our CEA subdomain website.
March 22: We opened ticket sales, but not yet promotion.
March 29: We finished confirming keynote speakers.
March 30: We launched a separate website which we thought would better accommodate our content.
March 31: We sold our first ticket.
April 3: We started promoting EAGxBoston on Facebook.
April 5: We started promoting EAGxBoston through emails to other related groups. We started seeking sponsorships and in-kind donations from companies and organizations other than CEA.
April 11: We got a quote for video recording through our venue’s preferred provider.
April 13: We got a quote for alternative audiovisual services in our secondary room.
April 16: We asked volunteers to come to our organizing meeting, to talk about what they’d be doing day-of and get to know the organizing team. We started sending out the speaker guide.
April 19: We confirmed that we had enough funds available to cover the conference expenses.
April 20: We confirmed our audiovisual equipment and services needs with all relevant parties.
April 25: We finished confirming panelists and moderators.
April 26: We finalized all catering plans, including donations.
April 27: Food Not Bombs offered to pick up leftover food from the conference.
April 28: We sent out the attendee guide.
April 29: We sent out the volunteer guide. Randy spent 15 hours picking up supplies and food along with prepping food for the conference.
April 30: We held EAGxBoston.
May 1: We started sending thanks to volunteers and speakers/panelists.
May 10: We sent a post-event survey to attendees.
June 8: Last reimbursement request came in and finances finalized.