Driving Growth through DevRel

Michael (00:02):

Welcome to Crossing the Enterprise Chasm, a podcast about software startups and their journey moving upmarket to serving enterprise customers. I'm your host, Michael Grinich. I'm the founder of WorkOS, which is a platform that helps developers quickly ship common enterprise features like single sign-on. On this podcast, you'll hear directly from founders, product leaders, and early stage operators who have navigated building great products for enterprise customers. In every episode, you'll find strategies, tactics, and real world advice for ways to make your app enterprise ready and take your business to the next level.

Today I'm joined by Martin Gontovnikas, also known as Gonto, the co-founder and general partner of HyperGrowth Partners. Before HyperGrowth though, Gonto spent seven years at a little company called Auth0. He joined as the sixth hire as a developer advocate and grew with the company to eventually become the Senior Vice President of Marketing and Growth. Gonto saw that company go from starting at an ARR of almost $0 to over nine figures and was personally responsible for leading a team that drove a tremendous amount of that revenue growth for the company. Since founding HyperGrowth in 2021, Gonto has been an advisor to a lot of companies you probably know, like Vercel, Retool, and Ashby. He's had a front row seat to some of the explosive growth in the developer tools market and been instrumental in orchestrating and fostering that growth. Gonto, welcome to the podcast.

‍

Gonto (01:25):

Thank you for inviting me.

‍

Michael (01:27):

So, really excited to dive in. There's a ton of stuff we can talk about today. First, I want to go back in time, go back to when you joined Auth0 in 2014. What did the company look like then, and what did you really see as your mission in that role?

‍

Gonto (01:40):

It was funny because it was a weird conversation. I was actually interviewing with them for a software engineering role even though it wasn't the thing specifically I wanted to do. But I interviewed with them because Guillermo Rauch, the CEO of Vercel told me, "That company's interesting, you should meet with them." Then as I was meeting with them, I ended the relationship with my ex-girlfriend. Then I realized that I didn't want to be a coder anymore. It was like, "Shit. This is not what I want to do." I followed Shane's word, who was a developer advocate, but back then I didn't know what it was. I was like, "I like what this guy is doing. He writes blog posts. He speaks at conferences and stuff like that." Then I looked up what he did, read a book on developer evangelism, and then went back to Auth0 to say, "Hey, I like the product. I love what you guys are building. Big believer in it even more because Guillermo Rauch believes in it, but I don't want to be a software engineer. I want to be a developer advocate."

I was explaining to them what I think I could do. When I was interviewing with them, I had already built two open source frameworks that were popular. One was called Restangular, the other one, Factory Pal. I also leveraged that to get into the position, and that's why I ended up joining. But what I saw in the company that I liked was, number one, Matias Woloski, the CTO and co-founder, was very smart and very product focused. He had a vision of where the product needed to be. I got the vouch from Guillermo Rauch that I admired, and I thought he always sees life of a developer with a different lens. He has a different vision for the future. Third, it was a really good opportunity for me to do developer advocacy in a company that has never done it, and they had the opportunity to do some random shit or some new things that maybe others weren't doing.

‍

Michael (03:36):

You come in, and they haven't hired a developer advocate before. It sounds like you had done some open source stuff, but it was also new to you. What did you find successful with marketing to developers? Were you even thinking about it as marketing? Talk about that initial job.

‍

Gonto (03:50):

In the initial job, I wasn't really thinking it's about marketing when we were doing developer advocacy. For me it was more about, okay, if this product, the objective is to get developers to try it out. In the beginning, the objective was to get more signups. The second stage was, okay, now we have more signups, how can we get these people to activate? But those were the two main things that I cared about and I thought about. The question was, "How can we make this happen?" Back then I was embedded into the Angular community. AngularJS was version 0.6, 0.7. I had a belief that that was going to become the framework for front end.

Because most of the engineers that were using Auth0 were front-end engineers, my bet was if we start speaking in every conference of the world about Angular or authentication, and if we build relationships with the other influencers of Angular, which are either the core committers of the framework or the creators of other frameworks that are related to Angular. And if they talked about us and our blog post on Auth, then we would become the thought leader. We did this bet on becoming known on AngularJS and becoming known on jQuery, which remember, it was a lot of years ago, it was like the most used JavaScript tech at that stage. That was the week of the event. The idea of that was we want to become known there so that when anybody on those two thinks about authentication, they think of us. So, we did that.

