April 3, 2019
Ep. #4, Backend Development with Paul Biggar of Dark
In episode 4 of High Leverage, Joe is joined by Paul Biggar, CTO and Co-Founder of Dark, to discuss the challenges of backend development, a...
Your startup might launch as a developer tool, but companies with vibrant developer communities often expand into broader platforms, 3rd party ecosystems, and, with luck, into category leaders. In this finale session from DevGuild: Software-Defined Movements, our panelists discuss how others have won categories, sequenced their partnerships, and tailored their roadmap along the way.
Paige Paquette: Hi everyone. My name is Paige Paquette and I am the Partner and Co-founder of Calyx Consulting, which is a platform and developer strategy consultancy. And prior to this, I led platform and developer marketing at Slack. And joining me today are three incredible leaders who have each played a significant role in building category leaders and category creators.
First up, we have Ali Rayl, who was my former colleague at Slack. She is the VP of Product at Slack and is responsible for helping to make Slack’s 12 million plus daily active users, more productive. Ali joined Slack way at the very beginning. And I got to personally see her lead Slack’s largest, and I would argue most impactful team, customer experience.
Next we have Ed Anuff who has over 25 years of experience as a product leader at places like Google, Six Apart, Wired and many more. Ed led product and strategy for the Apigee API platform and helped shape it into the category leader that led to its acquisition by Google. And now Ed is CPO at DataStax. So, welcome Ed.
And finally, Chris Bach, many of you will probably know Chris. He’s the Co-founder and President at Netlify, which has a massive developer community of over a million active developers. They have raised about possibly over a hundred million dollars in funding. And Chris is probably most well known for pioneering for Jamstack developer movement, which is a global movement and community with events all over the world.
And we’ll get into that and what that means. Welcome everyone. Thank you for joining us. Glad you’re here. We’ve got a lot to cover, so I’m going to dive right in. I want to kick off by talking about the hot topic right now, developer community and customer love.
So Chris, you took a big bet on Jamstack the movement and that has since expanded to a conference and a huge Slack community and much more. Can you tell us, how did you weigh the pros and cons of investing in this like ancillary, adjacent community, like Jamstack versus your own branded Netlify community? How do you weigh that?
Chris Bach: I think that’s a very good question, because I think very often, when we create something, a business, we want to create a category. We’re excited about it. And we think what we’re doing is a new take on it enough to be a category. But very often I think it’s not the right idea.
I think two things you have to do is, first of all,
figure out what’s in it for everyone else. Is this actual a new category play or is it more sort of something that you want it to be.
And I think at the same time, you have to be super aware that it’s two brains essentially and it does create extra efforts. In a startup we’re all spread so thin. It is more work.
In our case, what we saw was that, hey, with everything that was happening from the powerful Git workflows, new frameworks, the browser becoming an operating system, the API economy. If you put a circle around it, you could actually witness websites and apps being built in an entirely new way. And they needed to be nomenclature to talk about. You couldn’t spend 15 minutes explaining all this small parts. You had to have that.
We needed to attach what we could see would be best practices to it as well. And we could see that everyone, even if we would be competitive with others in this space, everyone would benefit from having a common way of talking about it. With Netlify, we wanted to do everything we could to enable this to happen. To enable what’s essentially just a better way of building for the web, in our opinion.
We wanted to be the glue, but we would also be very dependent on the ecosystem in general maturing Netlify wouldn’t be a relevant player at all if it wasn’t for a vast, now rapidly maturing ecosystem of the Jamstack. We had a lot of reasons for why it was the right play for us, but I would say in many cases, it probably isn’t actually. I think the category play is something that you shouldn’t make lightly. It is a lot of work and you do have a lot of risk involved.
Paige: I hear you that Jamstack was a key piece of why Netlify grew so much. Now that Netlify is where it is today, how do you decide how much to invest, like budget, marketing spend, head count into that community versus Netlify?
