Speed Matters: How Eventbrite Transitioned to Online Events w/ Matthew Himelstein


In software development, speed matters. The sooner you can get working software into end users’ hands, the sooner you can iterate and optimize your product.

In this episode, Matthew Himelstein, currently the Head of Product at Wonolo, discusses his time at Eventbrite. Matthew shares how he helped Eventbrite quickly build their virtual experiences platform in 2019 just in time for the pandemic.

We discuss:

  • How Eventbrite quickly pivoted to online events
  • The importance of end-user feedback in the development cycle
  • Why speed matters in development and in decision making

Want to hear more stories from high growth software companies? Subscribe to Application Modernization on Apple Podcasts, Spotify, or check out our website.

Listening on a desktop & can’t see the links? Just search for Application Modernization in your favorite podcast player.

You are listening to application modernization, a show that spotlights the forward thinking leaders of higro software companies. From scaling applications and accelerating time to market to avoiding expensive license and costs, we discuss how you can innovate with new technology and forward thinking processes and save some cash in the process. Let's get into it. Today we're talking with Matthew Himmelstein about building products with speed and agility. Matthew is head of product at Winolo, a company that provides on demand staffing platform that is changing the way businesses find talent for their immediate labor needs. Interestingly, Matthew Pioneered online events at event bright, leading the team responsible for building its online event platform. This came in handy as event bright pivoted its business in March two thousand and twenty, moving from physical events to virtual events. In this episode we'll talk about why speed and agility is important when building products and will speak to Matthew about his journey at event bright, in his role is entrepreneur in residence building out their online event platform. This is a really interesting story that really speaks to wide speed is important. Here we go with our guest, Matthew Himmelstein. A. Matthew, welcome to application modernization. It Ross great to be here. Yeah, really excited to have you on the show today and learn a bit more about you and your journey. Can you tell us a little bit about yourself and you know what you've done so far and what you're currently doing? Awesome. Yeah, so, Matthew Himilston, I'm currently the head of product at a company called will know Lo. Will Know Lo helps power on demand staffing for most of the United States seventy billion dollar industry. Prior to win a Loo, I've actually co founded three companies and been a product leader and others, and you right before when I was actually an entrepreneur in residence. That I could have called like event right and you know, really helped leadership team their focus on my growth, which what I brought to Winelo, and it's kind of growth long really three areas, business growth, professional growth and then personal growth. It's kind of like how the edict I I kind of bring to the team. Awesome. So we spoke before jumping online now about your journey at event bright and I think there's a really a compelling story there. Can you tell us a little bit more about, you know, your role at event bright and and what you did there in your time. Sure. Yeah, so I was a product leader at a coming called a ticket fly and there's a tangent interesting story of ticket fly getting inquired by Pandora and then two years later actually getting sold to vmp right, and I had various product roles from like ticket ticket management entry to actually managing like the Android IOS applications at event right. But eventually that evmp right I joined, internally referred to as the accelerator Pro Room, as an entrepreneur in residence, and the program was specifically designed for the company, that more people the company to focus on constraints of the business that we're not necessarily like immediate like fires that, you know, we should consider putting out, you know, investing into addresses the company like scaled. And the very specific thing that I pitched on is actually two thousand and eighteen identified that one of the constraints for eventp right was that everybody needed to attend to the live event in person and if we focused on something around online events, that would unconstrain the business of the creators being able to like get attendees from all around the world. So I talked to leadership in two thousand and eighteen, you know, got a green lighted and then obviously we went invested in you know, prior to the end of two thousand and nineteen we had launched we're actually seeing some things go up into the right and then, you know, March two thousand and twenty happened and, you know, luckily for evmp right they you know, we had invested in this and it ended up being, you know, online events and to be the primary revenue driver all throughout the summer of two thousand and twenty. Yes, so you got it out just...

