1. Library
  2. Arrow Icon
  3. Foundations For Repeatable Revenue w/ Figma, Netlify, Apollo GraphQL, Avo
FEB 17, 2022 - 58 MIN

Foundations For Repeatable Revenue w/ Figma, Netlify, Apollo GraphQL, Avo

  • Marketing
  • Sales

In this DevGuild: Dev-First GTM panel, leaders from Figma, Netlify, Apollo GraphQL, and Avo share hard won lessons from their time building repeatable revenue machines.


In this DevGuild: Dev-First GTM panel, leaders from Figma, Netlify, Apollo GraphQL, and Avo share hard won lessons from their time building repeatable revenue machines.

Stefania: Hi everyone. I'm Stef. I'm the CEO and Co-Founder of Avo. Thank you for joining us on Day Two of DevGuild. I am incredibly excited to moderate this panel of powerhouses that have extensive experience with going to market and building repeatable revenue machines, which is the theme of this panel. First of all, we are here to learn and to try to extract as much knowledge as we can from these incredible people. We will focus particularly on how teams with bottom-up approaches to go-to-market have built these repeatable revenue machines.

Today with us we have three people that have that experience from quite different angles. There are definitely themes to it. We already talked about the bottom-up motion, but they are all very unique. Every go-to-market motion is different from the other because you have all monetized bottom-up. Can you tell uswho you are, where you come from, and what your monetization model looks like in practice? We'll start with Amanda, please.

Amanda: Hi Stef. I'm Amanda Kleha and I'm the Chief Customer Officer at Figma. I oversee aspects of our go-to-market strategy, including marketing, self-service, sales, as well as customer support. A little bit about Figma. We are a web-based design platform for teams that build products together, teams like Microsoft, Notion, New York Times, to name a couple people who use us. And they use it for every part of the product development process from brainstorming to design to even build.

So it's an end-to-end platform that you can build on. And our go-to-market motion includes a self-service model where you can come try the product for free. You can upgrade with a credit card on your own, if you wish. And then we also have a sales team that helps sell as well. So we have a freemium self-service and sales motion all together, and they work together. And part of why I was attracted to working at Figma.

Stefania: Excellent. Thank you, Amanda. Over to Chris, please.

Chris: Hi, I'm Chris Bach. I'm Co-Founder of Netlify. It's a platform that web developers use to build, scale, and maintain web projects. And we are here because we saw that there was a new architecture for the web and where you would decouple all the web layers from the business logic on the back end. And that would need a viable workflow to work with that at scale and a way to tie it all together, an orchestration layer if you will. And that's why we built out Netlify.

We had about a million developers last year. Now we passed 2.4. So it's growing really fast. The category is growing really fast as well. And we, SaaS of course, we have a generous free tier. We have self-service plans, and then of course we have a lot of enterprises that run us. We run mission critical apps for like Twilio, a $60 billion market cap company that has the actual core product/the locked in web console is a Netlify application. Sprint just moved over six and a half million store fronts to Netlify, and so on.

What I think is unique about it compared to legacy platforms of the past is that this is a very wide agnostic platform. So rather than be specialized in e-commerce or certain use cases, because of this separation, what we're seeing is that it really caters anything you build for the web, whether it's an eCommerce or a .com or marketing site. So even like kiosks down in the restaurants with all the menus running, there used to be some .net server running on a Windows NC computer over in the corner.

They're now all sort of becoming parts of the same workflows. And that's what Netlify is championing. We're actually working very closely with both Figma and Apollo. And it's a lot of fun to be part of the new web economy.

Stefania: Nice. Thank you for sharing Chris. Over to you, Matt.

Matt: Thanks Stef. I'm Matt DeBergalis, I'm the Co-Founder and CTO at Apollo GraphQL. We're an open source company. So our bottom-up motion is about helping developers adopt a Graph. Instead of the traditional way of connecting clients and servers with point-to-point APIs, GraphQL allows teams to write queries instead of code. And for a developer it's a night and day difference in experience. You can have better tools, you move faster, all the benefits of a modern tech stack.

For a company the benefit is, "how do we ship more software more quickly and how do we adapt to a changing market?" If you're in retail or if you're in media, it's a race to be one of the top companies in your category from a digital experience point of view. And the ability to do that quickly, and to do that with changing market conditions and opportunities, to be able to take teams of developers and rapidly create new data rich customer experiences and get those on the mobile device on the desktop, on the kiosks, all of that is the difference between being really successful in the modern era and being a laggard.

