Jon talks with Katie McLaughlin about organizing PyCon AU, assembling a team of conference organizers, and many of the core roles required to start a tech conference.
Note: this episode was recorded in June of 2022 and published at a later date.
Jon: Hello, and welcome to Conference Talk, the podcast where we take a backstage look at all your favorite tech conferences. Joining me today is Katie. Katie, do you want to tell us a bit about how you got into conference organizing?
Katie: Hello, I’m Katie. One time at a conference, I had a conference organizer pull me aside and said, hey, do you want to run this conference after me?
Jon: Awesome. And what have you been up to ever since then?
Katie: Oh, running a little conference known as PyCon Australia for the last four years.
Jon: I think it’s probably a bit bigger than you give it credit for there. I think most folks familiar with the PyCon family conferences are familiar with PyCon Australia and all the fantastic talks that come out of it. So some years ago, someone pulled you aside and said, “hey, run this conference”, and you’ve been doing it by yourself ever since. Is that about right?
Katie: Yes, myself and dozens of other people having to put this whole thing together. Aside from PyCon Australia, I’ve also helped run other events as part of a team of dozens, but you cannot run a conference on your own. Even a fake conference, which I have run before, has been a cost of at least two.
Jon: Fake conference is, I think, a wonderful topic for another episode because I do want to hear how you convince the world that a conference that didn’t exist does. So you mentioned these dozens of other people involved. That’s really what I want to talk about today is all the people involved in running a conference and how you go about putting together that team. So you say dozens, is that where you would recommend starting out for, say, a brand new conference?
Katie: Absolutely not. A brand new conference needs a nexus of people who are willing to put in the time and the energy over much longer than what you’d expect to get an event running in my experience. Even taking over an event, it’s 18 months of work for a core team for a two or three day event. It is not an insignificant amount of lead up time and not an insignificant amount of actual work that needs to be done.
Jon: Yeah, I mean, that’s a big time commitment. More than I think most people are really envisioning when they say “I want to start my first conference”. So how many people are in that core team, especially when you’re just getting going when you’ve got that 18 month lead time?
Katie: I found that this is all… Dear listeners. This is all based on my own experience. Most of this is going to be PyCon Australia, but some of it is going to be based on other events that I’ve helped out with. But in my experience, a good team of five or six in a core team that starts off early and then adds more to it as different roles and requirements happen is probably best. And if you have five or six and some of them might have to drop, some might have to be added, but you at least have that core nexus of people that can get stuff done for the long haul. And that is one of the major things because when I say 18 months, I’m talking about events that require spaces that have lead times such as conference centers, where you cannot book something for next month at these places, you have to book far in advance. And so that is going to be one of the main timeline considerations is where you’re going to be running your event. And so you need to have that energy from that time all the way throughout as you build and grow and put on a really good event.
Jon: Yeah, so you’ve got your core group of five or six people. You’ve all understood that this is going to be kind of a long haul, 18 months out before you’ll really see the event come to life. At what point do you start kind of delineating the different roles? And on that note, what are the different roles you want to involve across these different stages of conference planning?
