Tooling your way to a great devrel team

Cristiano Betta: Anyway, hi. I'm Chris Samuel. Thank you all for joining. I was gonna ask the question, like, who here is setting up their DevRel team, but Rob kind of beat you to the punch on that one. So thank you all for putting up your hand for that.

I work for Brainfeet PayPal. I'm going to be talking about tooling your way to a great DevRel team. It's not gonna be as animated as Rob's talk. There's videos, there's GIFs. This is the animation I have for you.

Woah. Woah. Yeah. Let's do that again one more time. That's it.

That's it. That's what you're So getting. I'm senior in Bell Radford at at our team. I've been doing this for three years. And when when I tell people that I what I do, they're like, oh, you're you're one of these people who travels a lot.

You you must be like, you know, this sounds like a lot of fun. And I'm like, well, you know, that basically makes me an expert in kind of at work safety manuals, like airline safety manuals, which I don't know, like the rest of you, I probably sleep through it. Who here sleeps through those by now? Yeah. See, this is the hardcore.

Yeah. Yeah. Yeah. Yeah. Exit there.

On that note, the where are the exits? We should probably know. Anyway, so I travel a I do a lot of other stuff on our team as well. So one of the things I do is one of the other titles I have on my team, although it's it's changed a lot, I've decided to give it this title is master builder, which, yes, is a LEGO term, but that's a LEGO me. So I'm I'm allowed to.

Right? And I will do. I think any developer or analyst has a history. You you can't see a developer or evangelist just as what they are in the role that they do. You need to look at a wider picture.

So for me, it's I've done about six years of startups, some consulting, some freelancing, mainly here in London. I've been CTO, lead dev, DevOps, which is basically doing DevOps slash, which is IT support for anybody who needs it. You know, my mom mainly. So a lot of that kind of stuff. And since 2013, I've I've been in the first paid vault devs, then we changed that into Progentry dev.

So lots of fun stuff there. Now what does Progentry do? For those of you not familiar, we power payments for pretty much these companies. If you don't know them, use them. They're pretty cool.

We also like to have fun with Ascio. This was in Mission in San Francisco just to show that that's real. That's me and Joe, like, actually there. I travel a lot for battle hack. Who here has been to battle hack?

Yeah. Awesome. Cool. So battle hack is like a crazy for those of you who don't know, it's a hackathon, but we you know, when you say, oh, a hackathon organized by PayPal, that might not be that we like to have a lot of fun. So this year, we've got a full on neon style.

Last year, we had Vikings and, of course, the the the winning, you know, beside the trip to the San Jose, we we give away axes as trophies because we just like seeing hackers on the underground with an axe because that's the worst thing that can happen. But that's one of our programs. Right? So this is our program for developers. We have a whole bunch of other programs like Commerce Factory we ran last year.

We're kind of facing that one now. There's a media series that we did. We run Blueprint, which is where we give startups free processing for the first, like, eighteen months of their existence. We have a program where we give them away office space in our office to to six startups in six months, And then, like, shameless plug, tomorrow we have we we're starting this, so we have a a talk series happening in in skills matters where we have the CTO of the Obama campaign of 2012 coming by to London to just talk about working for the White House. That's a lot of different programs that we have and a lot of different stuff.

And like Rob said, like, once you start doing all of these different things, these these things we didn't start doing these all in one go. Right? But we started to build these over time. The thing is, you know, to really get the value out of it and explain to the rest of your business what you're doing, you need to build the tools to actually be able to get out of this the really the information, the metrics, etcetera, that want to be able to tell the rest of the company that you hello? My mic is That's fine.

That was fine? Okay. Sorry. I'll keep it a bit closer to my mic. So I had the you know, in in 2013, we tied a lot of this stuff together with, you know, sticky tape and glue.

We had our events in Eventbrite. We had a whole bunch of stuff in mail Mailchimp. We kind of, like, tied everything together. We had some stuff here, and we had to pull it in here, and then put it in there. And, like, if if I wanted to get some statistics about, like, okay, how many how many developers have we interacted with in 2014?

It would have taken me a long time to come up with that answer. And that is a problem. So there's a couple of reasons that I want to start building some new stuff for our team. One of them is that I wanted to program. I am a programmer.

I'm a developer. I wanna build stuff. So far, nine days of travel this year according to trip.com. There's probably about 99 new JavaScript MVCs by now this year. You can't stay up to date if you don't have some projects projects to work on, some projects to play with, some projects to try out something new.

But the more important one to me is that as your team grows, which hopefully if your random person team is doing its job well, you know, you'll you'll get the opportunity to grow out a team. You know, you need to be able to do more with that team. You can't like, even if you don't grow the team, you wanna do more next year with the same people as you did this year. So for that to work, you need to start building tools, you know, that allow you to do you know, always eliminate these repetitive tasks and simplify them. Now the way I see that is, you know, we're a startup.