...in time before the well changed. I remember sort of following event bright because we've used it internally at our company, and then, you know, I just thought that they were so quick to pivot to online events. But you know, it's interesting to hear this story that you've been working on this prior to help us understand sort of that process and how you able to get this out to the mokeut sort of so quickly. Yes, is a it's a great is a great question. I think what's started as like kind of like a research investigation into what really constrains event right in terms of it's our ability to grow. There are several areas that, like, I could have focused on, but it just became really apparent that live events are subject to, you know, geographic, local restrictions in terms of how most people want to get to the event. So you can, like the business analysis itself, like really surface that in like right away. But then it was trying to understand if eventp rights current competitive advantage of event distribution would actually fold over and extend itself to online events. When I say competitive advantage, one of the reasons that like people use of empright it's like, I mean super simple, like creating publish an event. But once you public the event of Empor it distributes to your event throughout its ecosystem and makes the events incredibly easy to find and Google. And that was one of the things that we tested at are that I tested out servicing my events if they would rank, you know, with Google. And the interesting challenge there is that the reason that event right is so successful with ranking events and Google is because it's all predicated on location, but online events essentially have no location because anywhere in the world, you know, you can go to an online event, and so it was really interesting challenge and that, you know, I worked with one engine you know, one of the engineering teams, to like stand up like a really simple ab test. We ran it for about two weeks and we got some data and like that data and of itself essentially creamlined entire project because we were able to say, Hey, there's a business constraint. We can show the you know, the map of the numbers behind like what does would mean if we changed it, as well as the VENTP rates competitive vantage. We have some data that shows that like, yeah, we can make these events service and being more easily distributed. So that was really under a month of work to kind of like jump start the entire project right. So you looked at the constraint, you looked at your competitive advantage and then stow it. Opportunity took us through what happened. You know, you, you, yeah, yeah, let us tell us all. Yeah, so the project gets screen lighted and then you really entered this next phase of like six months of development to try and like prove out that what you have like works. And randomly the engineering person who is most interested in my project end up being rene massage, happened to be the technical cofounder of that bright wow. I got to work with Renault, I'm then a couple other really smart folks to just kind of, like as leanly as possible, try and improve the online events experience. And so there's some pretty low hanging fruit that, like just using it from a first principle standpoint, you could just say this is not intuitive, like how can we make this better? We sped up development by focusing on only front end components that we already had built. We didn't want to build anything new from scratch. Just like reuse existing things, even if it's not like the best user experience, because it would just get us to market faster. The other thing we did is I started like interviewing, you know, customers who were, or creators as we call them, who are already like using online events on a event right and they were really it was very interesting. We found that the number one web platform being used on them, right again, this is end of two thousand and eighteen, was zoom. And what was interesting about that is a zoom had an integration with paypal to do paid zooms, and the number one use case on a vent right though, was...

...zoom plus paid events, paid ticketing. We dug in and I actually talked to these creators. They all said the same thing. So, like I've used it them right in the past. I wanted to do an online event. I tried their paypal immigration, I didn't like it and I just figured that I could make this work kind of like standalone. And they just used it them right, essentially as an advanced ticketing system and then just emailed the link. So the lowhonging fruit there was like how can we just make this as super simple as an easy to do as possible, and it became very apparent that like an extra step in the online event creation like flow could simply be create like a standalone online event page where like your zoom would be hosted it. And, you know, we made it as flexible as possible so it could be worked with like you know, it was hang outs of the time, but like Google meet or, you know, blue jeans or whatever, because you're supposed to link, and we gave a bunch of tools to the creators to like manage when the links can be distributed, if the reminder emails, all the kind of things that work. Just best practices. But we really focused on understanding the user need and the problem and just bring that to markets as quickly as possible. And you know, I like to talk to the team and say about some of the the the minor details of, you know, managing time and, like you know, sending out their miners, and kept reinforcing the fact that details matter, right, if you're an event creator, right, the the details matter in terms of can I send this on the exact minute that I want to send it? Can I send the reminder like exactly two hours early? And those kinds of like knobs for the end our end user, our creators, became incredibly powerful because, like you know, they could customize their online event experience exactly how they wanted to. Right. That's a couple of nun gets Fasti so that insight that you really should have uncovered by talking to your customers. As you know, the platform they're using, event brought really too to do the pay ticketing. So you got that insight and then you went back to your team and started to developing out, you know, the additional functionality. I think that's a really good piece of advice for some of the leaders that are listening. I mean, what else in that sort of sort of trying to full you know, should lead us be thinking about, like I'm a big fan of listening to customers and iterating. Mean, do you have a specific approach that you like to take when developing? Yeah, I mean it's you got to come at it all all sides right. So, like, obviously they're insights from customers who are you using your product new and unique ways, particularly ways that you hadn't really expected. There's mapping that to the business impact that you know it would you know, it really does. Like at the end of the day, if the attendees can be anywhere in the world than you as a creator. You don't have to market to just, you know, Berkeley California, like you can market to, you know, all the United States. So like once those two things intersect, then it comes down to really understanding the users incentilization, and there's a very clear like incentivization like mapping exercise you can do between, you know, the end user in your business and try and understand like how the value exchange like works and just one of more basic frameworks that are out there that I don't think it's written about enough, but actually picked it up from the cofounder of my space, aberwork comb kind of shared it with me as like a very easy way to understand, just because you us just want something like that's great, like you, as a business, are going to like deliver to them. Put how does that value exchange work? How do you make sure you're getting the most out of it and how you make sure that they're getting what they want? And it's as simple as, you know, signing up. What do you get as the end user? It's like, well, if I register stir than I get, you know, access to the platform. What's the business get? They get your contact details. I mean it's that simple. But as you start applying that framework out to like any new feature, you have a better understanding of how the value exchange worked and it becomes incredibly powerful, like a scale. So...