Katie: So we’re going to presume that you’re running a technical conference here. Technical conferences need three things. They need a venue. They need attendees and they need speakers. Your venue is going to be one of the things that you’re going to need someone to help with a lot, especially when you start thinking about, oh, “how many rooms do we need?”, “how are they going to be set up catering?” Then you logistics that kind of thing. So you’re going to need someone who does that kind of thing. I found that that is probably a role that should be a dedicated person, but it is also one of the visionary aspects. So depending on how you set things up, you can either have a dedicated person for this or the main director does this. Now, I’ve been the main director a few times now, and I found that as long as the director doesn’t have any dedicated roles, they can jump in and help out wherever. Second thing you need is attendees. So you’re going to need someone who is going to be helping out with your ticketing and the registration process, and that in theory can be a dedicated person as well. All these people need to talk to each other, of course. Your selling of tickets needs to be coordinated with how you sign up people on the day, how they get their lanyards, that kind of thing. Third thing you need is speakers, and you absolutely need a dedicated person to run your program. Selecting speakers is not an easy feat. So of all these three core roles, there’s probably also the director and maybe a co-chair to take over. They’re the three main things, but there are going to be so many other roles as well. You’re going to want to have a person who doesn’t have any other work who’s helping out with safety and code of conduct and spinning up their own team. The program chair will have to spin up their own team for reviewing content. The venue person will want to spin up their own on-site volunteer group, and these different groups have different canons as well. Your reviewers will want to be onboarded as your CFP is open. They will then review stuff when your CFP closes, and then they’ll spin out. Your on-site venue group will spin up just before the event and spin down as soon as the people leave the event. Your code of conduct group will want to spin up early to make sure that everyone is safe and have communication channels, especially for any interpersonal things with your team itself. Therefore, they should be a person with no other roles, and anyone who helps with them early needs to make sure that they’re not part of the team itself, and they will have more of an on-site responsibility as well for physical attendee safety, and then after the event is over, then they’ll deliver their reports and spin down. It is very much a very scalable group of people, and this is where the dozens come into it because the amount of slides that you need at your closing to thank everyone that helped is always bigger than you might think, and you will always miss someone. You can never remember everyone.
Jon: Yeah, I think that’s a great measure of just how many people are involved in putting together one of these things, right? The closing remarks at an event like this, you spend a significant portion of that attempting to thank everyone that was involved, but if you list out all of these names, it’s the dozens of organizers and volunteers, all of your speakers, all of your sponsors, the venue staff, catering staff, I’m sure plenty of other people that I’m not appropriately thanking off the top of my head here.
Katie: And even though you raised, like, I’m not thanking everyone, I completely forgot to mention some roles such as sponsors because for PyCon events, it is much more accessible if some of the cost for attending is offset by sponsors. So you need someone to help wrangle the sponsors and then do all the things that the sponsors get as part of their benefits. So you need a dedicated person for that. You need to have someone who liaises with both the venue and the registration to make sure that your registration signup ticket form has spaces for the people who get into arguments with food sometimes or have different dietary requirements and making sure those get passed along. And the logistics of that venue staff are really great for this, like having suggestions of let’s make sure that the main thoroughfare for the majority of people food is this way and make sure that the special dietary requirements is this way. I’m probably going to be in half an hour from now remembering more people that need to help out. But those were two of the big ones that I just remembered based on your list of people to thank.
Jon: Yeah, so I’m trying to keep track here. I think we’ve already surpassed our five or six core organizers with number of roles. Just those kind of director chair level roles. We’ve at least hit five or six and maybe gone over that. So do you tend to see organizers picking up more than one of these roles in a given year?
Katie: It depends on the size of your nexus group. I know that sometimes it can feel like I know that I felt that as a director, it’s just like, “oh, no, I can pick that up, I can get this done”. You can’t get this done. But having a core team of five or six that are dedicated for the long haul and then bringing in new core members like your volunteer coordinator could come in later on. And then that five or six turns into six or seven. Your program chair may want to only do the program stuff and then they don’t feel like they need to be included in the core team anymore for ongoing discussions because program is done. So they may step aside. And of course, remembering that at any time, these are all volunteers who aren’t being paid, who need to do other work to get paid and life commitments and such mean that at any time, any one person may have to bow out unexpectedly. So there’s always going to have to be more work than you think has to be done and just make sure that your leads do not have dedicated roles. It means they can’t be the flex point or get like a gopher who can be a person who can just pick up whatever’s happening or help find someone to pick up whatever’s being dropped.
Jon: And you touched on this a couple of times already, but I think it’s worth diving into it explicitly. What are some of those kind of key touch points of when you are onboarding and offboarding different organizers and volunteers? Because if you have your two, three, four dozen people all involved in the same room for the entire 18 months, I think that may be quite a lot to handle.