So for us, we've really had to build two things. We've had to build a bottom-up motion that's really fundamentally about developers and open source and this mega trend of moving away from the old way of doing things to the Graph. But we've also had to build a motion around the strategic implications of that for larger companies. If you're shopping at Walmart, you're on Apollo. If you're watching stuff on Netflix, you're using a Graph. If you're reading the New York Times, you're using a Graph.

These are strategic decisions that require a lot of thought about how you take that bottom-up motion and translate it into something that spreads across the company.

And there's a lot inside that.

Stefania: Right. Exactly. Thank you for sharing that. There's so much to unpack in everything in just your intros already. So today we would love to cover, for example, how on track have you stayed with your revenue model, for example. Did you always intend to build a bottom-up company? And I think there's a lot to unpack also, and we'll talk about persona discovery and all that stuff.

I want to just start immediately with this. Did you always intend to build a bottom up company, and why or why not?

Chris: Yes, we always did. I mean, what we saw was that there was a better way to build for the web. The web was in danger of going into walled gardens and losing out to proprietary app stores, because the underlying architecture wasn't good enough. And so we saw that there was a way for the web to be many times faster and more performant, more secure, more scalable, and much more compatible with modern git workflows and the workflows that we're hearing about here, but it would require this viable workflow.

And so Netlify essentially bet everything on a market trend that simply didn't exist. And so the only way for us to succeed is if there's a massive adoption of this, if it becomes as it is becoming now, it'll be synonymous with how you build for the web in the first place. And so it only becomes relevant for us, and we are only able to create a really meaningful business, if we get into the hands of millions.

So it had to be bottom-up. It starts there, it starts with the hobby projects. And then you grow and they take you into enterprises and so on. But yeah, in our case, there was no doubt at any point. It had to be bottom-up.

Stefania: Okay. Awesome. I would love to hear from Matt on this next, because Chris just talked about the trajectory from having the mass and then going into the enterprise and Matt had already touched on that a little bit. Matt, can you share your take on that, your intention regarding bottom-up?

Matt: Yeah. Let me say, it was always going to be bottom-up and I've been a fan of open source since I was a kid. So I think it's the best way to spread the best technology to the most people and to make that technology, particularly infrastructure, the kinds of stuff that benefits from thousands of hands improving it day over day.

And it's 2021, every company has to be a software company. And I think we would all agree that the tools that a software company is going to use are going to be tools that developers pick or the designers or the product managers, but that's the motion.

It's not just about scale. I think it's even maybe more fundamental that it's hard to see a different path, whether it's a cloud-based offering or an open source offering, it's got to be something that developers are going to pick, that it's easy to get started with, that you've got an early win or an "Aha" moment or some kind of sort of magical experience that really clicks for people and makes them want to do more and makes it a sticky experience for us.

A lot of these ideas are they make the most sense when they spread to a market that's essentially every company. And so I think you have to have a strategy for how that's going to work. You can't hire enough sales people to sell it to every single company really, and you need a way to make the economics work, not just as a startup at a small scale, but even at the largest of scales where I think the benefits that bottom-up motion are in some ways the strongest.

Stefania: Excellent. As a follow up after these answers, I want to talk about that sort of motion of turning the bottom-up leads into actual revenue and turning people into paying customers. But Amanda, would love to hear your thoughts on that as well from Figma or other stories, if you have them.

Amanda: Yeah, absolutely. So the founders, when they were first building Figma, wanted to get some feedback from potential users. And what they did was they looked at, "who do we want to have as our users?" And so they made a short list of people that they wanted to go show the product to.

And I'm sure most founders do this, but what I think is unique about the story is that because it was the first browser-based design tool, it was new and people weren't used to not having a native app. And so they got a lot of feedback that was like, "Oh, well, this is interesting, but it needs a lot of other things in it." And so what I like about the story is that they went back and built those things, but then went back to those same potential users and showed them like, "Hey, we listened to you and we built this."

And I think that second part is what a lot of founders maybe get distracted and end up not doing, but they really won over the hearts of some early influencers and had a bottom-up motion in mind the whole time, because

the mission and vision of the company is to make design accessible to all. And with that kind of mission, you can't create a company where it requires a sales team to sell.

So they launched a freemium version of the product at first. And I'm a big believer that if you're going to have self-service and sales, you have to start with self-service because it creates the right DNA in your company to figure out not only what's the product you're building, but what's the experience of understanding and educating your users about that product. And that in itself is its own product to build. And you have to do that upfront. Otherwise it's very difficult to engineer later in your business. So that's what we did at Figma.