I could imagine as you're pitching this idea and trying to get some support internally, when you're framing it up like this, you able to get the resources you needed to, you know, get it done. Let's speak a little bit more about the way that you accelerated the development. was there anything that you know you thought you did really well, or anything that you're doing now that you know could be applied to really accelerate the development of these features. Yeah, you know, I think the overall philosophy is that you need to minimize the time to get working software and to end users hands. And it's great to have nice designs and it's great to like have like a you know like a blanky product requirements document, but at the end of the day, particularly when you're developing something like at do zero to one mark, you know, you really need to like get that end user feedback and figuring out, you know, even how to cut one day out of your development cycle is super important. So at of them right, we tried to you know minimize working on like net new things and try to reuse like existing components and pages and styles, and we're able to do that and really, you know, the end in project. I think we sip the one and like three months pretty good. However, one of the things that have m bright that still can be improved to this day, is there, you know, overall release like process, and so vent bright still has like old school style of like you know collect everything together and do like a like a weekly release. I think they're up to like two or three you know releases a week, which totally works. It's actually, you know, a great. You know pattern as long as you're doing things finance scale. However, my new company will know low we actually do CICD, and so, you know, rap constant provements in the constant ployments. And so, like any engineer throughout the stack, can be releasing to production multiple times per day. And as you start to think about it, if I'm just trying to cut one if cutting one day off of my cycle for the vampire project made an impact and I could only release two or three times a week, then in a world where I can release multiple times a day, that shaves me, like you know, almost five days. And so you know, it is really important. It really lets lets you valid things, let you get them in and using hands, get that feedback directly, and I think it's, you know, just one of those subtle keys success that's needed in order to really build out great products so that build out great businesses. HMM, that's interesting. If you were to kind of give some advice, to tell listeners that are maybe in the situation where you were event bright M and they want to get to where you are your current company, like, what advice would you give them to make that leap? Yeah, you know, I think like figure out like from a transition standpoint, like it's pretty easy to figure out where your end goal is and then work backwards. Right. I think the easiest things, if you're on like a web back like it regardless, like mobile APPs. Focusing on CICD for your back end is probably the place to start, because it's lower risk and it's easily like version control. Like if you push up new code on the back end, that has nothing to do with any like front end code, nobody sees it. You don't need to even like word about feature flaking it right. Conversely, on the front end, like anything you push to the front end like needs to be either like tied up directly or behind the feature flag so that it doesn't either break the uiux experience or, you know, potentially like can be hidden, and that just takes additional work, additional you know, services need and like a little bit more structure. So, yeah, to break it up nowhere you want to go and then get started. Awesome. I want to track back to event right now and here a little bit more about the journey. During that time where things should have got a little quirky and you need, you know, the business really needed to move to relying on online events for its revenue stream. One personally, but then also, I can imagine, you're supporting a lot of creators where their business basically went to zero. So that you've done this two...