Katie: Yeah, a lot of it is to do with like, you have your volunteer coordinator who then coordinates the volunteers. So they might have their people and then your program team will have their content reviewers, so they would have their people. So at no point should you be trying to manage everyone at once, you should be delegating. That really helps a lot. But you also need to make sure that those people who then delegate onwards have the power within their domain to get stuff done. And of course, at any time, like anyone can ask questions, can like raise things. There is definitely a semi informal chain of command thing, but as long as you delegate, you don’t have to do all the work. And that’s coming as a conference director who learned how to delegate after.
Jon: So we’ve covered a lot of the points I have in my notes already. I think the next major piece is how does this team change in subsequent years? How do you get from your core group of five or six people in say year one to how long has PyCon Australia been going on?
Katie: Since 2010. So a few years.
Jon: Yeah. So I imagine you have not only a larger, but a very different team from when the conference first started.
Katie: You say that. And yet last year, in 2021, we had as one of our core directors, one of the directors of the first one. But that is a story that will come up later. When you’re running an event for the first time, it’s completely different to inheriting an event. And I inherited PyCon Australia. PyCon Australia is an event that’s run under the auspice of Linux Australia. And Linux Australia is a Australian based non for profit that does a whole number of things, including being a entity that helps with the financial and insurance stuff for running various open source events in Australia. Some other events that there’s that you may have heard of include Linux Conf Australia and local WordPress, WordCamps, Drupal events, Joomla events, Gov Hack back in the day, and the Open Source Developers Conference. But they are an entity that helps with some of this stuff. But part of that is they have their own processes in place for some of the bigger events to ensure continuity and handover between different organizers. Now the two big events that they run are Linux Conf Australia and PyCon Australia. And so we’ve sort of picked up on some of the things that Linux Conf Australia does, including bidding for cities where if you’re going to be running an event that is going to be hundreds of thousands of dollars in ins and outs between venues and catering and everything else, you probably want to have a budget before you start committing finances. And so one of the things you have to do is say “hey, we’ve put in the time and effort to consider how many people we expect to turn up, how much we expect them to pay and how much it’s going to cost to host them”. And so putting together that budget before you get the keys to the event is a thing. And subsequently, when you sign up for an event, you probably want to sign up for a specific amount of time, be that one year, two years. And so making sure that you have people who can be enthused to take over from you is a thing, which is why I was asked in a hallway two years before the event I would run “hey, do you want to run this next one after I finished it?” So I ended up embedding in that team for the six months between when I was asked and when they would officially hand over. So I was able to learn all this stuff by osmosis and thorough documentation. And I did the same thing when in theory, I had a person who would be embedded in my team for the six months before their event. But then this thing happened. Not sure if you remember, Jon, everyone had to stay inside for a while there. But in theory, it would be a would have been a two year commitment for this particular event hand over to the next one, but still be around if they needed help. And then in theory, another two year handover, another two year hand over. PyCon US does similar with the two years, which has plus sides and downsides. Some events do annual handovers. Some events round robin between a group, which is also fine. I mean, in all of this stuff that I’m talking about, you’re putting in so much time for a weekend or a couple of days, and your attendees are not going to see all the work that goes into it, but they are going to see the event and experience the event and want to come back hopefully to your event. And that is what you’re aiming for at the end of the day is a happy, healthy, learning, educational, entertaining experience for a couple of hundred your closest mates, descending on your city of choice to learn about some geek goodness.
Jon: Can you go a little bit more into those different handover processes you mentioned and the one or two year cycles, the round robin option, and how you actually handle that six months of learning by osmosis and things of that nature? Because this is all well and good as long as everyone is still enthused to continue running these events year after year, but it takes a lot out of people. It’s a lot to ask for someone to be working on an event for 18 months and then immediately six months before the event happens, begin planning the next one.