Stefania: Awesome. Everyone talks about Slack as the first or as a really strong example of a bottom-up motion, but that applies to everyone in the company using that solution versus all of the companies that are here on this call. It's a different thing. It's a specific persona to solve, a specific work problem that's not just general communications for all the team. 

On this note of identifying how you can turn the mass, the bottom that adopts your product into a buying persona, can you talk a little bit about that journey of when you started realizing that you have some pull? How do you then start turning those people into actual paying customers? Matt, maybe going over to you.

Matt: This question was one of the big struggles and challenges for us. We understood pretty early on, the persona of the developer that was going to adopt Apollo, would have to make the switch to GraphQL. We understood the benefits. We went through a lot of the things that we've talked about, where you watch people using it, you ask questions. The first killer feature of Apollo is that we took poll requests. You get that feedback loop pretty early on. But what we found is that the buyer's very different.

Stefania: The buyer is very different from the adopter?

Matt: Yeah. The instinct I had initially was we would serve the needs of the developer and at some point we'd find some features for which the developers or the developer teams would pay. And it turns out if you're in that conversation, you're having a pretty tactical conversation about specific features that a developer or a manager values.

And the problem isn't that a manager has one credit card limit and a director has another credit card limit, and a VP has another credit card limit. The problem is what the VP wants is completely different, unrelated in a lot of ways to what that manager or the developer might want. And if you go to the VP and you try to have a conversation about the tactical stuff that's important to make that product-led growth in the bottom-up journey work for the developer, it doesn't get you very far.

So what we found is we really needed to understand two personas pretty early-on for this to work. And that second persona is when we think about how we do marketing, for example, there's a whole set of things we do for the buyer persona that connects to what we might do for the developer persona.

The way Apollo works is that as part of my team at Apollo, we have a DevRel team and a Developer Experience team. And that's the group that actually does a lot of the content that's really designed for the developer persona. That team owns the documentation, the blog, a lot of the communication we do at that level. And then we have a marketing team and a lot of the work they do, whether it's case studies or content that's designed for the buyer is really, it's a different value prop and a different message and it's designed to drive a different kind of conversation toward the sale.

For us, once we figured out what that looked like, it made things a lot easier. I'm an engineer by trade. The instincts and the intuition that maybe serves me well from a product design point of view, serve me less well in that go-to-market space, and we found that we had to really look very carefully at that whole journey that ends at the purchase to really understand that.

Stefania: That is awesome. That is immediately sparking questions around team structure, because you talked about two different teams owning that. So I would love to double click on sort of how you've built your go-to-market teams, from person one and then onwards. Amanda. I would like to throw the ball over to you. Where are you in the world? So I know where to throw the ball? 

Amanda: I'm in San Francisco.

Stefania:So over from Iceland to San Francisco. You talked about the user journey and the product-led growth. And I really agree, if you're going to try to make sure that the adoption of the mass of the company is going to be the driver for the big deal, then that big mass within that buying company, they have to have a really pleasant experience.

So can you talk a little bit about your learnings on building that product-led experience? How did you identify the things that were maybe triggers for turning people into paying customers and things like that?

Amanda: I think to Matt's earlier point, today's tools are often chosen by developers, by designers, by product people. And so we're in that space where we knew that we had a bottom- up motion to execute on, but early-on we had enterprise-sized companies wanting to use Figma. And so it's easy to have a single developer sign up and maybe get some colleagues to start using it. But very quickly, if those designers are excited to use Figma broadly at a company, they will start doing some internal selling on behalf of us, which is fantastic, but

you get IT involved and they have some security questions and now you have legal requirements on your go-to-market motion that is not a bottom-up kind of conversation.

So, that's how that happens real quick.

But I think for us, as a design platform, you think it's just designers using it, but the product itself, what's beautiful about it is you can use it really easily with other people. And so in fact, two thirds of our users are people who collaborate with designers, they're developers, they're marketers, they're product managers, they're researchers. And so we have designed the product to make it very seamless to share files, to incorporate other people into your projects.

And as a result, even though the designer is the initial user, it very quickly has other personas using the product and other people involved in the process. So we will often, at large companies, because you no longer have a centralized design team, but rather design teams that are within business units or within product lines, we'll land in one team, grow there, and then we'll expand to other teams.

And when you're working with enterprise-size companies, they want to have the security conversations. They want to have the legal conversations. They want to talk to sales. You might think that people don't want to talk to sales, but they actually do in many cases. And so if you are a company that hires sales people that know how to talk to your personas and users, that's okay. That's not a scary thing to have sales. When I joined three years ago, we already had a couple sales people because we were already selling to enterprise-size companies.