...trains of thought here. One event right needs to make some money, but then you've also got to really help your creators. He took us through maybe those months there and to if how you approached, you know, getting this to the mock and to make it happen. Sure. Yeah, so, going back to the original like green light of the online events project, which we had code named digital experiences, we had already understood that we that aren't we could help distribute online events through the Seo Competitive Advantage, and so I think it was like the beginning a queue for of two thousand and nineteen we actually launched the overall feature set, and so we, you know, within six months of like you know planning it, to actually like you know, develop me it, we were we were live and you know, obviously there's playing stuff like work on what we got there, but one of the things we're able to do is start ranking these pages, and very specifically pages of groupings of events that were built to rank highly for Google search. So we call them like Seo, optimized pages and these things. In United States, like within three months you could already see the trend of them going up into the right. In terms of the the core Messiers, we would be looking at his sessions from search, which kind of Corre likes like impressions, but then also like click throughs, and then we could tie the clickers to, you know, ticket sales and like we were able to like, you know, look at the analytics and see like what percentage of online event ticket sales for coming from these sessions from search that were organic search sessions, and so that was already happening. You know, it a pretty good, I'd say, like linear rate into queue. For there's a big bump at the end of the year for New Year's like online events. And again, this is two thousand nineteen, the two thousand and twenty New Year online events on New Year's tanked because, as you could imagine, people were going to real like right New Year's New Year's events. So you know, come back into January, like things are starting to like recover, which you would expect from any kind of seas little business. And then what happens? You know, covid starts to perculate. That's really like the end of January. You just kind of like start to get the reports of what's happening in like China, and February it's you know, it's already United States. Remember, business is okay. We start talking internally, you know, at the highest levels, of like what this possibly could mean, you know, and then obviously the being a marsh, happens and everything just, you know, plunches. It was just kind of really interesting because we had percentage of revenue targets for online events that are bent bright and in like a very terrible way, we like blew them out of the water. As you can imagine, online events did like continue to rise in terms of like revenue, but like obviously the rest of fun rights aren't you dropped, and you go look at like they're, you know, q one reporting from from two thousand and twenty. You can just understand, you know, what happened throughout the rest of the summer. But I say, so things are going south. You know, Julia Hearts to see comes over means has in Matthew. This is online events thing, ready to go and say yeah, like shared it with the data. We immediately started talking about partnerships with which they've launched now, so I think I can talk about him. Like deeper partnerships with zoom and Vimeo we identified as the best Webinar and best live stream platforms, like at the time. That just fit the bill and really understand had a really good like pitch to them because out of the gate we could see, we just really understood exactly where event right fit in. Right, like zoom doesn't want to go build an advance ticketing system. There's too much nuance, like refunds, like all these kinds of things that event right like handles and handles well at scale. The same thing for Vimeo. And so, because they didn't have great advanced ticketing monetization engines, they had different kinds of, you know, Sass based, you know monetization. It really fit well. And so those partnerships launched from a PR standpoint almost immediately and then actually, you know, over time of them right ended up investing in deeper integrations to make it even more seamless. And there's no way that I can say, like I knew that covid was coming. But again, going back to the fundamentals, right, like we're able...

