Eventbrite’s Brian Rothenberg on growing a marketplace
Five years ago Brian Rothenberg joined Eventbrite to start its cross-functional growth team.
https://art19.com/shows/inside-intercom/episodes/c9e87463-f12a-437d-bbc6-180118fb940dFast forward to today, and Eventbrite is the world’s largest self-service event ticketing and registration platform – processing more than 3 million tickets each week. Going back a bit, Brian was a co-founder at SkillSlate, a local services marketplaces similar to Thumbtack. That company would go on to be acquired by TaskRabbit, where he helped 3x core business volume as co-head of marketing.
I hosted Brian on our podcast to learn why a company’s mission and growth are inextricably linked, how he built momentum for Eventbrite’s growth team in its earliest days, the nuances of growth for marketplaces, and more. What follows is a lightly edited transcript of our chat.
This is part of an ongoing series of interviews about unlocking the potential of growth. If you enjoy the conversation and don’t want to miss the rest of the series, check out more episodes of our podcast. You can subscribe to it on iTunes or Spotify or grab the RSS feed in your player of choice.
Adam: Brian, welcome to Inside Intercom. Can you give us a brief rundown of your career to date and a feel for what you’re doing today at Eventbrite?
Brian: My career actually started back when I was in college, coincidentally in the event ticketing space. I started my first company, Eventticketsnow.com, when I was a freshman out of my dorm room. I was one of the early power sellers on eBay and built a business there, but realized I didn’t want to be a ticket scalper my whole life.
I went to Yahoo in the early 2000s, when Yahoo was synonymous with the internet. I joined as a junior product manager and focused on their local marketplaces and e-commerce businesses. I was a consumer-facing PM and actually doing a lot of growth stuff before it was “growth hacking” – things like SEO, conversion rate optimization and lead generation. I was on a small web property that didn’t have a lot of funding and no marketing, and we had to figure out how to grow it, so that was a great experience.
Then I went back to business school thinking that I wanted to become a venture capitalist. I thought it was this sexy thing and would be a lot of fun. I did a short internship at a VC fund called Canaan Partners, learned a lot, but realized I have more fun on the operating side and can come back to VC later down the road.
At that VC fund I met my co-founder of my next company, SkillSlate. That was a local services marketplace focused on helping people hire small businesses rather than local services companies. It was very similar to Thumbtack; we started about the same time. We raised a $1.5-million round of venture funding from First Round Capital and others, scaled the team, got somewhat of product/market fit, but ended up selling early to TaskRabbit. Upon joining that team, I led online marketing and user acquisition growth for about a year.
When Eventbrite came knocking, it was just too good to turn down. I had the opportunity to join Eventbrite as their first head of growth and helped build out the first cross-functional growth team. I’ve held a number of roles here, ranging from interim CMO to head of growth to GM of our entire self-service business and product. I’m still doing growth stuff here today and having a lot of fun.
Adam: What was Eventbrite like when you joined back in January 2013 compared to today? How has it changed?
Brian: It’s grown a tremendous amount. There were roughly 150 people when I first joined. We’re now at more than 700. We’ve seen a rapid growth in our international expansion. We have 11 offices across four continents. Much of the team is outside of the United States.
We’ve also expanded beyond just focus on the event creator side of the marketplace to more of the demand side as well – how do we drive more ticket buyers and create the virtuous cycle between both sides of the marketplace? It’s been a lot of fun.
A tale of two marketplaces: SkillSlate vs. Thumbtack
Adam: You mentioned your startup SkillSlate, which was acquired by TaskRabbit. That was a success, but Thumbtack reached that next level of growth. Looking back, what were the levers they may have pulled that your team didn’t? What lessons still stick with you from your experience there?
Brian: First of all, I have a lot of admiration for Thumbtack’s founders. We’ve been talking for close to 10 years. One of the key differences was that SkillSlate took a local approach, focused on New York city only, whereas Thumbtack took a broad, national approach in acquiring supply.
There are trade-offs between the two, and clearly their method won out. I can’t say that’s the root cause, but they also they did a good job of building tools for the supply side to engage the supply before they had robust demand. Things like tools to manage their business more effectively and an easy-to-use “publish my listing to Craigslist” functionality, which created a lot of utility before they had a ton of people coming to find local services in their market.
That made a huge strategic difference, because they were able to engage their supply until they could build the demand, which typically takes a lot longer. They created a path that became a runway that was much longer than ours when we were trying to balance supply and demand all at the same time.