Stefania: Awesome. They want someone to help you navigate the legal and security processes. It's like, "Please, I don't want to do this on my own. Help me." So, can you talk about the data that you might be using to identify those sales opportunities? Are they mostly when the internal teams ask to talk to sales or how do you use qualitative or quantitative data to trigger an up sell motion?

Amanda: We absolutely have a lot of people who fill out a form on our website because they want to talk to sales. So we pass those onto sales right away. But then of the people that sign up for Figma, we do a bunch of segmentation, enriching those leads in all the classic ways you would do that, on the demographic side, on the product usage side. So we determine if some of those might get passed to sales right away, or do we wait and see how they grow a bit before we pass them to sales.

So there's a whole kind of decision tree that happens on the backend to determine that. But it's based on product as well as the demographic profile.

Stefania: I'm excited to double click on that. And also you were touching on that, Matt, in an earlier conversation with account-based sales and things like that. But before that, just quickly over to you, Chris, where are you in the world?

Chris: I'm also in San Francisco.

Stefania: Right. So maybe Amanda can take the ball and just throw it over to you.

Chris: It's not as far a throw as from Iceland.

Stefania: No, it's not as far. So it won't take long for the ball to arrive, but can you talk a little bit about how you turn your bottom-up motion into opportunities and into revenue and what that journey looks like?

Chris: Absolutely. Very much echoing Amanda, a lot of that that applies directly to Netlify as well. Netlify is a very broad platform. We go from local development all the way to a global multicloud edge network. And we'll have tons of code schools using us because you can deploy a web project in seconds. Success is very immediate. But we also have very big enterprises and of course they want to hear about security and compliance and so on.

And I believe in our case, there are different needs, but it's very important to be a way of how one feeds into the other. Otherwise I think it's very hard to have two. Then it's a dual funnel motion. And that to us is a little hard. And so we look at a bunch of different things, but

we've never been interested in trying to monetize the individual developer. That's not the point.

If there was a magic buttons so we never have to do that, we would have pressed it long time ago, but there's different kinds of usage, both in volume and in advanced feature sets and so on. And what kind of usage is it for? Who's the client, the demographics? I mean, there's so many things that plays into this.

I would also say that modern sales, for a while it's not always about closing, but always be helping. If you instrument your product in the right way, these people will graduate to want the help and then they receive it from you. And it's a collaboration and as long as you make sure that it's a fair value exchange, and it continues to be so. And I think that's super important to always strive for. 

We look at a multitude of different things when it comes to knowing when to reach out. And again, just like in Figma's case, a lot of them will reach out to us because that's how the product uses this instrument. And you want to talk about security and compliance. Or you want to say, "Hey, we want to make sure that we have dedicated build environments." Or "we just want control. We want more insight. We started with one team. Now we have 20 teams. And hundreds of developers. We need to consolidate." 

Stefania: I love the "always be helping." I totally relate to that. And that feeling when you are able to, as Amanda mentioned, with rapid feedback, show people that you are listening and that you will keep delivering or fix the things that are broken and just make the things better that people really want. 

We try to have it as a policy when people ask us questions and they don't find the answer in the documentation, then update the docs and share that docs link with them. I think that's a really powerful thing as a mindset for the company to frame how important good documentation is.

So we've talked a lot about how we turn the mass into revenue, but can we talk a little bit about the top of funnel? Early-on, what were your first inbound channels and how did you figure them out? And what did you do once you figured those things out? Anything from community to product marketing, to creating the right content. And it sounds like, Matt, you for example mentioned earlier that you now have different teams that are creating different content for different purposes, for the leader versus for the developer. 

Can you talk a little bit about that top of funnel and the inbound channels?

Matt: So, our top of funnel initially was the open source developer adoption. Do you market to developers n the classic way? Of course not. I think documentation is the best marketing content there is, to your point a second ago. You've got forums, you've got events, you've got all the things that developers really gravitate to. I think, video is the next big frontier on this. And we're starting to put a bigger and bigger investment into shorter-form video content for developers.

That leads to open source adoption, that leads to all the great things that ultimately culminate in somebody's using this thing for something they care about. And one thing we found was sometimes they call you and they want to buy. It's not even clear why they want to buy. They're not even sure why they want to buy. And you can have some success there early-on. We had some examples where teams using Apollo, they reach out to us. We think it's a good sales signal. Maybe they even filled out the contact sales form. And turns out they're pretty happy.

