In 2014, our first year, we had room for 120 people. After subtracting about 40 spots for speakers, organizers, and sponsors, we had about 80 tickets available. We gave people who’d submitted talk proposals first crack at getting a ticket, and they claimed about 35 tickets. Finally, we opened up the remaining tickets to the general public, and they were all claimed almost immediately, with about 100 people signing up for the waitlist. Since it was our first year running the conference, we didn’t know how popular it would be, and we were pretty floored by the response. We were happy that so many people wanted to go, but sad that we had to disappoint a lot of people who hadn’t been online at the exact right moment.
In 2015, then, we decided to do things differently and hold a ticket lottery. About 320 people signed up for the lottery, and we offered tickets to a randomly chosen subset of those people.1 The lottery approach was arguably more fair than the first-come-first-served approach we’d taken in 2014, because it didn’t require people to be online during a very narrow window of time in order to get a ticket. Unfortunately, in 2015 we had a much higher no-show rate than we’d had in our first year. (There were basically no no-shows the first year, which is pretty amazing for an event with 120 people!) We hypothesized that with the lottery approach, the people who ended up with tickets weren’t necessarily the people who were most interested in attending.
In 2016, we moved to a bigger venue with room for 250 people. We had a few more speakers and sponsors than we’d had in previous years, but even so, we still had room for about 200 regular attendees. We decided to drop the lottery approach, and to my recollection, we didn’t set aside any tickets for talk submitters, either. When we opened sales, about 90 tickets sold out within a few minutes, and then then it took several hours for the next 110 or so tickets to sell. The waitlist accumulated about 120 names. This was useful information to have: it told us that there were about 90 people who were really excited about going, and then a couple hundred people who still wanted to go, but were more casual about it.
The bigger 2016 conference was a success, and so in 2017, we upgraded venues again, to a space that accommodated 280 people, and we decided we could return to offering the
Historically, !!Con tickets have sold fast — really fast. Last year, according to Eventbrite’s data, we sold out in about 22 minutes. Eventbrite had an 8-minute maximum time window to hold a ticket before buying, and so depending on how many other transactions were in flight at the moment someone visited the site, they might have seen an “unavailable” message sooner than that —
Lottery pros and cons
For the first couple of years of !!Con, in 2014 and 2015, we had small venues that only accommodated about 120 people. After setting aside tickets for organizers, speakers, and sponsors, there weren’t that many tickets available for the general public. The small number of tickets we had in 2014 sold out almost immediately. Therefore, in 2015 we decided to set up a ticket lottery, so that people’s ability to attend wouldn’t depend on them being online during a very narrow window of time.
The upside of the lottery is that it doesn’t require people to be at their computer within a very narrow window of time in order to get a ticket.
In 2016, we moved to a venue that was twice the size, and since a lot more tickets were available, we decided to discontinue the lottery.
The choices weren’t quite independent: we asked people to tell us if there was another person that they didn’t want to attend without, and in that case, I believe we made sure to either choose both or neither of those people. I don’t know exactly how we did this. ↩