Blend Developer Experience and Enterprise Sales, with Vercel CEO Guillermo Rauch

Michael Grinich:

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 Work OS, which is a platform that helps developers quickly ship common enterprise features like single sign-off. On this podcast cast, 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 Guillermo Rauch, the founder and CEO of Vercel. For those of you unfamiliar, Vercel is a developer platform that makes it incredibly easy to deploy modern front end applications. They're also the creators of Next JS, which is a very popular JavaScript framework for building applications. Recently, Vercel announced their series C financing raising over a hundred million dollars at north of a billion dollar evaluation. The company has attracted developers from Facebook, McDonald's Uber and many, many other organizations, both large and small. Guillermo, welcome to the podcast.

Guillermo Rauch:

Thanks for having me.

Michael Grinich:

So let's just dive in. So, first of all, give us a quick update on Vercel. Where's the business at today? Where's the team at? What's your current focus?

Guillermo Rauch:

So Vercel is a front-end platform that encompasses the three phases of publishing a website or web application online. Developing, which for us, it starts with our open source framework. Next JS previewing, collaborating with your team. Meaning every time you make a change or you push code, having a URL where you can test out your front end and then shipping to really high performance edge network that automates all the cloud infrastructure for delivering at scale, delivering for lots of traffic or delivering highly dynamic and user personalized server-less workloads. So any website in front of the project, you have, you just import it to Vercel and we handle the rest.

Michael Grinich:

And I noticed a few months ago you had a Next JS conf where you talked about Next Live, the live version of Next JS. Maybe you could talk about that briefly and kind of your current focus as you've been expanding and building this platform.

Guillermo Rauch:

Yeah, in many ways, a lot of what we do today is the same thing we set out to do years ago, which is we want to give developers and their teams real time feedback over the evolution of their projects. So one of the things we did when we launched Next JS is that you would see your changes in a browser in real time on local host without having to press a refresh button. Today, I think in the front of the world, it becomes sort of the default, but for a lot of people, a lot of developers it's kind of like dark magic. Wow. Like I can change a style. I can change a react component. I can add a function, remove it. And then my browser just reflects that change in real time. And we've been working a lot and like reducing the latency of that, like from keystroke to screen in like 10 milliseconds.

And what we've realized in the process is that more and more and more is happening directly inside the browser run time. It's not happening in No JS. It's not happening in local host. It's happening directly inside the browser and browsers have become a lot more powerful. So we saw this incredible opportunity to start moving more and more of the developer process directly inside of web browser, using technologies like service worker and web assembly, such that instead of standing up a developer environment, which can take sometimes your entire day or weeks, just downloading the right software and the right dependencies. You can now hyperlink anybody to editing an Next JS project collaboratively with our team in single digit seconds. So it's very much like turning Next JS into a [inaudible 00:26:42] experience.

So doesn't mean that developers are just going to develop inside the web browser, maybe not, but we all have to make a quick change. We all have to jump in and change copy or peer program in terms of finding a problem with the teammate and debugging it. And this is a niche that I think Next JS live will grade a lot of value for.

Michael Grinich:

So this is pretty groundbreaking. And I can totally imagine startups using this small companies, developers just, signing up and starting to use this, but you also have some huge customers already using both Next JS and also Vercel for hosting huge properties of theirs, like the IBM or Staples or McDonald's of the world, and, have grown to this unicorn status in the last years. What were the timing and kind of events that led that to occur? I mean, I think that's pretty unusual. How did Vercel get these users? Where did that growth come from?

Guillermo Rauch:

Yeah, I definitely think it's unusual because really hard. And there's all this catch two and two problems all over the place, right? Because you start in our case, focused on developers and obviously developers have a hugely high bar for quality and like the tools that they want to implement, but then you have to also jump the hurdle of like, it has to be production ready. It can't just be something that's like really cool in your local tests. And then I think the really difficult part is that now, because Next JS is an open source framework, there has to be a comprehensive developer platform and collaboration workflow around it that companies need to pay for. So how do you meet those requirements? How do you meet that bar of, okay, you build the framework, now you have to build the platform that like you mentioned, companies like Washington Post are relying on day in and day out to build their software. And that's quite a big leap to go from, from a small startup to fill a small startup, but having that kind of client.