Adam: That Craigslist example is a great example of meeting your audience where they already are.
Brian: Exactly, plugging into liquid, existing networks, rather than solely trying to create their own. Another learning was that we almost tried to over-engineer the solution to how we acquired supply. We create this robust web crawler. We used Mechanical Turk by Amazon to extract data from these Craigslist-type postings to build our own local services listings and profiles on our site. Thumbtack used a much more simple but still scalable solution. They also crawled the web and used technology, but they actually built a 300-person team in the Philippines to help structure a lot of data and invite these service providers to join their network.
Thumbtack was able to engage their supply until they could build the demand.
While it seems like a much less scalable approach than an automated program, it was effective. They didn’t have to invest as much time in the technology and it was effective in the long run.
Adam: Did Thumbtack also reach product/market fit faster?
Brian: They definitely [found it] more so than SkillSlate. Then, comparing them to TaskRabbit too because I had some insight there, Thumbtack is effectively ServiceMagic 2.0 (now Homeadvisor). It’s very much, “Enter the request you need and we’ll send that out to a pool of service providers, and they’ll bid on your job.” It’s the same model, but they effectively just built a better mousetrap in that regard.
They’ve of course innovated in many areas since, but TaskRabbit was effectively trying to create this peer-to-peer local sharing economy model, which was new and different, and it just didn’t get to pure product/market fit quite as quickly.
Getting growth off the ground at Eventbrite
Adam: Shifting to when you joined Eventbrite, take us back to those early days. You co-founded the cross-functional growth team. What was it about the company’s position at that time that made decide to deliberately invest in growth?
Brian: The founding team stayed small for a number of years, so they didn’t raise a lot of capital. It was our three co-founders building the product, and they really worked on building a product that met the market’s needs, ensuring they were getting product/market fit.
Fortunately, the product had a lot of built-in levers that helped it grow organically, which carried them for a very long time. Virality was a huge one. People loved the product, so it spread via word of mouth. Another built-in viral component is that when an event creator comes on the platform they inherently market their event to attendees, so that’s classic member to non-member virality. It spread that way, which then fueled another viral loop. A number of Eventbrite’s event creators first learned about it by buying a ticket. Then they had this aha moment of, “I’m organizing events too. I can use this simple tool.” These things were all happening organically, and it was helping the business grow.
When Eventbrite then raised a big round of capital, it was really about creating the processing systems to scale. Not only to do so in the United States, but to scale globally as well.
Adam: That aha moment of “I’ve been buying tickets through this marketplace, and I can actually use it to organize my own events too”, what was your team able to do to help get them to realize that? Or was this happening through virality itself?
Brian: Interestingly, it was happening organically. I was able to actually spot that during my interview process. There was one other person inside the company already digging into this lever, so we connected and really hit it off. When I joined formally we teamed up and started to form that initial growth team.
That was the first area that we focused on. We looked into the data to see if this was happening organically, and the answer was yes. We had a number of questions around whether we could actually move that conversion rate up of starting out as a ticket buyer, converting into event creator. We ran a number of qualitative surveys to understand, for those that had converted, what was the aha moment? When did they first understand? Then, with that information, we started running A/B tests in the product to see if we could actually increase those rates of conversion.
For example, on the event listing pages where people would buy tickets we just started increasing the Eventbrite logo size so that people would see it more frequently and understand, “Oh, this is Eventbrite.” We also added a call to action to find events or create an event. That really showcased the two-sided dynamic of the marketplace. A lot of people starting clicking on “Create an event” to find out what that meant. Those are just examples, but there were a number of others. We shipped those tests and we saw that the numbers were increasing. We realized we were onto something.
Adam: How did that challenge evolve as you moved upmarket? At a smaller, local scale you’re offering something that’s totally new, but as you try to get larger organizers into the marketplace they have to switch from an existing solution, right?
Brian: Yeah, Eventbrite started out really as a new solution for an underserved part of the market, so there were small event creators that didn’t have any online ticketing at the time. That’s creating a market, but it’s very different when you’re trying to capture market from existing players that are higher up in the marketplace.
That was part of the thesis for why I formed the growth team, how do we go upmarket? I helped build the first lead gen programs as we were scaling our sales team to help us move upmarket, all of our marketing automation services, really the B2B side of marketing. The early days of Eventbrite were much more B2C, so a lot of the classic B2B growth tactics applied for us as well. One of the things that I love about this job is it’s a mix of both. It’s really core B2C growth tactics, and also some of the B2B.