Katie: So I can speak for my experience here whether it was a successful handover or not is left to others to decide. So it just so happens that calendar wise, Linux Conf Australia happens in January and PyCon Australia happens in August or September. So when you have similar groups of people hanging out every six months, you end up getting the wonderfulness that is, oh, there’s six months until PyCon. Would you like to have a PyCon next year? So literally the conversation in the hallway that I mentioned earlier was January 2017, where in the August of 2017, I got to go up on stage saying “hey, in a year’s time, I’m going to be running an event”. So that’s where the 18 months comes in. For those six months, I was sitting in on the weekly or bi-weekly core nexus calls listening to what was happening. I inherited some documentation, but I was also creating my own things. And in all of this with a inherited event, if you are taking it over, you’ve probably attended. And so there are some things you want to keep for the continuity of the event, some things you don’t, one of the things that I didn’t keep was the branding. We rebranded our entire event, and that cost money. We actually paid a wonderful designer to help us rebrand. But then the next year could choose to keep that or develop their own. I know that PyCon US tends to have a different color scheme and iconography each year, and that also helps to identify which year it is. We ended up choosing to keep our branding consistent for the last couple of years, which has helped and hindered because sometimes it’s hard to tell if you’re looking at any things like which year was this from. Besides that, the six months that I was embedded in this team was really helpful because I was able to see behind the scenes and learn about the stuff that I wouldn’t have otherwise been able to learn. But I also had no responsibilities or roles within that team just to learn and ask questions and see what was all about and make my own notes and my own plans for my own team. And so the six of my team got up on stage at the end of the last event going, hey, in a year’s time, we’re going to be running a thing. Doing that really helped for continuity of the event as well. Because in a two-year cycle, you would think that after your first year, you would know when your next one is going to be running. But once you change organizers, it’s very useful for attendees to know, okay, we’re running again next year. If you already have a venue and a date, that’s also grand. And then they know that, like, “Okay, I’m going to come back. I’m going to submit a talk. I’m going to think about these things”. And then I run my event. And then while I run the first event, and then I’m like, “Okay, I’m going to run the second event. And now I need to start onboarding someone else to take over from me”. And so I got them into the organizing group just as I had. With no responsibilities, they were also planning their own ideas, their own things that they wanted to duplicate or make new or revamp, which is all within their prerogative. I mean, any changes that make an event better, but also dropping things that are felt to be not worth the effort is absolutely fine. Such as mentioned at the top, a little conference called PyCon Australia, it was little when it started. The last couple of years, it’s been bigger than Linux Conf Australia. And at that scale, you can’t do things like having an all attendee dinner. Because no venue that you can afford will want to host 700 people for dinner. And so you end up getting the thing we could have done is have the separate dinners that PyCon US has where you can choose if you register early to have a dinner with 150 people as opposed to 2000 people. But we felt that that was too complicated. So what we had was a “find your own dinner” thing, where we suggested that people form groups of six and go out into the Sydney CBD on a Saturday night and try to find tables. And we did have someone who preemptively reserved some tables at some restaurants. And so that worked really well, especially when it’s a case of, hey, you can’t know more than one person in your group to try to make new friends. And all reports are that that went well. Things that you try, things that you don’t try. And a lot of this you can’t just learn by listening in because you’re only one person. So sometimes documentation is helpful. But the documentation that you inherit or you take is going to be of some use, but it’s always going to be a case of there are going to be other questions. And as a director, if you give a whole bunch of set of instructions, the next person’s not going to do everything. You can’t tell the next person what to do. You can absolutely offer advice if they ask. But if you want to direct another conference, you be the director and do the work. Don’t just tell people what to do. But the documentation has really helped for at least some of the processes that we found that helped. But it’s always evolving, always changing as we learn better ways to do things, that kind of thing.