Chris: I think that’s another great question, because that’s a really tough one and there isn’t a right answer. There’s both the creative possibilities. When you’re communicating values that are inherent to the category versus that are specific to your own services. And the other thing is that, well, it requires some upfront investment as well. Where do you put your resources? And I think it is tough. You have to consider, the category exercises have to be about expansion, they have to be a lot less company specific.
They have to be about the enablement of others. Both other players in that ecosystem and your users.
And where’s that line, how much do you want to put into conferences? I think that you have to sort of look at what’s available. There’s no right upfront answer to it. And the others of course, what are the specific KPIs that you as a business needs to meet and where do they meet in the middle and so on. When you do it really well, you’re furthering both causes at the same time. But it’s not always possible. And I think it’s something you have to be aware of.
Paige: Well, speaking of things that are hard to quantify. Ali, so much of why Slack grew in the beginning, in my opinion, was customer love. And much of that love, I can say personally was driven by your team and the insanely high standards that you set. Most people think about customer support and enablement with things like tickets closed and time to resolution. But I think what really sets Slack apart was around this much harder to quantify thing around customer love.
How should companies think about investing in customer love? Is it something that you actively tried to cultivate or was it just a by-product of your team and the culture, or was it something else?
Ali Rayl: I think it’s not something we intentionally tried to cultivate, it’s more just inherent in who we are. We didn’t start Slack with grandiose visions of conquering the world, we just love to make software that people love. And here’s the thing that we think people will love because we like it quite a bit. Building Slack, we had this really unfair advantage in that we were using our product every day to build our own company. We had a ton of surface area as customers with the thing that we ourselves were building.
We also had the ability to decide this thing is too annoying. This thing is not good enough. This feature needs to exist. Whatever it was, we had the wind at our backs in terms of making a product that we could love. Also recognizing that there are millions of people who had used similar products and had the same expectations and muscle memory. We were able to kind of take,
What do people expect? What do they need? And then say, what do we need to layer on top of this to make it truly valuable?
And then where do we feel like we need to make it better to put it in our customer’s hands?
And the aggregation of all of those decisions, that’s what lands with the customer support team. So, at the very end of all of this, all these decisions, all this work, that’s what lands with the customers. And then this just has to be a really consistent throughline, which is when our customers talk to us. It had to feel like an extension of the product that they use. And especially at the beginning, we didn’t hire our first marketing person for almost a year after we launched Slack.
The only brand exposure that people would have for the most part was the app on their phone or their desktop and us in support and Twitter, if they contacted us there. It’s really about creating that … Well, one, just being true to who we were and what we really cared about, which was making great software. And then two, carrying that experience through. So, so much of what we did in support was just, how do we support everything that we’ve done up to this point to have these real Slack authentic conversations with people once they get there?
Paige: Totally. But I think there’s like a real tension there that I want to name, which is that so much of the customer love driving things, are almost at odds with more traditional, quantifiable metrics, like time to resolution. You want to close as many tickets as possible. For companies that are watching this, how should you quantify customer love? Is it quantifiable? Or how are you going to know if it’s worth the investment?
Ali: For us, I think the best quantification was word of mouth. So, if you’re trying to grow a company, you have a few different paths, you can hire an army of salespeople and go sell. You can hire a marketing team and go market, or you can do word of mouth. And all of our growth was word of mouth.
It’s the cheapest and the easiest to do if you can develop a platform that supports it. So, if people are willing to go around and say, “hey, you should check this out.” Not only is it basically free to you once you’ve put in the investment to make your product worth recommending. But that recommendation also carries more weight than any of your advertising or any of your sales ever could.
I think that the top line metrics that I would look at is, are we getting recommended? NPS is one way, like one strict and very industry standard way to quantify this. It has its problems. It also has its benefits, but ultimately
are you growing organically? And if so, you’re doing something right, because you’re providing people with enough value that they’re taking it with them and telling others.
Paige: Totally. There used to be a wall of love showing all the tweets, which is standard practice now. But yeah, it was very tangible.
I want to switch gears a little bit and talk about acquisitions and partnerships which is another super common area that I get asked about all the time. Ed, last year DataStax acquired a company called The Last Pickle, which is a great name. One of the most prominent Apache Cassandra consulting companies who had also developed some very popular open source tooling. Can you talk to us about the strategy of investing in open source tooling and why it’s important?