Building a cross-functional team
Adam: Often we’ll see startups build their growth team through a single function, like marketing. You’ve cautioned people against this. Why? What’s the attraction and what’s the risk?
Brian: I don’t think there’s a perfect, prescriptive solution. A lot of it depends on the company, their openness to cross-functional teams, and also having people in the seats who have the skill sets for the job.
When growth is run through a single function, functional leaders can sometimes be protective of their domains. They want to protect their own kingdoms and own their areas, rather than form these cross-functional teams where ownership is less clear. It’s like, “Can I get this credit even though there are other people from other functions working on it?” You like to think that doesn’t happen, but in practice it really does.
Instead of asking, “What do I own?”, ask, “What do I owe to the business? How can I service the business?” Also, inertia’s really powerful. When it’s always the way it’s been done we think we should keep doing it. People don’t fear change, they fear loss, and the loss of that ownership. It’s important to have empathy and talk through it, but also to illustrate the upside of trying this.
I’ve also seen companies try to create this cross-functional growth team, and for whatever reason, maybe a prior incarnation was not successful, they’re really soured on future attempts. Was it not the right time in the product’s evolution? Did they not have the right people? The right structure? It’s hard to know.
If you really want to do this right spend a little bit in advisory shares. Bring on an advisor who’s done this before to help coach the team, give some executive air cover and help get it off the ground. That really improves chances of success.
Adam: What was the original makeup of the team that you built? Did you have dedicated resources from design, product, etc? Did you have an analyst right away?
Brian: It was cobbled together. Kevin Hartz, our CEO at the time, was very bought into the concept. I initially reported into our head of marketing, but I made sure through the interview process that the head of engineering and the head of product were bought into the concept. As I got in, it turned out one of those people was not quite as bought in, but still we persisted.
Adam: How did you get them him past that hurdle?
Brian: We said, “Let’s set this up as a six-month test.” We highlighted this attendee-to-organizer viral loop that we thought was an opportunity. We highlighted our key hypotheses and how we wanted to test them. We said, “Give us three engineers, one part-time product manager and myself, and we’ll test and we’ll prove it out.”
I was pulling a lot of the analytics to start and pulling in our marketing analyst to help. We started to show these successive wins and how it was really driving the business. Through that you start to form a snowball.
One tactic that I like is to create cross-functional learning sessions on our experiments. Show (the experiments), get people to vote on what they think the winners were, and then show them the results. You start to teach. People will guess wrong almost half the time, so you have to test. You have to experiment. A lot of it was about ingraining the growth and testing DNA into the company, and, again, positioning it as a test. If it works out, scale it up.
Starting with activation
Adam: How did your team know where to start? You mentioned the viral loop, so it sounds like you went the activation route.
Brian: It’s not that I believe you solve activation first. I look for light and heat that’s already existing. There has to be signal of a strength, and something that’s unique to the product or service is typically best. Fortunately, Eventbrite had so many of those levers.
Not all companies are equal. Some just have inherent advantages. I’d much rather apply our growth tactics to a company that has these built-in levers than start fresh.
If you are starting fresh and you can’t see where this light and heat is you have to think about where you are in the evolution. If you’re more early-stage you’re probably not going to want to focus on retention, assuming you have core product/market fit. That’s just table stakes. Don’t do growth until you have a baseline of retention.
Let’s assume you have product/market fit and baseline retention. You don’t have a huge pool of customers, so focusing on retention early may not be as impactful as moving up the funnel a little bit more and focusing on top-of-funnel conversion and activation and onboarding. Typically, I’ve seen a strong correlation between effective onboarding and retention, so you can get two birds with one stone on the activation piece.
Successful SaaS onboarding drives more than just acquisition, lifting both new and recurring revenue. For example:
+ reduced support contacts
+ higher NPS
+ virality
+ share of wallet
+ retention— Brian Rothenberg (@bmrothenberg) November 29, 2017
Adam: Going back to the concept of buy-in, is there tension between balancing the need for finding quick wins versus more strategic, long-term plans?
Brian: It’s definitely a balance. If you are on the all-strategic, perfect analytic solution, and that’s going to take two years, you’ll never get the program off the ground. If you’re doing a bunch of micro optimizations, you run into the problem of, so what? If you went from a green to a blue button and got one-tenth of a percent increase in conversion, it literally doesn’t matter.