This is our team, 2013. When I started, we were about 11 people. End of last year, we were about 30. That's insane. That's, like, crazy earth.

Or we could grow that even further, obviously. And we're we're not a lot like this. We don't wear suits. We're also not a lot like that, except for Joe, occasionally. We're a bit more like that, asking people.

Actually, this is is us at the end of

Speaker 2: I think it's at the end of

Cristiano Betta: last year. There's only, like, half of the tea. Anyway but for startups, it's important to build side projects. It's important to build these tools that really allow you to kind of simplify the processes inside of your company to do more. I mean, the good examples are companies like Slack, which is a funny story because the guys who were involved with that were also involved in Flickr, which just like Slack is a side project.

It's a side project that was built as a as a tool to help them do what they wanted to do for the game that they wanted to build, and the game didn't quite get where they wanted to go, so they pivoted and changed. They at least had something to come out of that. So for us that is the tools that allow us to do evangelists in bed. So why is that for us? What have we built so far?

We've got a couple of cool things. So in 2014, we launched the The Valhall Hack website and I've realized this is a twenty fifteen site, but it's the same system powering both. We have our full ticketing system, which admittedly only ever breaks in London because it's the only community that, you know, we have about 600 people trying to get 50 tickets. And that's when concurrency really, really, really, really, really kicks in. Barbara's looking at me like, yeah.

I know. I know. It's twice in a row. But it's fun stuff. So this, you know, this is custom built.

It means we can pull in our information ourselves, and we can actually start properly looking at the statistics in one system. So we now have a pretty good database. This is the behind the scenes of the Biohack system. I had to obscure quite a few things because I'm not allowed to share that information, but we can we can go in and we can see what the we can fully customize the entire system, we can customize the events, we can customize how it works, I can see exactly all the tickets, who signed up, what they signed up for, when they did, whether or they've cancelled. Like I said, there's a bit of a CMS element to it where I can fully customize the way, you know, the text that is being shown, which year, this is actually for the year.

Can define new cities and do recurring cities, but the really powerful so, you know, there's a lot of CMS stuff. I can update the the judges and specify they belong to. I can update the, you know, the partners that we work with and which cities they are attending or FAQ. But the really powerful stuff to me is in this kind of stuff. So we have fully customized email system that allows us to just send out emails specifically to specific segments which are completely customized, and it's all localized to the right time zones.

So we don't run into issues where it's like we're sending somebody in Chicago or something at 9AM London time, which which would be messed up. But the nice thing is, like, fully automated, and it's fully integrated in the same system. We can even schedule releases. So we can schedule full on ticket releases when they go out, how many tickets, which emails need to be sent out to it. So something that would, you know, if we would have to do that for every city, because we're doing 14 cities a year.

Like, if you do one of them a year, that's okay. If you do 14 cities a year, you need to start scripting the hell around this to make this this fluid as possible. It's not just a few clicks, and we've got a new release set up for Citi, including all the emails. To do that in, like, Eventbrite with Mailchimp with, like, all the kind of different systems, it would take us half an hour. So it doesn't sound like a lot, but if you think about the amount of emails that we're sending, the amount of stuff that we're posting, the amount of cities that we're doing and every year over and over again,

Speaker 2: this is not

Cristiano Betta: the stuff you want to do. This is not what you signed up for. Right? You signed up to have fun. So that's just an example of how you how we can specify a confirmation reminder.

You might have seen this, this where we ask people to actually confirm whether or they're still coming. This is something I nicked from my days of running BarCamp. We did that and it turns out to be a pretty good metric for seeing if people are attending. Our ratio of confirmed to actually attending is pretty good. So as you can see oh, yeah.

That's actually what I was gonna talk about next. So for London, this is this year. We had three eighty two people sign up, about 30% of them canceled. So when we ask you to confirm, you have two options, confirm or cancel. So about 30% cancels, about 60% confirmed.

So there's still about, in this case, about 10% who did not reply. That's actually quite a high respond rate for us. Normally, we get about 60% that actually responds. So that's either confirmed or canceled. The rest just drops off.

But the interesting one is that we have about eighty eight percent attending, and I I'll come back to that kind of on a global scale in a bit, but actually, like, in general, it's about ninety to a 105% of the confirmed that attends. So this is 8088% of confirmed actually showed up. This is really useful to our team because we run we have a fully dedicated operations team that run our events for us, and they they're always afraid about, you know, you know, what if we get 226 people? Because if we get 226 people and we only have food for 200, then it's going to be a bad event. Right?

Being able to provide them with this information and this prediction model where we can say, no. No. Don't worry. So we're not gonna have 226 people is extremely important because we could look at this also from a year to year basis. So I know that some cities, you know, 105% shows up.