Ed Anuff: I think that this ties into a lot of the previous conversation. DataStax brings Cassandra, the open source NoSQL database, and we bring it to business customers and enterprises. But for us, a lot of the love comes from the community, comes from the people, it’s a product and technology that’s built collaboratively through an ecosystem that comes about this. It’s something that people are doing because they’re fascinated and interested in the technology.
The Last Pickle, this is an organization that’s been involved in Cassandra since day one and has built a lot of tooling. Per your question, what’s the importance of investing in tooling? Well, this was tooling that actually grew up in the ecosystem, and this is what developers were using, people who were building Cassandra. And so to us, it was very important that we would create a home to sustain this work.
We spend a lot of time worrying about the Cassandra ecosystem itself. Is this something where word of mouth is occurring? Is this where the people who are using Cassandra, having such success with it, such a positive experience with it, that they themselves are the biggest evangelists for it, and are the people who advocate for it to all the other developers that are building new applications?
How can we help them be successful? Tooling is intrinsic to that.
Paige: And how do you think about measuring ROI with something like an acquisition of open source tooling?
Ed: Yeah, that is the perennial challenge. We pay a lot of attention to a variety of signals. Most people go and look at the first order of things on GitHub and so on. How many stars or followers do you have on a project. But we also go and try to look at mentions. We try to look at how often we get mentioned within mailing lists, on Stack Overflow, hits within blog posts and so on. You’re able to go and sort of get an extended share of voice type analysis that really goes and says, okay, people are aware of this. People are talking about it or using it.
It’s always an indirect metric. I think that’s part of what everybody discovers within developer oriented communications and so on is that you have to go and say, okay, this metric is a proxy for what the awareness is within the ecosystem. We use a lot of that and it helps us triangulate on how people are getting benefit from these things.
Paige: Chris, speaking of acquisitions just last week, Netlify announced the acquisition of FeaturePeek and also launched some new features that leveraged FeaturePeek’s technology. Can you talk to us about how does this play into your strategy and do these acquisitions, like the one with FeaturePeek change the nature of your community? How do you think about that?
Chris: Absolutely. Super excited about the FeaturePeek acquisition we launched last week. Where they play in is that, what Netlify traditionally did was sort of introduced the notion of. all you do is push a code to Git, and then we build it up for you. But you had all sorts of staging environments. So, essentially we just replaced it with what we call a deploy preview. It really minimizes all the efforts that goes into that part of development.
But at scale, and that’s where Jamstack is going, it’s just becoming how we build for the modern web, collaboration is key. Essentially building anything for the web is a team sport. Once it becomes more than just minor projects.
It’s a little hard to work together when you have developers, they live in Git. And you have all these sources of truth, project managers in JIRA or Trello. I get something and I take a screenshot and I send emails, and where does this go?And FeaturePeek is essentially a layer over the deploy preview where everything is consistent there. That’s persistent in all the project management tools in GitHub and so on.
For us, what it does is it expands that community to also start involving a lot of the stakeholders that aren’t necessarily developers.
And a lot of people come into play and there are a lot of stakeholders in a lot of website productions. For us, it continues what we’ve always wanted to do, which is enable developers. And now it’s also saying, we want to enable more stakeholders in the production and the FeaturePeek example is great. The FeaturePeek acquisition is a great example of that.
Paige: That makes a ton of sense,. You want to get more cross-functional stakeholders using Netlify. How has this playing into the Netlify community strategy? Do you think that you’re expanding it?
Chris: I do see it as an expansion. Before it was very developer focused and it absolutely still is. But you’re adding that layer, the developers in the real world have to interface with a lot of people in their organizations that aren’t developers. They can be clients or run their own company. And so how do they work together with them in the most efficient manner?