You have to take it step-by-step. Get some quick wins, invest in some of infrastructure, get some more wins, some bigger wins, and go back to the infrastructure again.
The importance of mission-driven growth
Adam: You gave a really great talk at GrowthHackers, all about mission-driven growth. Essentially your message was, “If you don’t deeply know why you’re growing you’re really closer to not growing at all.” What does this idea look like in practice, and what happens to a startup that grows for the wrong reasons?
Brian: The talk was about how a company’s mission and growth, and the strategy involved in both, are inextricably tied. The example that I provided is people will build the right things if they understand what they’re building towards and how they are uniquely positioned to do it.
I’ll share an example. Eventbrite was all about open accessibility – reach the masses, empower the masses, and democratize ticketing. These are terms that our founders would frequently talk about. People understood that this is core to our ethos.
The product was initially launched to be only for paid events, but some of the early users were actually putting zero dollars into the price field for tickets. Some companies would have viewed that as a loophole. They’d have closed that loophole and said, “We’re just going to monetize.” Our founders said, “No, this is happening. It’s organic, it’s open, and it matches our mission,” and they let it happen. That ended up building one of the biggest growth levers for the company, which is freemium.
A huge percentage of our tickets sold are actually free tickets, but that drives trial. A lot of those initially free event creators convert over to paid. It’s a huge lever for us. That never would’ve come about if it wasn’t tied to the company’s mission.
Adam: It reminds me of the Simon Sinek idea of “start with why”.
Brian: I literally pulled in the Simon Sinek talk into that conversation. Related to your question around a company not growing if it doesn’t understand why, I’ve seen it a number of times. TaskRabbit was growing like a weed for a while, but there were some top-line growth things that were masking underlying problems. It’s led me to believe that if you don’t know why your product or service is growing you’re just one step away from slowing down. It will inevitably happen.
I like to view a business as an onion. You’re constantly peeling back the layers to more deeply understand it. Even if you deeply understand your business, you have to always keep trying. Then, if you feel like you have your business down, you have to be peeling back the layers of your customers’ motivations. They’re so tied together.
The nuances of horizontal marketplaces
Adam: Eventbrite is a horizontal platform. Speaking to that, what unique challenges has your team faced that someone in growth at a different type of product company might not have to account for?
Customers are often faced with a paradox of choice.
Brian: I love horizontal platforms, but they can both be a blessing and a curse. They can be great when you’re early to market. For Eventbrite, it was online ticketing long-tail events. That didn’t really exist. For TaskRabbit, it was local peer-to-peer services. They were a pioneer of the sharing economy.
One of the challenges is that customers are often faced with a paradox of choice. There are too many options on the platform. At TaskRabbit people would come [to the product] and say things like, “Great idea. I can’t believe I didn’t think about this,” and then bounce, because they didn’t have a core burning use case in mind.
The outcome of that was if you look an old TaskRabbit homepage it had grocery delivery and regular deliveries, housecleaning, and a bunch of other categories, that all these other startups picked off. They literally built billion-dollar companies, like Instacart for grocery delivery, because they had a better vertical solution.
Adam: This conjures that famous Craigslist graphic showing Airbnb, Etsy, StubHub and all these other companies built out from it.
Brian: It’s the same exact thing, and that’s a real challenge. One way to combat that is horizontal marketplaces verticalize over time. Sticking with the Thumbtack example, if you go and request a dog walker on Thumbtack, you get a very tailored experience. It’s asking you specific questions, like how big is your dog? Do they need to be fed? It makes it feel like a unique native experience.
Eventbrite also has verticalized over time. We’ve taken vertical go-to-market approaches. If you look at any of our recent acquisitions, we’ve been buying players in the music space, like Ticketfly. We continue to build vertical solutions for music, and we’ll probably do so for other categories. You have this horizontal platform that serves as a great foundation, and you build verticals on top of it.
Adam: Another thing your team has done well is localized over time. Eventbrite is now in 180 countries and four continents. How have you approached growing these marketplaces abroad? Is there such thing as localized product/market fit?
Brian: Gathering in person is really a universal human need. It’s a global job, and organically we’ve grown and sold tickets in 180 countries. We are active in more than 18 markets around the world, and we have offices in 11 countries on four continents.
The approach that we’ve taken is that it’s first about adapting the product and achieving local/product market fit, because you can have great product market fit in the U.S. and really fail miserably if you’re lacking, say, a key payments piece in a local market, or the localization isn’t great. That’s where we start, and it comes from strong local understanding and in-market customer insights. Teams go on the ground, understand the local customers, and figure out what we need from the start.