...to identify that attending five events in person. It was a constraint to the business and created business ricks. So how do you alleviate that risk? It just fortuitous that like covid actually like triggered like that risk and that scale for them. Right. Yeah, looks such a fascinating story and you know, I was just thinking about the time he's you know, the timing was perfect. You had enough time in between you've sort of launching it, you know, from the Seo up to my page. perspect dips. You had some of that ranking already, you know, to go. So, you know, I understand the Sei, but maybe for the listeners, like what's an example of a SEO optimized page is? It's something like Coptail vestul events, a set last and nice, popular along right. Yeah, so I think there's a couple of factors that kind of like like make things like I see it. What when we say Seo optimize, where we meet? So like there's actually like the page itself, right, the the creator page, where like people go to buy the tickets and making sure that like it loads fast and you know, it has like good seo, you know, visibility in terms of the headline, all the tagging. I think you could do a google search on like what that means and then really come up with some like really great frameworks. I think Canada actually comes to mind as having like a good, publicly available like style guide for for what you want to do. So there's that like out of like a foundational level. But then one the vmpright does is it takes the IT takes these internal search traffic and then looks at that from like a data pipeline perspective and uses the into the search traffic like to figure out which pages are getting the most traction within the system and then tells that back to Google and says, Hey, index this search result page so that, like we think it's good based on this search criteria, and then we auto tag that and send the back to Google, which, if we're at the scale that of them, rights that if we're seeing that, you know, internally, things are have good quality, Aka getting clicked into and tickets bought, when we put them back to Google, like for the most part we're right and we have this self reinforcing loop of more and more event right pages like are constantly raking. Actually you run into a high end problem of Google only let you rank for so many different topics that like them bright is capped. That's how that's how effective that pattern is for for wow, interesting. So we've heard about the event brught story. We've talked a few about a few short of lessons for the audience. Is there anything else or any other advice you have for you know, lead is at these high growth software companies that you really want to share today? Yeah, I mean I think, like you know, going back to the theme here, is it speed matters and it's you know, a lot of this story is about like development speed, but the other like leadership speed. You need is the speed to make a decision. And lucky enough at event right. But the leadership, you know, came in hat like when I pitched them, like they had plenty of questions, like hey, like our whole business around in person events, like right, you know, this is kind of like totally antipical and typical to it. But to their credit, you know, they said, okay, over the contraing prone of view, like yeah, like let's do this, and what's the harm of having, you know, one entrepreneur residents like go poke at this for like three months and see if there's anything that there. And so, you know, they made the decision rather quickly and I've been at other companies that I won't name more like yet lengthy, you know times to just try and make like very simple decisions. And you know, at leaders, at the leadership level, like I think what people forget is that there's the cost of making the decision and the other aspect to look at is the cost of on making the decision and at the cost of unmaking descision is relatively small and it doesn't really matter like who's right or wrong. If you want to go left or right, just go just whoever has the strongest opinion. Follow them. Let's go left, and if...

...they're wrong, will then UN make the decision and go right. And as long as you understand what you're looking for with the metric space approach of like how to valuate if left is the correct, you know, decision, then it's pretty easy to like unmake the decision as quickly as possible once you have the data. And I think that's gets lost a lot because, you know, people fall on their sword or they need more you know, I need more information because I always want to be like make the right decision. And just going back to the remember I story, if you don't listen to your users and like listen to the data, like, you're just never going to be successful. And to be successful at scale means to constantly be iterating and will constantly looking and trying to have forethought into like what possibly could go wrong and to be ahead of that. Yeah, yeah, I'm a big fan of using facts and data. This is opinions. You know, when should bring in some of these ideas to the table. I like to tell the leadership sometimes, like I'm a fan of math, math, math, right, right, yeah, you know, I think sounds like you should have had the daughter. You had the math, you had it in place and that was, you know, really helped make that decision recently, quickly. Awesome. Hey, well, matthew, it's been a pleasure having you on the show here. I think your story here is super fascinating. Again, I'm the timing and the speed is really essential. Thanks for joining us. Awesome. Ross will, thank you so much. I've been listening to show since you guys put it out and I'm excited to listen to all the rest, although if I listen to this episode I'll probably cringe at the sound, but I really appreciated addy. Thanks, Matthew Jared. Application modernization is sponsored by Red Hat, the world's leading provider of enterprise open source solutions, including high performing Linux, cloud, container and COUBERNETTI's technologies. Thanks for listening to application modernization, a podcast for high growth software companies. Don't forget to subscribe to the show on your favorite podcast player so you never miss an episode, and if you use apple podcasts, do us a favor and leave a quick rating by tapping the stars. Join US on the next episode to learn more about modernizing your infrastructure and applications for growth. Until next time,.

In-Stream Audio Search


Search across all episodes within this podcast

Episodes (28)