You can build a feature at some point in your SaaS. We have a SaaS element to Apollo. And so there's a sale there, but the enterprise-ility stuff, what we found is sometimes didn't actually resonate with the team that's using it. They're happy, it's a good product. They had a question or two, but, "Hey, doesn't your boss want you using Single Sign On?" "I don't think so." "Well, should you ask him?" "No, I don't want to talk to my boss about this. My boss doesn't know I'm using this stuff." Sometimes you're pushing on rope.

And so for the first however long it was, it was all inbound. And we were talking to people that were already using the product and it's one of those things where if they're happy, you have one problem, and if they're not happy, you have an even bigger problem. And what we found was the big shift for us was we don't necessarily need to go through or want to go through the team that's using it to have the real sales conversation.

So we have this whole other funnel really. And when we built marketing at Apollo, which is a much newer function than the product endeavor that I was talking about, that's really more about "what are the signals about the companies that are using this stuff?" And some of that can come from our product, but a lot of that, maybe signals about events they're attending, or job posts that they're putting up, or any information we can gather that shows us what the scale of Graph adoption might be in the company.

We have this thesis that every company is going to need a Graph, that's going to come from a bottom-up initial desire for the Graph, but that's a strategic decision. And it was mentioned earlier, sales has a role to play, and a big part of what enterprise sales really does is help people navigate multi-stakeholder strategic decisions across a company.

And so there's this sort of second funnel almost where we can think about having a conversation with the right person, finding the right champion to help us have that conversation. We're not trying to do this with anybody that's not using Apollo. So there's always part of this where we can say, "And guess what? Your teams are already doing this. And here's the value they're getting or here's what they've found."

And now they do want have the discussion about enterprise-ility. It's like, "Wait, I've got how many teams doing what? And I can't see all that in one place or that is completely contrary to our policy around how security is going to work," for example, and I think security for all of us is probably a big driver of the enterprise skew. So that's the thing that took more sort of creativity for us. And that's what unlocked a lot of the interesting things.

And I'll say one other thing on that point, which is for a lot of us, I think PLG is attractive because it's got a certain incremental character to it. You have happy users, let's go get some more, let's add a product feature that drives additional growth or a viral element or a sharing element. I love the way you can share designs in Figma. And that's certainly how I first learned about it is some designer put a URL in front of me and I was like, "This is cool. How do they do that in a browser?"

And that's great because you can feather in as much investment or as much time as you have. The top-down piece, what we found is you really need all the pieces in place for it to work. You can't muddle your way into a strategic conversation for an enterprise sale with a larger company. You've got to have legal, you've got to have the account execs, you've got to have sales engineering, you've got to have a certain element of success in place. All these parts have to be talking to each other and there's just a lot more intricacy to that.

It's less of an intuitive thing, we found. We needed to really, until we had all those pieces in place, it's harder to see the results of that. A message I'd have for the viewers is if you have the instinct that there's going to be an enterprise motion or a top-down motion, think from the beginning about what that looks like. It's true, I think Amanda said, "If you don't have PLG early on in your DNA, you're going to struggle to get it later." I 100% agree with that.

I would also say though, if you're not thinking from the beginning about the monetization and you don't have it in your DNA, the understanding of what it's going to look like to really have these strategic conversations, if that's part of your motion, that also is a wrenching cultural change. And I think the earlier you can sort of think about all that fitting together the better.

Stefania: Thank you so much, Matt. I think that's a good trigger to throw the ball over to Amanda. Where are you in the world, Matt, you said?

Matt: I'm also in San Francisco.

Stefania: All right. Short journey there. Amanda,I would love to hear a little bit about the inbound channels from you as well. But there are things to double click on also, and what Matt was saying in you have to have both DNAs in your DNA.

It's interesting to talk about with PLG, how can you build a muscle of a PLG, where you have the right data to identify your revenue strategy as well, and how you're going to price your product based on your learnings of the PLG? Matt touched a little bit on, there are some people that go and reach out to sales for something. Is there something in your PLG learnings that helped you learn things about your pricing strategy as well?

Amanda: Sure, sure. So early-on, there was a lot of community aspects to how we thought about inbound. Well, I should say first and foremost, it was about building a great product that people wanted to use. And I know that's obvious, but if that's not where you're laser focused at the start, it's just not going to happen after that.

So building a great product, but then also building in that spread through the product was key for Figma. Being able to really easily share files with other people in the browser, a lot of our inbound signups are from people that have been invited into a Figma file. So that's key. So the product drives a ton of inbound. You can also post your own Figma files and those drive organic traffic for us as well.