Jon: Yeah, I may ask you to do a deep dive on documentation, best practices in a future episode because I think that’s one thing a lot of us organizers struggle with. There’s so much to do who has time to write it down. I wanted to touch on one more thing before we wrap up because I think it’ll be relevant to a lot of organizers who maybe don’t have that existing support structure. So you mentioned that Linux Australia handles some of the higher level aspects and provides a bit of additional structure to PyCon Australia, Linux Conf Australia, and all of these other events that are sort of related. What are some of those aspects that say someone starting their first event might need to account for when building their organizing team?
Katie: One of the biggest initial costs apart from workforce time is going to be venue deposits. And that is going to be something that’s going to secure your venue. But depending on the scale of your event, once you start talking to conference centers in Australian dollars, we’re talking 200,000 Australian dollars total venue costs and then catering. The deposit that you’re going to need for that is going to be maybe 10%, 15% or something. I’m not sure of any person who would be actively willing and knowledgeable enough to not throw down tens of thousands of dollars of their own money to kickstart an event. Now, there are ways around this. I have heard people who pre-sell tickets to attendees to then afford the venue deposit, but that takes a lot of trust, especially for a new event. I’ve also heard the stories of “never put down your personal credit card” because the guy that ran the first Linux Conf Australia paid for it on his own credit card, never organized another Linux Conf Australia again and was paying off his credit card for a number of years afterwards. Those are the kind of things that having an organizational structure really helps with. It’s just that bootstrapping process. And I know that some of the PyCon regional events have that sort of thing as well, where the profits from the previous year can help kickstart the next year. But when you don’t have any of that, you’re going to have to start small or you’re going to have to think of a way to just make this possible. And taking out loans is not something I would recommend, even without considering the fact that there have not been a lot of in-person events the last couple of years because of reasons. And starting a new event now, deeply consider the comfort levels of your attendees gathering in person, regardless of where you are in the world and what your local requirements and restrictions are. People are going to be scared right now. So keep that into consideration and do not presume troves of people at your events because things are weird now.
Jon: Yeah. I don’t have anything to add there. I think it’s a weird time for organizing events.
Katie: I will say, on the documentation side and the weird side, one of the things I had the energy to do after my original two events was write down all the things that worked for me. And the theory was that I could publish that, make it a public GitHub repo, and offer that for anyone to consume. By the time I’d finished writing out a first draft of that, we weren’t doing in-person events anymore. And so the use of that wasn’t very good. It was great to get it out of my head. And I hope that someday someone else might find something useful in there. But one of the things that I am very regretful for but I had no control over was the fact that I put so much energy into making sure continuity for my own events and then passed on that to my next organizer who was able to announce that we will be in a city in August 2020. They weren’t in that city in August 2020. But I mean, acts of God, it’s fine. But one of the things that I’m deeply grateful for is they did pivot hard to an online event. And so we had that continuity. And then decisions were made to do another online event using only previous organizers. And then we decided not to run again this year. But I think that might be a topic for another time. Yeah, I think so. Two more things before we wrap this one up. One is you already, I mean, all of this is great advice already in terms of documentation and finding the right people to build your event as well as the starting small or finding an organization to partner with. Any other kind of last words of advice for say an organizer who’s really trying to start their first event in maybe 2023? One bit of advice as a core Nexus lead organizer or anything that requires on-site presence. Get accommodation near your venue. That half hour that you’ll save each day getting from where you’re sleeping to where you’re working is really going to help. Even if it’s a cost, maybe see if you can, like if that’s not accessible for you, maybe see if you can incorporate that into your event costs. Because as one of the lead organizers for your event, you are never going to know what’s going to break. And so being available as an active idle person throughout the event as early as possible as late as possible is really going to bring ease in case anything comes up. Also, do not expect to see anyone at your event. I could not tell you anyone who attended my event apart from looking up the speakers and my keynotes. But you are putting on an event for hundreds of other people. You are not attending the event you run. If you happen to have everything organized and you get to see a talk that you’re not introducing, you’ve done real well. But know that this is a sacrifice, a responsibility. And you are doing this for the benefit of hundreds of others at personal cost. And know that ahead of time. The benefits of having done the thing is great. But on the day, it’s going to seem lonely and a lot. But it’ll have hopefully been worth it.