Well, if you can have everything consistent with a single source of truth, where you’re not compromised, you still get to have your code live in Git. The project managers or whoever it might be still work in Trello. Or in the Asana or JIRA or whatever it might be. For us to bring that together means that you’re sort of creating more fertile grounds for community to be developed around that part of the collaboration as well. I see this as an empowerment of what we’ve got, and an expansion into talking to some stakeholders that aren’t traditionally so much technically molded in what we’re doing.
Paige: I want to take us a little bit back in time to 2016, where you were already in talks with Heavybit around the idea of Jamstack in its infancy. And the world was already familiar with BitBalloon, but Netlify was just in its infancy and maybe only existed mostly in roadmaps. So, why did you choose to build education and partnerships around a movement of Jamstack before finally having your commercial product of Netlify?
Chris: I would say BitBalloon was commercial, but it was just an MVP to make some observations about the space. “Does that mean what we think it means?” And it proved it. And then we said, okay, well, what’s the end goal here? The end goal is to build websites in this much better manner. What is our role in that? Well, our role is that this is going to need viable workflows. Let’s try to be a conduit for people and developers and companies to be able to make the parts fit together.
And again, that needs an ecosystem. So, Netlify came out of private beta in early 2015, and we were talking adjacent with that. With, “hey, this is part of a new way of thinking.” And in the beginning, they’re quiet small, but you stuck with them and you’re going in concentric circles from that. That’s always been my philosophy. And just like Ali was talking about. It’s about word of mouth.
It really is about making sure that the people that get it will help spread that word to much larger circles. And so I think in our case, Netlify was a bet on the expansion of the category. And we had to push forward ourselves. So, completely intertwined. I don’t see it as pre-launching a commercial product.
I see it as a mission to bring a version of the web to the world where Netlify provides some of it and the ecosystem provides some of it as well.
Paige: You’re literally creating a market. What were the most impactful partnerships in those early days?
Chris: In the beginning, we had to define our own space, what is this? Where do the borders go? What does it include, or what it doesn’t? Who are the opinion makers, like the founder of GitHub and so on? Who do we see as pillars in this? Even if they’re just like people that are very vocal in it. Who do we share a worldview with? I think that defined the early partnerships. Which was more like individuals who reached out, and we showed them what we made. And we said, do you think that this makes sense to you?
Because if they’re the choir and they think, nah, it doesn’t really solve the problem, then we should probably go back to the drawing board. And so in that we made a ton of connections. The founders of Heavybit obviously have a history at Heroku before. And that was part of that play of the one-click deploys and the simplicity there. That’s how we got that way.
Git is obviously a huge driver of the modern web as well. And an essential part of our worldview. That’s how we got connected. Then they become more commercial partnerships. And in the beginning, tons of the partnerships essentially were with open source projects. We basically said, “hey, this is super cool. You’re making this build tool with a site generator. We’d love to have a one-click deployment into Netlify for example.” They would be like, “hey, it’s great. It’s easy to get started with my project. It automatically deploys and it has a place to live. And so it’s cool.”
I think you have to find fellow enthusiasts, you have to find people that are aligned with where you want to go. And then the commercial aspects will follow.
Paige: And then build it together. And speaking of software improvements ChatOps, Cloud Native, NoSQL, these are all movements that people associate with your work, Ed and Ali. Can you talk to us about what your teams have done to proactively steer conversations around these movements, if at all? And does this actually benefit your commercial product lines?
Ali: It’s a good question. In terms of proactively steering very specific conversations around, for example, DevOps or ChatOps, those are best facilitated through really targeted communication mechanisms. So, we speak at PagerDuty conferences. We’ve worked on some targeted blog posts about how you can use Slack for this. We write code examples that we put out that show, here’s how you might manage an incident in Slack. Bridging the gap between, “I have a concept of managing an incident or deploying code or whatever in Slack” to, ” it actually works.”
Providing some specificity in there so that people can see the steps along the way to get there is really important, but also to make that general enough.
This is the beautiful part about working on devtools. You can be like, here are the broad strokes to get there. And here are the tools in the toolbox to get you there.