Michael Grinich:

When did you start splitting your focus there? I mean, thinking about starting off focused entirely on the developer experience to get those initial users. I'm curious if you can think back on maybe when this other side of the house started building and you started saying, Hey, when you started thinking about the focus for the Washington post side.

Guillermo Rauch:

Yeah. That scaling into bigger comments. For sure. And this is something I haven't really talked about much, but it's a great question. So one example is I used to be really obsessed with the idea that we could have like thousands of signups, right? We're hugely popular, but I started caring a lot about what people were using the technology for. And I wanted to make sure that it wasn't going to be a technology that it was just for a quick test or a side project. I knew that Next JS was designed by grabbing inspiration from the largest sites on the internet. Like we started out with surrendering, which we knew was what google.com was doing for search and what facebook.com was doing for newsfeed. So I didn't want to settle for like, oh, cool, like a small team is going to use it for some tests. I wanted the technology to live up to its eventual goal, which is to serve enterprises.

So we started saying, okay, that's really cool. Like there's a lot of self served interest. We could continue down this path forever. It's growing, it's working. But then I started sort of focusing in more like, okay, let's just start thinking about this early experiments that are happening at this companies. How can we grow them into much more than just a handful of developers being excited? And then the thinking started shifting around, like not just like, okay, let's make things extremely awesome and fast for developers, but how can we arm developers with the features substantiations, materials, documentation, integrations, and workflows that the rest of the company needs to turn that from a small idea or experiment into something that becomes kind of like the lifeline of the company? Like if Vercel gets shut down, the Washington Post shuts down or Airbnb shuts down.

Michael Grinich:

It sounds like enterprise, or really just this team focus for bigger organizations cross-cut through everything in terms of the product, not just bigger dollar amount, but actually fundamentally changed the product itself. How you're thinking about it.

Guillermo Rauch:

Yeah. I mentioned this in the series C announcement. I think what a lot of founders need to be cognizant of when they're creating tools is my tool has a early adopter who's a developer, but that developer has stakeholders and, and users beyond that. So if you just think one layer in you're going to get stuck, most likely you have to think one layer in and then two layers deeper, even kind of imagining those conversations and requirements. And frankly also just asking, right? Like, hey, what are the requirements that you have in your company has for using more of yourself? We sort of kind of pulling down that thread and learning a lot more about what else do we need to add to our platform to meet those requirements?

Michael Grinich:

How did this turn to the commercial side as you started thinking about actually sales and revenue? Because I think what you're mostly describing right now is kind of product changes the use that gets you in and how it spreads within a company. But obviously you need to make money on the other side, and grow revenue. How did you approach that? Did you do it at the same time? Did you do it later?

Guillermo Rauch:

Yeah. There were kind of two sides of the same coin for us because a lot of these requirements were being driven by starting to create our customer success function. Our sales function, we hired our first CRO and we started getting into those conversations and finding the places in which we're lacking, not just documentation, but like the ways that we service our customers and the ways that we help them succeed. So they kind of happen simultaneously and I think they really go together. Like you have to take that very seriously if we want to succeed in that space.

Michael Grinich:

How much of enterprise is a focus for you today in terms of how you think about the customers you're serving or you know, revenue? Is it entirely the focus? Is it 20% of the focus or 20% of the revenue? How do you stack today?

Guillermo Rauch:

I like thinking about, again, both being priorities for us, obviously it's not a great answer because like when everything is a priority, nothing is a priority. But I do think that we have to continue to focus on making the developers succeed. There are strong advocates in those enterprise conversations, right? A lot of high level enterprise executives are turning to their developers were like, Hey, what's going to make you happy? What do you think we need to buy to succeed? On the other hand, we have to recognize that with how viral Next JS is, it's being adopted by lots and lots of enterprises that need help. And sometimes they're not just browsing for software that they can self-serve and install. They're expecting a certain infrastructure to be there, to do business with us. So we're focusing on developing that side of the business as well.