We went to the conferences, we built relationships, and then people started to share our content. Since then, Brad Green, Jeff Cross, Ben Lesh, all of the creators or people that were related to AngularJS were sharing content about authentication from our blog. Everybody started to come to ask questions about authentication to us. What I think we did well there is that when somebody asked a question about authentication, we only recommended Auth0 on the cases we really thought it was going to be useful. If they were just starting login with Facebook and login with Twitter, we were like, "No, you shouldn't use Auth0 just for that. But if you need this, this and that, then you should come back." That honesty built trust. That's how we were thinking about it. It was more about how do we create awareness by becoming thought leaders in these different spaces?

‍

Michael (06:14):

How did that evolve? Obviously, Angular today is not as popular. There's so many other frameworks. In the interim there's been a lot of other changes too. Can you talk about as the marketing functions started growing out of that from that high integrity, building trust with developers, what did the next step look like in terms of marketing and how did you have to change?

‍

Gonto (06:34):

From there, the main thing that we saw is that writing content and then being part of the community started to work out. Then we started to do the first hires for marketing, which were reporting to me even though I was just Developer Advocate. What we were hiring in the beginning was engineers that would write. I had a personal belief, I still do, that not all DevRels are the same. Some companies think that DevRels all do the same, but some DevRels are better at speaking, others are better at writing content. We decided at Auth0 to split that role into two. We had content engineers that were only writing content, that were some sort of junior DevRel and then we had DevRel that was more building with the community and working with them. And those two that were different were participating on different technologies.

We started to focus on Angular and then Angular started to grow. Once Angular grew, we were like, okay, Angular already is now the established tech, we need to bet on another one. So we bet on React as the next one when it was small. Did the same shit. We started to build blog posts, started to go to the community, and then we did the same for Next.js and for a few others. I will say we were lucky on picking the winners on who was going to become the front-end tech. We did pick some that didn't end up becoming that popular, like Ember. We did a lot of stuff with Ember.js, and Ember.js disappeared. So we did a few good bets, some were bad. But that was a lot of the strategy in the beginning.

Once that was working, we started to focus on a couple more things. One was how can we improve distribution and how can we get more people to see our content? The two things that we did for that one is actually do SEO. Back then, there wasn't that much SEO work being done, so we started to optimize for specific keywords, for specific things that we knew that people were searching. We did a lot of interviews with developers to understand what do you want to learn about authentication? What things are interesting to you? Why are those things interesting? And back then, Okta was our competitor and Okta was writing content for “what does single sign-on mean? What does refresh token mean?” And what every developer told us is, "I don't give a shit about what single sign-on means. I don't give a shit about authentication." What they told us is if I'm told that I need to implement authentication and I'm blocked, then I'm going to Google it and try to find either an open source framework to do it for me or how to solve the errors.

So then our strategy with SEO was targeting that. We built something like hundreds of landing pages where we tried to implement authentication with Angular, with React, with Next. And every time we came to an error, we built a landing page. This is the error, this is how you solve the error. If you want to use multi factor or some advanced features, you should use Auth0. We also started to sponsor open source frameworks because those were the ones that people were searching. And our main thing was we don't want the banner, we just want in the readme, something that says, if you don't want to build authentication yourself, you should check out Auth0. That's it.

We also sponsored Passport.js and redesigned the website to look more similar to ours, for example. And all of these strategies came from better understanding the habits and trying to double down on how developers think and interact with authentication as a problem.

Last thing that I'll mention is the other thing we found is that a lot of developers were searching like ‘authenticate React with active directory’, ‘authenticate React with Facebook’. And what we did was taking the data from our docs so the content wasn't duplicative, we created 12,000 landing pages. Each of them was a combination of authenticate, some SDK or some technology with some identity provider. We got the code from Docs, so the pages were at least 50% different, which is needed for Google not to consider it duplicate. And each of them was giving us approximately 0.25 signups per month, but that of 12,000 was a shit load of signups. So that was also a successful strategy. That's a bit of how it evolved. And what was interesting is that all of these evolutions meant that we were hiring people for the team to manage each of the things.