And everyone’s like, “oh yeah, okay. I see how to do this.” And they’ll pull out the parts they need and put in the parts that they need to get it there. This is the thing that’s been really useful for us. Working with folks working on the platform is that we can kind of sketch out a vision. We can show a few paths and then people just take it and run with it.
And this applies to not just NoSQL, DevOps, ChatOps, whatever. This is really how the platform has taken root, is by working with developers and giving them a lot of tools, a little bit of vision, and then seeing what happens.
Paige: You can also source these stories from the community too. There are so many developers who are excited to talk about how they do that. And it’s so powerful.
Ali: Folks go to conferences and they’ll see, Slack is just kind of a side note. Like here’s how we do this, this and this. And it just happens to be in Slack. And I’m sure you remember it, we’d reach out to them and be like, hey, can we talk more about the Slack part of that presentation? Because that’s really interesting.
Paige: What about you Ed? How do you think about leveraging this movement and tying it to the commercial benefit of the product?
Ed: Well, a hundred percent of our customers come to us through essentially the movement. They start with their mission. They’re trying to build some new type of application. They know that they’re going to build it in a different way. They know they’re going to do their development in a different way. They want to do it on open technology, on open source technology. They come in and this ecosystem helps them learn about what’s possible. They find that out through the success stories, people talking about what they’re doing.
That’s the great thing about the Cassandra community. It is that from day one the successes of how large companies, so people like Netflix, have built on top of Cassandra was part of the key motivation. These things built on top of themselves. We spent a lot of time thinking about how do we amplify the voices of the community? How do we let the community, the successes and activity, speak for itself.
As long as that happens, as long as there’s energy in the Cassandra community and it’s growing and it’s dynamic and people are communicating their experiences and that’s propagating, then us building a commercial business is pretty easy.
So, we spend, on top of that. We spend a lot of time going and saying, okay, where are the leverage points where we can amplify the voice of the community?
Paige: Just showing rather than telling people why something matters is always more powerful. The best case scenario is where you have a community that’s robust and healthy, and everyone cares about it. But of course, there’s also the danger of being put in the wrong box in the wrong category. And I know this is something that we talked about earlier that you have faced.
Ed, can you share an example of a time where you were miscategorized and what did you do to resolve that?
Ed: This is one of the things that every developer organization goes through. As a startup, you grow up and where are most of the developers? Well they’re at enterprises. And so as you evolve your messaging to that you start to deal with a lot of folks, analysts and so on, whose job it is to help enterprises learn and evaluate these new technologies. And the standard thing that will happen is they’ll try to put you in a box based on what they already know.
At a previous company, Apigee, we were trying to help companies that wanted to have an API program. We wanted to mirror the success of companies like Twitter or Facebook or Amazon that had successful API programs and develop ecosystems around that. We would provide those capabilities to companies. Many enterprises took advantage of this, because they wanted to have developer programs. T
he analysts would look at that and say, well, this is integration middleware, which was a different and previous category of software. And if you squinted just right, you might go and say, oh yeah, maybe they’re in the same category. But that fundamentally did a disservice to the users and use cases that we were trying to solve. And so for us, the way that we got around that was to really identify who are the people at these companies who were excited about APIs, who led these things.
What we found was it was a bunch of different folks. But what they had in common was they all essentially were the product managers for their APIs. They didn’t call themselves that, but that’s what they did. And so what we did was use a little bit of thought leadership and evangelism to go and get people to use that term. And so then we would go back to the analysts and say, hey, we’re creating products for API product managers.
And that finally was the tipping point for the analyst to go say, “oh, this is a new category. What we were talking about when we said middleware, that was for enterprise integration specialists. You folks were onto something new, these API product managers.” And once that happened, it snowballed and the whole category took off.
It’s a little bit like what Chris was talking about with Jamstack. Once you sort of broaden it and you create a name for the movement, suddenly even people who are trying to put you in this box before suddenly actually become your allies. They’re like, “I now understand how this is fundamentally a different thing.” But you have to help them get there. You have to basically create the new box that you’re in.