We then tend to take a long view. It’s generally about building supply of events in a local market before you layer on the demand side, because you really have to have products on the shelves before you welcome in the shoppers. Otherwise, they bounce and they’re gone forever, so that’s really important.
Then, for us, we found that a mix of central platform-wide benefits, like we can run a global A/B test in our onboarding flow that benefits all markets. We segment to see how is it performing across markets, but usually, more often than not, you see a benefit across markets. That’s one aspect, but still you can’t get all the benefit unless you have typically at least a small in-market team for local awareness, positioning of the product, evangelism through events, etc.
Adam: As someone who’s worked in horizontal marketplaces for so long, have we hit a saturation point? Is there still room to build in this way?
Brian: I don’t think that there’s a lot of room for more horizontal marketplaces. We’re 20 plus years into the history of the internet. A lot of rocks have been overturned. It doesn’t mean that new ones won’t start up, but I think the next big marketplaces will start in what looks like a niche and expand from there.
Using at Airbnb as an example, when I first saw it in 2009, it looked so niche. It was air mattresses in shared spaces. I think a lot of investors probably overlooked it, thinking, “Is that really a market?” But people adopted it. Then, they expanded their market to more of the vacation rental-by-owner space. Now they’re going up against hotel chains and experiences and all sorts of things. It started looking like a small market and actually has expanded into an incredibly huge market.
Brian’s quick growth tips
Adam: To wrap up, I’ve got a few rapid-fire questions for you that we’re asking all our growth guests. Number one, what’s your favorite growth tactic that you think is also underused?
Brian: It’s classic marketing, but segmentation pricing and product packaging is incredibly powerful, and we’ve had a lot of success at Eventbrite there.
Adam: One book that’s most influenced your thinking, and why?
Brian: A book I’m rereading right now, The Score Takes Care of Itself by Bill Walsh, the great 49ers head coach. He turned one of the most losing teams into a dynasty football program. It’s based on my belief that effective growth programs that scale require effective leadership with all the cross-functional coordination and good coaching of a team.
“Conventional wisdom leads to conventional results.”
– Bill Walsh, from “The Score Takes Care Of Itself”
— Brian Rothenberg (@bmrothenberg) March 15, 2018
There are a few keys that have stuck with me from that book. One is about focusing on process, which produces results, not on the actual results themselves. Another is that failure is key. You have to test and fail in order to find those eventual wins.
He also says that conventional wisdom leads to conventional results. You have to try new things, otherwise you’re just going to get the same thing as everyone else. Then, in an organization, it’s not just important for individuals to know their own role, but they should be aware of the roles of other people as well, and how they’re all interconnected. In a very cross-functional area like growth, that’s more important than ever.
Adam: Your favorite recent onboarding experience, and why?
Brian: Unfortunately I lost my iPhone X this week. I got a new one, and the upside was it’s a delightful experience unwrapping the whole thing, reactivating it, having all of your data sync from iCloud. It’s a totally easy experience. They’ve really nailed that whole package, unboxing and activation flow.
Adam: One app or tool that you can’t live without these days, other than your iPhone X?
Brian: SimilarWeb is pretty awesome. It’s great for understanding competitive dynamics, breaking down competitor analytics, and really reverse engineering some growth tactics that you can try yourself.
Adam: Speaking of growth tactics, the emergence of messengers: are they a game-changing new channel for growth? Or too soon to tell?
Brian: I haven’t spent a ton of time here, so I can’t say definitely, but it’s clearly where people are spending a ton of time. There’s a social graph involved with that, so there’s something there, but the question I ultimately have is how open will those platforms be? How open will they be at scale?
Look at Facebook. They had a very open apps network and apps ecosystem, which Zynga built a whole multi-billion-dollar business on. Then, Facebook shut them off and Zynga crashed. That’s the long-term question in my mind.
Adam: Last question. What’s a common mistake that you’ve seen growth teams make when it comes to running experiments?
Brian: One I alluded to earlier, running tests without enough leverage or impact to really matter, where a winning test is still a “so what” result. Very related is when you find an area that’s really high-leverage and you’re testing repeatedly in that area, but after running a number of experiments you start to see substantially diminishing marginal returns. Teams will sometimes stay there for too long before jumping to the next thing, thinking, “Oh, we’re going to eek out another big one,” and it just doesn’t happen.