But we weren't thinking about, "Oh, we're a marketing team. We need to have a content team, a product marketing, a this, a that." We were just thinking like, okay, these things just worked. Who is going to write now the error pages? Oh, we have the idea and it just worked to create these automatic landing pages, we need engineers who can then optimize that. And that's how the team started to grow up. And I would say that the team was a mix of mostly engineers up to the time we were at $25 million in ARR when we actually started to do more typical marketing as well, so we could continue to double or triple at that stage.

‍

Michael (11:44):

Tell me more about those people that you were hiring, the engineers. Where you  were finding engineers that knew how to write or liked writing, wanted to do that type of work versus writing code, actually, is probably what most engineers think about in terms of writing.

‍

Gonto (11:57):

So we did a couple of things. Number one was we subscribed to all of the weekly magazines, like JavaScript Weekly, Ruby Weekly, Python Weekly, et cetera. We saw what were the blog posts that were being most successful and we contacted the authors there. The other thing that we did is there's people who prefer to have breadth to depth. So part of the argument of if you're going to do Content Engineer is you're going to learn so many new technologies. We want you to write how to do authentication for Rust or for this new version of Next.js that is still undocumented, and I want you to look at the code. So getting people that were excited about their breadth was the other one.

And the last one that I'll say is some people were looking forward to becoming Developer Advocates or DevRel, but they didn't have the experience yet. So because of that, what we were telling them is, ‘this is a tour of duty for you.’ It's like you're going to come here and you're going to work with us on building Content Engineer for two years. After two years, if everything is going well, we'll promote you to become DevRel and then you'll start going to conferences and do a lot of the other stuff as well with training. And those first two years, we also offered them that they could pick one conference or two conferences per year where if they got approved to speak, we would pay the tickets, the hotel, even if they just speak about authentication, as a perk to them. It was like a junior DevRel team for us.

Last thing on this is we also had a guest author program where we paid people to write content. And a lot of the people that we hired for the content engineering team started as guest authors and then actually enjoyed it and then we got them to change to full-time, working on that.

Michael (13:42):

Tell me about when you guys started hiring or maybe you started doing the traditional marketing stuff that maybe most people think about as corporate marketing or marketing more for commercial use. When was that? When did you start doing that? Why did you start doing it?

‍

Gonto (13:57):

Every year we built something that we called the L2R, the Lead to Revenue Model. The Lead to Revenue Model basically had data on how many new anonymous users we have every month on the website, what percentage of them convert to a signup. From the signup, how many of them have a high score. From those, how many convert to an opportunity? How many of them click and talk to sales and connect with us? How many of those convert to self-service? But the whole idea was from visitor up to revenue, whether it was self-service or opportunities, we have data, month over month, on how it was working. And then every year when we were planning for the next year, what we said is, okay, let's assume we're going to continue growing the way we're growing now. Where are we going to hit next year?

And let's say we were at $5 million ARR, we did that math and we saw that we did nothing and just continued doing the same, we were going to hit $7.8 mil in ARR. That meant that we had a gap of two. So the gap of two, the idea was to think about experiments or things that we were going to do or areas and then look at what would be the impact. For example, at one stage we tried Concierge, which was like engineers that were talking to people who were blocked in the dashboard to help them convert. We thought that if that was effective, we were going to increase our conversion there 10%. So then we mapped out increasing 10% conversion from March, up to December, and we looked at how that affected revenue. In the beginning, all of the experiments that we wanted to do were going to be enough to hit the revenue targets that we wanted. However, there was one year when we were approaching $25 million ARR that we knew that with just inbound, just sign up and just doing experiments, we were never going to hit $25 to $50.

At that stage, six months before, we were like, what is the bet we want to do now? And the bet we wanted to do was let's do some traditional marketing. But instead of targeting developers, let's target Directors of Engineering, Directors of Product. We knew that getting that to work was going to take six to nine months. So we started that six months before the year we actually needed that to happen. So with that, we hired Kerry Ok, which then ended up being the SVP of Marketing at Auth0 when I left, and she's still at Okta now, to run this team. And I remember having conversations with her in the beginning that she asked for one or two months heads up if she was going to get fired because everything that we were doing in the beginning in the first couple of months wasn't working out.