Jon: Yeah. I mean, I really appreciate all the event organizing and general community involvement you have. So I know this can often be a thankless job. But there are a lot of people out there who do really, really appreciate it.
Katie: And if you really, really appreciate it, get in touch with your local conference organizer and see how you can help today.
Jon: I want to make that just the tagline from now on. One last thing. I’m going through my notes here and seeing this time on here. So PyCon Australia does a two-year cycle, you said, right? Yes. And you were approached or signed up, however you want to phrase that, in 2017. Yes. But it’s 2022 now.
Katie: Yes.
Jon: What happened there?
Katie: So I was approached in early 2017 to run 2018-2019. So I was able to announce at the end of the 17 event that I would be running 18 and 19. In theory, then at the end of the 19 event, we announced that the next team would be taking over for 2020-2021. However, something happened in the world which meant that wasn’t possible. These things are going to happen, hopefully not as frequently and as big as this current asteroid that we’re still cleaning up from that is a lot. But one of the things that I felt responsible for was wanting to make sure that the next event got off to a good start. And it just so happened that one of the core members of the 2020 team came down with a disease of note and so had to drop from the event. And this was very early on as well. I happened to live in a country where we had a slightly delayed effect from all this stuff. But we were still seeing cases all the time. And when it’s a person, it’s never a case of numbers, it’s a case of a person. So I was approached with a, “hey, one of our core team members dropped out, can you help us make this happen?” And so I did. Because I was asked and because I felt partially responsible, even though I’m not responsible for the current health crisis, as much as my anxiety wants to make me feel responsible for making people hard pivot to online events. But I was not to know. So 2020 happened with everyone trying events for the first time, trying virtual events for the first time. And that went remarkably well. We had a bunch of people come in, we had new platforms, we were trying, we had a whole bunch of different things going on. And it went remarkably well for an online event that was being run less than six months after everything went down. We did have a bit more time to pivot than some other events, but I think we did remarkably well. The decision was made that we would try it again in 2021, but the core team for that one was only previous organizers and not any of the 2020 team, who we told them that all your core team, you are not helping in 2021, you are having a break. Because in theory, you are going to be running two years of your own event. We don’t know when that’s going to be. But we are not going to make you do the work this year. So we will take over. And last year’s event was an event that happened. But it was a lot to try to undertake, especially from some of our core team that have been doing it four years in a row, which was double what we said we’d do. So the decision was made because we could not find the continuity of nexus team members to continue for this year, that we would not run this year and focus all the energy on an in-person event next year. And so one of the only things I did this year was make sure that our 2022 page was up talking about all the local meetups that were happening throughout the year. So people in Australia can go to https://2022.pycon.org.au and take a look at the list of meetups happening in their local area, which means that you can quickly choose to attend an event that requires minimal travel for you and still meet up with your Python friends. And those meetups run at a much shorter cadence than conferences. You can spin up a meetup a whole lot faster than a conference, but that’s a different podcast.
Jon: Yeah, I think that’s a great note to end on. And I want to thank you for sharing all of this with the audience and coming on the show. Is there anything else you want to pitch before we end? I know everyone go to the 2022 website, find your local meetup.
Katie: For those listening to this podcast who aren’t in Australia, if you just go to https://pycon.org, you will see a list of regional conferences and also right at the bottom of the page, there’s a calendar of events that will include meetups in your local area. If traveling on planes isn’t a thing you’re comfortable with, but you are comfy with a train or a bus, most major cities have something happening. And if yours isn’t happening, maybe you get in touch. If you feel like you have the capacity to run something smaller, if you want to try out running a gathering, meetups are a great way to start. Get in touch with your local organizers if that is something you feel like you have the energy to do. Because smaller events right now are probably the way that we can gather safely.
Jon: Wonderful words to end on. I really appreciate it again and come back again soon.
Katie: I just might.