We did a lot on the SEO organic side early-on. Then community, so many customers early-on didn't want to necessarily share that they were a Figma customer, but because we invested in social channels and creating relationships on social channels publicly with many of our end-users, I could show proof points of many companies using Figma because their end-users are talking about it online, even though the company PR department doesn't want us to use their logo.

You need those early proof points in case studies, but you can get them in different ways. You don't have to write a case study for it necessarily, so those were maybe different elements of the early days thinking. It certainly wasn't advertising. I'll tell you that. We do some advertising now, but that wasn't where we were focused early-on.

Stefania: And how did you build that team? Who was the first person to start building out a community? Who was the first person that started obsessing over people posting on Twitter and things like that?

Amanda: Yeah, the first person hired was a marketing generalist who was passionate about the design community and could be an overall athlete and do a lot of different things. And then we also hired a design advocate or evangelist who had actually had a podcast himself and had a lot of connections into the design influencer world. And so we would create content like live streams where we would invite people to use the product together and show it off in a fun game format.

So it was thinking about how do you create the opportunity for people to see your product and be involved in it? And without a lot of commitment to it, it was more in a fun way. So we did that. And so there was community building by hiring someone in the community to be part of it.

I think what was important was that that person was really honest about what was exciting about the product, but also maybe what wasn't quite there yet with the product, because it set a tone of this company is really authentic about their building of something. And they're going to display the truth about where they are in the process of building a new experience that's exciting, and it's going to change the way we work. So that was key to just the brand in general. But as far as your second part of your question, which was about pricing-

Stefania: Yeah. And product-led triggers for deciding, for example, cutoff points. "You get access to this if you subscribe" or things like that.

Amanda: You yourself can sign up for Figma and start using it as an individual today. Where we ended up putting a pay gate was if you... Well, I should say the magic moment is when you use it with other people, because then you see how it's different and the real value prop behind the product. And so it was important to us that you can experience that on the free plan. So you get a taste of it and then we want you to upgrade to a paid plan because you want to repeat that process multiple times.

So you have to, as you're thinking about pricing, figure out what is that magic moment? How can I incorporate a taste of it in my free offering and then have people upgrade when they want to repeat it over and over again? So that's how we thought about the packaging of it.

Pricing-wise, I think that depends on what you're selling, willingness to pay, your competitive dynamics. So all of those inputs go into it. I know that when I joined, we had one paid plan and we had another one in beta that was more like the enterprise tier and we had many customers on that enterprise tier in beta and we just talked to them about their willingness to pay. So they played a role in helping us determine where we should price it as far as that goes.

Stefania:Right. So qualitative was more of a governing factor than quantitative data. Is that what you're saying?

Amanda: For us, yes.

Stefania: Okay, excellent. Thank you for sharing, Amanda. So over to you, Chris. Can you talk about your inbound channels? Because what's interesting about Figma obviously is that as a designer, you are sharing visual things all the time, but you're not necessarily sharing your deployment system.

Chris: But again, we were trying to push forward a new architecture for the web. So I think the beginnings, you have to try to define what does that mean? What is that? And who's the key players and trend setters and where would you draw an invisible line? And then interact with those people.

One of the things we saw was that early on a lot of the innovation was coming from site generators. And so we, for example, created staticgen.com, which was the largest open source directory in the world for site generators. And that also gave us access to a lot of those creators and contributors so we could start the conversations there. That also meant that it was easier for us to figure out just by being able to measure the popularity of those, like GitHub Stars and so on, which ones that we would interact with the most and write tutorials around and so on.

If you are early in a space where you are a king of a hill, but it's a very small hill, rather than having a small sliver of a large pie, if that's your approach, then something like building up these resources makes a lot of sense. Because it doesn't take a lot of resources and it is going to be a lot harder later on, because all of a sudden it is a big category and getting the SEO to have this prominence is going to be a lot harder.

The other thing was that the first 10 people we hired, they all had developer advocate in their contract. We don't advertise, we demonstrate the product by showing how it integrates with other parts of the ecosystem and so on. And so it's not that everyone had to stand on a stage, but it's setting the tone that, "Hey, we build it and then we talk about it."

And the first advocate we got was someone that we had just seen in the space and he was learning, again this was all nascent. But he was good at talking about it and blogging about what he was learning as he went along We thought, "Hey, that would be great." Someone that's going to be helping building out Netlify. The early audience would say, "Well, how do I even start here? What's the step one, two, three?" And he would share his own journey there and that resonated a lot and developer experience has always been extremely important for us.