The way that I think about it is I want to capture as much opportunity as possible in a mission driven way. Our goal is to make the web faster, both for developers and end users. And without that enterprise arm, I think we would be leaving a lot of business on the table.

Michael Grinich:

And I think the biggest impact also for something like Vercel and Next JS right?

Guillermo Rauch:

Yeah.

Michael Grinich:

The biggest companies, you can have the biggest impact there as well.

Guillermo Rauch:

Yeah. And the product, right? Like in order for it to be rolled out to as many seats as possible within the company, we have to have that enterprise approach. Otherwise, you might get stuck at just one team within a company.

Michael Grinich:

So earlier this year Work OS started helping Vercel power single sign on inside of Vercel, which were obviously delighted to be working together. Vercel had quite a bit of [inaudible 00:11:28] success before this. I mean a lot of these customers already on board far before actually adding single sign on, which for a lot of folks is kind of the first step for enterprise. How did you signing those enterprise deals, getting those folks on board when maybe you didn't check all the boxes for what they were looking for?

Guillermo Rauch:

I like to think that the demand for the product was so strong that we worked as much as we could with their security teams, with our champions, within those companies to sort of align our roadmaps. And we were very clear like, look, you really need our product. We've made it really secure. We have a lot of the foundations for you to rely on it in production. But we do recognize that the SSO feature is missing. A lot of that was like good phase in this side of the customers like, Hey, look, I'm partnering with work OS this is going to happen. You need our product today. Let's work on this together. And ultimately a lot of conversations about future roadmap end up being like that, which is something I didn't know in the beginning. I was thought Vercel is here to like just offer products from our R and D labs and just like lead the way.

But with a lot of this companies, it's just a close collaboration and it was overdue clearly. Like I think we could have done even more business if we had partnered with work OS earlier, but it's been incredibly well received. We actually went and asked to these folks, what do you think about it? Like, how do you compare it to the onboarding process of all the other SSO integrations that you've done with other tools? And people love it. So I think it lived up to that Vercel quality of even though it's not quote on quote developer experience, it is about that obsession over customers who are going to love our tool. And this is the nice thing about working on these enterprise features is that there's so much room for delight here as well. It's not just all like real time react components. It's also delighting all the other stakeholders. So I've been enjoying that a lot and we are doubling down on a lot of those features and capabilities.

Michael Grinich:

I think this is something that we share as like a value as like thinking about that design and craft and quality, which is typically not present in most enterprise software.

Guillermo Rauch:

Hundred percent.

Michael Grinich:

I want to ask about your sales team also. How is your sales team structured? You know, early on, how did you go about these initial sales? What type of sales folks did you look to hire? You know, thinking about it from the engineer perspective, a lot of engineers don't like sales, they don't like interacting with sales. They don't like talking to sales. How did you go about building sales at Vercel, such that it'd be kind of sharing the same cultural values and yet be able to still grow the business?

Guillermo Rauch:

It started with customer success. So, it started with our support team, recognizing that there were basically different categories of clients with different levels of, let's just say demand/anxiety/tolerance or failure, right? So like we have customers that need to know about every single possible future roadmap direction. They're constantly asking very important detail oriented questions. When they get stuck, they want to have priority, they want to bump the queue. So, because, we are working with the technology that is pretty sophisticated, a lot of that, it starts with the support team and guiding companies into successfully rolling out our technology. The next step from that, which was like hiring our first AEs, a lot of that had to deal with like helping companies even understand the landscape, understand how the different offerings compare, why our product is more beneficial as a long term partner. So starting to sell more of the vision of where we're going to be in the future.