But we kept on it and once we became a little bit more creative, some things started to work out. Our first campaign that worked out was a direct mail campaign that she actually packaged all of the things in the office herself where because it was a digital transformation campaign, we were sending transformer toys to people and telling them, "If you want another one for your kid, this other person has it, so we'll change it." And the idea of that was to foster conversations between the people in there to drive demand. We got MassMutual, which is an insurtech company, through that. And that was the start of like, okay, maybe we need to do typical marketing, but we spin. And from there we continue to do direct mail, started to do more ABM, started to do trade shows, started to do webinars and more of the traditional marketing as well.

‍

Michael (17:31):

Doing direct mail and sending someone actually something physical and the mail, that's pretty different than building 12,000 landing pages for technical topics for SEO. It's almost hard to imagine two things that are more different on the marketing function. How was the cultural change around that? It seems like that's a totally different motion, not just slightly different. It's not saying that you were just doing blog posts or something like that, mailing people something. It's actually super, super different. How did the culture change and evolve? Where did you butt heads? I'm sure there's a lot of people listening to this podcast that are either going through that transition or anticipate going through it.

‍

Gonto (18:09):

In the beginning, on marketing, I would say 90% of us were engineers. Engineers who wrote content, engineers who were doing data analysis, engineers that were doing growth, engineers that were doing DevRel. And Kerry was sort of our first non-engineering hire even though she had worked at a technical space. So that was a bit hard in the beginning in the sense that she didn't feel comfortable talking about the product because she was scared that people were like, we're going to look at her bad, like, "Oh, you're not an engineer. What are you doing? What are you talking about?" And she was very cautious because of that.

The other thing that I think was hard in the beginning from a culture perspective was for engineers to accept that we were going to do direct mail or go to trade shows and they were like, "No, engineers are not going to like it. They're going to hate it." And that was a big clash in the beginning, but how we convinced everybody to merge the cultures is like we always talked about the culture of experimentation. The idea of a culture of experimentation is we're going to try something to see if it works and if it works, we'll continue. If it doesn't, we won't.

So why would we experiment and iterate on things that are more content and developer related, but be scared to try something more different than this? And that was the biggest part of the argument, both to the exec team as well as to the rest of the engineers in marketing. It was like, we're going to try something different for a couple of months. If it doesn't work, we're going to kill it, like anything else. But if it does, then it's going to work out. And you all think that it won't work. That's okay. Maybe you learn something or maybe we learn something unique.

And that's where the fear of Kerry being fired came from because I think she felt that things in the beginning weren't working out and people were mostly engineers. So it's like we don't want to do this traditional marketing shit. But what actually helped is that she's very determined as well. So she got the support on we're going to experiment, but she also was obsessed on making it work and trying and trying and trying. And she really liked this experimentation mindset that she didn't have that much before, even though they were testing things and embraced that idea. After it started to work, she also took time to learn about Auth0, how it works and the technical part. And because of that, she built trust with all of the rest because she took the effort to actually learn the technical parts of what it is and how it works.

‍

Michael (20:43):

What advice would you give people that are looking to make that kind of a hire, someone like that, what should they look for in candidates? What's a great fit for that first, maybe more traditional marketing hire, not as developer engineer focused hire?

‍

Gonto (20:58):

I think number one, I wanted somebody that had led a team on this because they could potentially start a team eventually, but that could roll up and actually do the work. I talked to a lot of people that were just managers, and when you ask them about very specific questions on ‘how do you do this?', ‘how do you implement this?’, ‘what were the steps?’, most of them didn't answer. So focusing on specificity of what are the exact things that they did to make it work, gave me a lot of confidence that they really did it and they really had some learnings and they knew how to do it.