Of course we do some marketing and enterprise marketing and so on but very much this organic approach in creating concentric circles in the water from demonstrating the product in itself, and the value and the partnerships. The ecosystem, getting them involved and making sure that it's a win-win for all and they will preach you and vice versa has been incredibly important for us.

Stefania: So this is really valuable insight. And there's a lot of talk about community right now. Matt, you didn't specifically talk about how you originally get people to adopt the open source product, but you mentioned documentation and things like that, but ultimately they have to discover it somehow. But Amanda and Chris, you both talked about community building pretty clearly.

Chris: Very much so. We just recently announced our latest fundraising. As part of that, we're creating a fund where we set aside and are planning to invest 10 million into the ecosystem and have a program attached to it because that's how much it means to us.

This idea that if we are going to be a conduit of an ecosystem, if we're an orchestration layer, then we are nowhere without the ecosystem maturing alongside with us.

So that is incredibly important. I don't think I've ever been in an enterprise call with someone where there isn't someone that says, "I just wanted to say that I've been using Netlify for my personal projects for years and I absolutely love it." I mean, the champion inside is incredibly important for us and remains so.

Stefania: Did I understand correctly that every person independent of a role, the first 10 people you hired, you hired them for a role, but you also included a developer advocate in their contract. Is that right?

Chris: It was part of the contract.

Stefania: That's awesome. So we are very close to time, but I do want to talk a little bit about identifying personas. It is just such a crucial part of building your product and company early on. Amanda and Chris, you both mentioned finding the cool kids in the space and I once had a conversation with the Docker founder and he said, "Developers are like most people, you find the coolest developer that wears the leather jacket and have them put on your leather jacket. And then all of the other developers also want to put on that leather jacket," which is hilarious.

How did you identify your first personas? Have you gone through any struggles in identifying them? Have you sort of switched a little bit? Have they been pretty consistent? Would just love to throw that over there as an open ended question. How much of a mix was that of qualitative and quantitative data research? Over to you, Amanda.

Amanda: The target initial personas tended to be people who were either very passionate about a particular design topic. They were maybe public with their thoughts on that and were prolific online and we could find that through data. Or they worked at design-forward companies. And so it's basically finding the first adopters in your persona group.

They often are the ones that are the most vocal. And so that's who we targeted first. I think over time, we've been able to attract people beyond that audience because as you get a critical mass and all of a sudden like, "Oh, it sounds like everyone's using Figma. What's wrong with me? I better go check it out." And so how do you get to that point is just first winning those first adopters, for sure.

Stefania: I got a great story about Figma today. There's a moment when you've been speaking a language for such a long time that you start dreaming in that language. And someone was talking about, "there was a moment where I realized that I had become a Figma power user when I started applying Figma keyboard shortcuts into sketch" or something like that. You're like, "Ah, okay, it's happening."

Amanda: I like that.

Stefania: Yeah. It's a good one. But you mentioned, you found them because they were vocal. So that's sort of qualitative. Were you able to feed that back into product analytics, and to identify then personas in another way through quantitative data?

Amanda: I think quantitatively more so on just who is active online and how broad is their circle of influence? There was a lot of quantitative research that went into that. And then McKinsey even has a study on design-forward companies. And are they more successful than companies that aren't? So there's research that you can tap into like that to come up with who might be good lists. And so that played a role as well.

Chris: Very early on, it was a little different. Because we were saying, "Okay, if this is the category, who are the ones that we can't do without? What are the key players here?" And "are they founders, or are they just very active on the scene, speaking about it and seeing. Do we have a shared vision? Are they also talking about the same things? Are they going to be likely to understand what it is?"

It was part of the product-market fit exercise. You want to get in front of them and show them something. And if half of them think that this is not a good idea, they don't see the immediate value of it, you should go back to the drawing board. Because if anyone, they should be able to get it. And funny enough, three or four of those 20, work for us now, and another three or four are angel investors. So, it is funny to look back on. 

Today, we have millions of people using the platform and we can see a lot more about at how they use it. And there's a ton of features like advanced language-based redirects. You don't really need that for personal blogs. So, that's a good indicator that's someone where it makes more sense to engage in conversation and so on.

But early on, I think it was very qualitatively, it was very much talking about what are the shifts in the mechanics and how do you buy it? Why do you buy it? What could you be doing instead? What's the overhead that you're aware of? What's the overhead that you're not aware of and what does that mean for where we fit in and what are our dependencies?

And all those things really happened from hundreds of conversations with people that we found relevant in the space, and not so much from looking at large data in our case. Which is often the case when you're starting out, but data eventually becomes incredibly important as you start moving along. Data like web usage and so on helps us move product roadmap positions. We have events and talk to clients and so on, but of course, quantitative data also takes up a big space now.