And that entails and relates to what I was saying earlier that we actually succeeded in having amazing early customers that believed in our future ability to add all these features that they were missing. So, that went really well because we sort of rolled it out slowly, such that we could understand, okay, what works and what doesn't work for us. And frankly, that's why we started also like just seeing a lot of our biggest successes land, because like you said, if we wanted to see next in Vercel reach it's ultimate potential that sales motion was extremely important. Some companies will just not even entertain the technology if they cannot talk to sales. So it's a reverse of what you were mentioning, which is like, developers don't want to talk to sales, but the company as a whole wants to talk to sales. They want to understand who they're partnering with. In many cases, they want to understand pricing in a more predictable way than a lot of this server-less usage based pricing models, right? And obviously they want to have a phone call with somebody to like, know who's going to be on the other end when I'm having issues. And when I need to make my next big step.

Michael Grinich:

It seems like some of the stuff you're saying is the things that bring a developer in why they get excited, whether it's usage based pricing or the specific technology piece. It's not actually what the end buyer might want.

Guillermo Rauch:

It's almost the opposite [inaudible 00:16:30] .

Michael Grinich:

How do you reconcile those two? I mean, us as a team that must be really challenging to make steps forward in terms of product, to just get alignment around what you're focused on, why, how you talk about it.

Guillermo Rauch:

Yeah. One way that we found that it matches up really well is that we can have one motion for the more scalable, bigger customer. And I think this is why security and as a so, and features like that have been always such a clean differentiator of what bucket of customer do you fall into. I always reflect on our past as a startup. Now, one of our biggest priorities for the security team out of our sale is corporate and enterprise security. When we're buying our first developer products like used data dog in the beginning and whatnot, we're thinking about corporate security. Now we're bigger, a hundred people plus. Corporate security is one of our top three priorities for our security team. Also, it correlates nicely with like, okay, when we go and select a vendor, we're going to do very ambitious things with them. Especially if we engage into this process of baking it off with like comparing it with their competitors.

And then on the other side, I can imagine, the investment on their side. Oh, let's put our best foot forward. Let's work with Vercel and let's establish a multi-year partnership. So in some ways, things become slower compared to the developer, just grabbing onto whatever product they happen to like that day. But also the impact just goes up dramatically, right? Like when we do partnerships with our providers, we tell them like, look, we're doing like tens of billions of requests a week. So the impact on their side is so much greater. And we're seeing this all the time. Vercel will host the Grammy's in 2022, which is, for us, like we started having our first paid customers just a couple years ago to host a world recognized event. And that happened again through the success of our enterprise motion, because for them they're going into in some ways unknown territory, they used to do all of their initiatives through agencies.

Now they're taking the development side of their company a lot more seriously and they need to choose the right partner. And they're putting a lot at stake. So security to me has always been a very important differentiator. The guarantees around infrastructure as well. When you're a developer, you're more going by the things that solve most of the immediacy of the situation. When you're looking at it from a enterprise angle, you're thinking about what are the guarantees that I get years down the line. So that's another way that we differentiate, but obviously there are a lot more, it's always a challenge of course.

Michael Grinich:

It seems like with a lot of developer platforms that start off being super developer focused developer experience, over time they drift further and further away from that. Sometimes even because of their success. You look at things. I don't know. I love Heroku early days Heroku. And I went to their website the other day and I was like, this has fallen so far from where they started. Do you feel that tension, like you feel like you're getting pulled away from it or there's a natural evolution. I know developer experience is super important to Vercel. So I'm curious if you can talk about that tension, maybe how to keep it a focus or what that feeling is.

Guillermo Rauch:

Yeah. I think it exists in many ways at more secure system, for example, from a product perspective can be a system that interest is more friction. At the same time I think you can solve that because a lot of these customers are buying developer experience as well. They want to give their teams a solution they enjoy using day in and day out. So I think as developer productivity becomes more of a priority for enterprises, the challenge would become smaller and smaller. Companies like Stripe have made developer productivity, sort of their focus of a lot of their go to market. Like almost educating the enterprise world about the value here. So I think that education process can help, but yeah, there is a tension and we see it all the time in products that we use, but I'm confident that we can still retain an excellent DX over time while we have a successful enterprise business.

Michael Grinich:

So a couple more questions before we wrap up. Wanted to ask what's next for Vercel in terms of enterprise? What are you working on right now without giving away your whole future roadmap? What's kind of the next step? Vercel has already taken down a bunch of these logos and a bunch of these customers. What are you working on now for those enterprise users?