The second part was we needed somebody that was number obsessed. There's a lot of people that are doing demand gen and that focus on how many leads they get or how many MQLs. I don't trust those people because a lead is a vanity metric. What I think is very important is to find somebody that is obsessed with the bottom line. So on demand gen, it will be pipeline or revenue. And as we were interviewing, we saw that for the trade shows, the direct mail, the campaigns they've done in the past, their biggest objective was pipeline. And that was a big change for us. When we also dig deeper into the data on, ‘okay, you created pipeline, how do you know it was for this and not for something else? How do you think about it? What attribution model?’ They actually understood about attribution, how to attribute it, what were the risks, what weren't the risks. So even though they were not an engineer, they understood metrics and decomposed a problem into smaller pieces that were measured somehow. That gave us confidence that they thought like an engineer, even though they weren't one.

So that's another one that I think is really, really important. And then the last one that I'll say is we do exercises. At Auth0, we did exercises for every role that joined. And the exercise was you have to do some work. Big believer on that. We had people that we were going to hire that f***ed up the exercise. So doing that I think is key. And in this case, the exercise was: build a complete unique campaign on what we would do and what would be the plan. And that was really helpful, understanding the questions they asked on the Slack channel. How were they thinking about it? Why were they asking those questions? How would they pick the metrics and how were they thinking about that? That was key. We gave a very loose problem because the questions they would ask were more important than the end results. To be honest, we actually hired her without finishing the end results just because we liked a lot the questions and how she was thinking about it.

‍

Michael (23:42):

Gonto, last question for you. You obviously advise a lot of startups now with HyperGrowth. You work with a bunch of growing companies. What advice would you give the next generation of entrepreneurs when thinking about this marketing and growth? And specifically folks that maybe are technical or come from the product side, less of a background in marketing that are just embarking on it?

‍

Gonto (24:03):

Two main things that come top of mind. One is not every developer is the same. Frontend engineer, different to a backend engineer, different to a network engineer, different to DevOps or DevSecOps or platform engineer or SRE. So understanding who exactly it is, is important. And then doing research to understand how they relate to your problem is key. We do a lot of what I call habit research, which is, you interview people in your customer space on your ICP, but they've never heard of you or they have never heard of your product. And the questions are ‘how do you learn about X?’ So for example, with Vercel, we did something similar and it was like, ‘how do you learn about new platforms to develop apps and to deploy apps? Do you want to learn about it or you don't care? What communities are you part of? Who do you follow on Twitter? Do you use Reddit? During the day, do you use any apps? What apps? Do you use more the computer or the phone? How do you connect with those things?’

And I think marketing ideas will come more from that, than from anything else. If you want to target somebody from a PLG perspective, you need to tap into their existing habits. So finding out which ones are those, and then creating strategies for that I think is key. So first step is doing that research to then think about the ideas. And then the second one is once you have to think about those ideas, it's better to be different than better. I'm a big believer of that. If you do the same as what everybody else is doing, but a little bit better, people are not going to remember, people are not going to see it. People are going to remember something that is different, something that is unique. And even if it's a bit shittier, how it's implemented, if it's different and unique, they will remember. So giving some time to think about more different or unique ideas is great.

A lot of times when we advise companies, they were asking us, "What frameworks do you use?" And of course, we have some frameworks that we try to apply to companies. But the reality is that if a framework would make you successful, every company would be successful. Success doesn't come from the frameworks. Success comes from the experience of trying something different and the experience of what worked and didn't in the past. And it's from understanding the ICP and then doing something about it. So not being tied to the frameworks and actually trying to find ideas that are a bit more creative or unique make a difference.

‍

Michael (26:34):

Gonto, thanks so much for joining us today and sharing a bunch of advice and stories about your time at Auth0.

‍

Gonto (26:41):

Thank you for inviting me.

‍

Michael (26:48):

You just listened to Crossing the Enterprise Chasm, a podcast about software startups and their journey moving upmarket to serving enterprise customers. Want to learn more about becoming enterprise ready? The WorkOS blog is full of tons of articles and guides outlining best practices for adding features like single sign-on, SCIM provisioning and more to your app. Also, make sure to subscribe to this podcast so you're first to hear about new episodes with more founders and product leads of fast-growing startups. I'm Michael Grinich, founder of WorkOS. Thanks so much for listening and see you next time.

‍