Stefania: Yeah. Exactly. Anything from you on that, Matthew? You've already talked a bit about it.

Matt: I agree. I think early-on it's got to be qualitative and it's got to be based on a view of who the product's really for and what it solves. I mean, the big thing I'd say is "developers" is a pretty broad brush and I don't think "developer" is a very good persona. You probably want something a little bit more nuanced than that.

What we've found, and I think this probably connects to the Netlify world and Jamstack and the Figma world too, is I think the future is app devs. I think a lot of the technology stack historically was built by and for backend developers. And you had AWS, you've got all the cloud native stuff. You've got every API in the history of the world was really designed by the backend teams.

And the reason GraphQL spread so quickly is it's the first API anybody thought to make that was good for a front-end developer. And if you learn about product engineers, you find they're very different in a lot of ways. The culture and the values are really different. They don't wear leather jackets the same way that the Docker people did. You know what I mean? And the whole move fast and break things thing from Facebook is a piece of this.

I think there's a really interesting cultural shift. When I was in school, I thought, "Oh, the cool technology's on the backend." But my view has come around to, "It's the app devs that make the cool stuff. That's where the value is." It's the companies with great app devs and great apps that are the companies we all respect and admire now. And yeah, they have great backend systems too, but I think there's a big shift in the dynamic.

The thing that really for us unlocked this insight was to start really trying to understand a growing number of developers whose needs hadn't been met with the old way of thinking about this. And not just the product, like how you price things, what support looks like. Everything changes on this dynamic.

My hunch is that a lot of the next generation of companies and products and a lot of the new tech stack is going to come from that culture and is going to reflect a different set of needs and values and experiences from that classic, "Let's sell to developers and ultimately get inside the data center" model.

Stefania: Yeah. This is such an interesting takeaway, and I have seen the same pattern, and those are exactly the people also who are typically quite data-driven. They really want to work with designers and be product-led and use product analytics and understand how their users are interacting with their products and the end-user experience. So quite relevant as a follow-up to John Cutler's DevGuild talk earlier today.

But I want to wrap us up here by asking if you have any quick takeaways for bottom-up founders that are just setting off their journey. Over to Chris.

Chris: When you're just talking about figuring out who the product and so on is for, and there's a lot of ways to think it into the product as well. So from the very beginning, we let open source projects use Netlify for free, and we have thousands of them, the largest ones in the world, and there's both Kubernetes on the backend and React and Vue on the front end and so on using Netlify.

We can see a lot from that usage as well. That actually helps us instrument a lot of what we should be focusing on. And we have tons of open source projects there. And have worked with Apollo and a bunch of other fantastic companies out there. And it's super interesting. So if you can, instrument your product to actually use that to figure out early learnings, to figure out what you want to do.

Otherwise, I just think whether it's bottom-up or top-down or you're selling shoes or whatever, make sure that you know where you want to go, but also that you have immediate goals. And then make sure that you are focusing on also getting value out for yourself on the process of getting there. Anyone I meet that's only motivated by the end goal element, you can say the pot of gold, they rarely make it. I think it's super, super important to be focused on the process of getting to those goals, making sure there's plenty of milestones that'll indicate success along the way. And for God's sake, enjoy it.

Stefania: Yeah. Enjoy the ride. Amanda, quick takeaway.

Amanda: I'll just reiterate, once you've designed and built a great product, think about the experience that you need to design because that's super key. What do you want people to do when they first see your product? What do you want them to do after that? And how can you design that so it's a wonderful experience and they can do that without someone telling them what to do?

Stefania: Yeah. That's actually a perfect input into how to design good data for product analytics. So perfect. Over to you, Matt. Quick takeaway.

Matt: Good luck. Work hard. Look, I think PLG is obviously here to stay. I guess my experience is, I wish I had seen sooner that it may not be the whole story. Knowing that you're not going to know all the answers up front, but thinking about what it looks like to have monetization and be open to the possibility that the monetization may not just be the intuitive bottom-up motion that I think a lot of founders really start with.

There's a lot of interesting opportunity in that space and none of it replaces the need for a great product that people love. So start there, but maybe spend your Friday evenings or Sunday mornings or whatever it is thinking about the other piece too.

Stefania: Thank you everyone so much for giving up precious time, for sharing knowledge, sharing with founders who are on their bottom-up journey, or everyone else learning why they should not do bottom-up, for example. But this was wonderful. Thank you all. Ciao.