Guillermo Rauch:

So, number one for me is security because I think Vercel's position is in a very interesting place. It's the first thing you touch. We're now working with healthcare providers, we're working with FinTech companies, we're working with multimedia companies, lots of E-commerce stores. So we're at that entry point where the customer comes in and there are a lot of things that enterprises need to succeed at scale. When we are in that position that we're working on. On the corporate and identity access management side, we're continuing to invest there, role based access control, more granular controls. I think something that gets underestimated a lot is just like the ability to build a security model about how the system works from the outside perspective. Like if I'm a security team that is adopting Vercel, how can I, it makes sense of the different data flows? How can it make sense of activity?

How can I audit the entire system? How can I develop an awareness of what endpoints exist out there and what kind of data is flowing through the system? So I think that's a great opportunity for us to lead with great solutions rather than try to be reactive to security questionnaires. And I think we're in the right position to execute on that. I think a lot about security on the supply chain attack side of things where comments are always anxious about like what's getting deployed. Role based access control plays a role there as well because folks want to understand what's eventually shipping to their "dot coms" and have the ability to have either more control. And this is where you mentioned BX can be heard. Some companies might be more protective of what makes it online. Wheres the platform was conceived in the idea that everybody can instantly make changes, right?

But I think again, like we can find good ways of reconcile, these two worlds. And I think most of the startups and most internet companies today have a long way to go on this security space. So that'll be a huge priority for the foreseeable future. And what I mentioned about guarantees as well, especially around isolation, the ideal state of Vercel is where you can truly call it your own developer platform. If I hire Vercel, they're going to give me the work OS developer platform where I can control absolutely everything in complete isolation from everybody else. And I think we're slowly but surely making our way there. We're obviously massively multi-tenant from an Ingres point of view, but there are a lot of really awesome ways in which we can isolate workloads and provide more ownership to the enterprise that's higher in Vercel.

Michael Grinich:

That's super exciting. I think we may become customers of that too. Last question for you. So obviously Vercel is still growing. It's still early days in terms of potential is here for you to accomplish, but it made a lot of progress crossing the enterprise chasm and going up market. What do you wish you would've known early on? Like what advice would you give your past self and maybe what surprised you over the last several years as you've been going out market?

Guillermo Rauch:

As an engineer I would say just to be aware of the taxonomy and the schema. I read this really funny tweet that said the other day, if you're selling guitars, don't call your table guitars because in the future, you're also going to sell drums. So call it items, but don't call it items because you're going to sell services. So just call the table sellable things, but then you may change your mind so just call it items. You have one table left. But I do think that thinking early on, and even like listening to podcasts like this, where like you will understand your future requirements. It doesn't mean that you have to obsess about them, but you can lay out some basic infrastructure from the very beginning. I was actually talking to a security expert today. I was talking about this in the context of centralized IDP because that company took security seriously early on, you're in a really nice position now. They can grow from a hundred to 200 to 500 people and not have to like deal with like five different IDP systems.

So I think startups can do a lot there. And I think there's also companies like work OS that can give you a lot of this goods out of the box and try to not go too far down the path of reinventing it yourself. That would be my number one advice. The other one is I think having that awareness of the stakeholders that are out there, just even knowing the titles, interviewing them, even if they're not yet even buying your product, just talk to the person that works on procurement, on security, questionnaires on the CSO. Every stakeholder that you're not that used to interacting with, perhaps. Obviously you could be building products for those people too, but just having that awareness of what questions and concerns they have will give you the right frame to build scalable products in the future.

Michael Grinich:

That sounds like some really Sage hard earned advice from the last few years. So thanks for sharing that. I think with that, we can wrap up Guillermo. Thanks so much for joining us on the podcast. Really appreciate your time and excited to see how Vercel continues to grow over the next few years.

Guillermo Rauch:

Thank you, Michael.

Michael Grinich:

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 work OS blog is full of tons of articles and guides outlining best practices for adding features like single sign on skim 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 Work OS thanks so much for listening and see you next time.