Some cities, 80% show up. So that information really helps us do catering rides, do the production value ride, make sure we don't don't end up overselling or or underselling our events. Full on checking, of course, we need that to be able to actually know who's attending, as Charlie and Jeff actually doing it on iPad. The other thing that we are massively oversimplified this year was the actual sheets of attendee badges. So the first year we got fully customized badges printed which were really awesome, but what we ended up with was about, well as you can see, we ended up with about 50% badges never being picked up.

And these things were expensive, so we started printing on transparent labels, and we just did that using a nice laser printer, which I thought was quite small, but my events team was like, no. That's a huge printer. So this year, we completely went, we did this. We're a lot more portable. So that's like, you know, we keep iterating.

We keep changing things as as we go. We can even print kind of soft badging so we can just specify, like, oh, you know, these are the staff that is actually present this year. You know, there's a lot more in there. Like, it's basically, like there's actually a roulette section in here as well, which for those of you were Biohagolash, you you know what that is for the rest of you or not, you're like, what the hell is that? I can show it at another moment.

And similarly, like, we've started collecting data about the hacks and actually, actually, you you know, know, what what what have they rehearsed. As they rehearsed, we keep track of all the information, like what technology are they using, what partners are they using, so we actually give this information to our partners as well. Other stuff we will clock, like a Twitter wall fully powered by Pusher since last time. Yay. And similarly, we've done a similar thing for blueprints.

So blueprint is where we give away free stuff, free transactions to our startups. It's a very selective program. So as startups apply, we work through our partners, through branch caps list, through accelerator programs and we give this stuff away. They, you know, we need a full on program to kind of be able to manage these relationships and we could have done this in, you know, any of the official, like, really cool, I wanna say Salesforce, but we don't wanna use Salesforce. Sorry, man.

The other stuff we do is we have a very cool tool. Well, it's slow as hell at the moment. I need to figure out why. It's a tool where we can actually keep track of all the events we go to. This looks like a spreadsheet because it used to be in a spreadsheet, but I don't know if you've ever tried to keep very structured information in a spreadsheet together.

What tends to happen is that some people don't know how to work spreadsheets, and they, like, overwrite things by accident so this is a lot more fault tolerant for us and the nice thing is we can export this into an iCal feed or a non iced iCal feed which means I can give my wife a an iCal feed of where I'm going which is extremely useful because I don't if you've ever tried to use the calendar features of Tripit but it's pretty rubbish. Other things I've worked on is Commerce Factory, like we have a whole bunch of demo code. I worked a lot on setting up the kind of coding conventions that we wanted to have around that so that we could reproduce the same kind of, you know, code sample standards across the board, and some other stuff that I manage, the URL Schrodinger, our Sartank applications, some metrics, etcetera, and our code of conduct. By the way, if you want a hackathon, use this one. It's fully customizable.

You can support it. You can create your own page, hack code of conduct. It's pretty cool. So how do we build that? Lots of different tech involved in that.

That's the fun bit. Like, we get to play with a lot of new tech. I get well, I get to play with a lot of new tech. Occasionally, I invite my colleagues to join me in that. So why does this matter?

Why does this matter to you? Right? So, terrific. So, you know, we're You're a a startup, especially if you're if you're one team and you're within your your small group of people and and within your bigger company, but you're just one person running your evangelism team, your startup, you need to treat treat it like that. So you need to be able to show your metrics and share them with your investors, which is for the rest of your company.

Right? So this is our growth in people, but if you look at other numbers, this is Adelaide. Right? We've grown from 10 to 14 cities. We had a 995% more ticket sales the second year as we had the first year, 223% more attendance.

About 115% were confirmed, Of the confirmed, sorry, of the confirmed attendance, which is amazing, and about 65 of those actually presented. We have 934 hacks this year this year that presented. That's a lot the a lot same stuff over and over again. That's 16 about sixteen fifty times recorded. Right?

Actually, those 487 were unique. And the cool stuff now is, like, we can actually pull out information and I can tell you, like, which languages are the most successful, which languages are the most popular. So the most popular 20. Here we go. And, yes, this is zero indexed.

Like, Somewhere in the ruby is there really low. Tap you note. Okay. This is how we increase our events. We went from a pretty ballsy, very low number and we about doubled that every year.

And sure, we had some new hires and we had some new people, but the only reason why we're able to do this is because we can do more with less each year. These are our event breakdown, and I can change that to one of this year. You can see this year we did a lot more meet ups. All in all, I think, you know, start looking, take away what you want from this talk, but I think putting somebody in charge or at least having a plan as a team as to how to, you know, how to be able to collect this data and how to build these tools so the teams do more is pretty important. I would like to invite you tomorrow again for our folks at Codenos and I'm happy to take any questions.

Thank you very much.