Paige: I love that you were giving a specific example there of literally finding this phenomenon was already happening. People were already being product managers for APIs. But maybe the category didn’t exist discreetly in people’s minds. So you went out and created a category with something that people could identify with and say, “yes, that’s me.”
Ed: Movements happen. The interesting thing about Movements is that they happen without anybody actually almost being aware of it. And so part of your job is to actually help give a name to the movement and help create the common terminology. And sometimes that involves you working very closely with the companies that you consider as competitors. But the goal is to create this common language that we can all use to help grow the category and amplify the movement.
Paige: I think LinkedIn also did this with the term ‘social selling.’ That wasn’t really a thing before. That’s super helpful because when people talk about category creation, it’s very hard to nail down a specific example of how to do that.
I think we have time for one last question and I’d love for each of you to answer this one. As category leaders today, what could you have done or would you have done five years ago to better prepare yourself? What’s the single piece of advice that you would’ve given yourself or to other founders who are going after this bottom-up go-to-market motion?
Chris: It’s a tough one to answer because you learn so much along the way. And you just want to go back and say, ?hey, like you should have … ” But there’s a lot of details in it. I want to piggyback on what Ed just said. I think there’s a difference when it’s a GTM that is about bottom-up developer saturation. If you have some enterprise software, very often even nascently you have to go through procurement and you have a version of it where your buyers are sort of buying it for the same reasons. And then you grow from that. You can scale it tremendously.
But in a bottom-up motion, very often the version of your product is going to be different to some extent and developed. So, people might use it for small use cases. And then it grows and it becomes an enterprise thing. And so the framing of your own product, a lot of it will happen by people self-architecting, and growing with you.
I think it’s important to remind everyone internally to make sure that you really have to develop your messaging and everything along the way, as the use cases grow tremendously. And that is a little bit different. And you can get ahead of your skis by communicating something that is very enterprise rich. I mean, basically it’s this whole dual funnel thing. You’re catering enterprises, you’re starting to get a lot of revenue from that. But the whole point is you’re building a standard. And you’re pushing forward in movement.
It’s really, really important to always have that focus on, “hey, this is only healthy if you keep exponentially expanding that top-of-funnel. And you keep exponentially push forward a standard.” And trust that the monetization is coming, of course it does if you’re providing something that has value for users or for companies. But it’s easy to fall into the trap of choosing one or the other. I think you have to keep that balance.
Paige: And it’s a tough nut to crack. Ali, what about you? What’s the piece of advice you would give other founders who are thinking about bottom-up?
Ali: There are a thousand things that I wish we had known. But one thing I wish we had really, really figured out was where’s our growth coming from? We were in the very fortunate position that growth just happened to us. We had a fantastic word of mouth funnel. It was all just organic. And it took us a little while to really understand, what is the source of this. Because the important reason to figure that out is like, how sustainable is this? And where do we need to go to next?
I think of this kind of like curling versus hockey. So curling, you are staying with the stone and you are just like, it is your thing. And you’re just making that glide as smoothly and as far as possible. And in hockey, you need to skate to where the puck is going. And I think that this is a little bit of a tough cycle to be in, because if you’re getting so much traction, you’re getting so much love and everything you do is incrementally just that much better for that audience that you already have. The cycle of feedback just feels really, really good.
You almost forget to look ahead and say, “we have this market nailed. What’s the next one? What’s the next one? What’s the next one?” So, I’d say focus on generating the love that is getting you recommendations. But don’t forget that your business needs to grow.
Paige: And the audience where your business needs to grow looks very different from the audience that is currently curling with you.
Paige: And as a Canadian, I appreciate the metaphor. So, thank you for that.
Ali: Doing what I can.
Paige: Ed, what about you? Any advice?
Ed: I think it’s really important to have a point of view, and don’t be selfish. Understand what your point of view is. Spend a lot of time, be opinionated that there has to be a better way.
There’s a better way of building your applications. There’s a better way of managing your data. There’s a better way of working with your colleagues. What we’re all saying is, there’s a better way. And always go back to that.
When we look at our developer events, our most popular event that gets large turnout is the introduction to NoSQL. There has to be a better way to do data. It isn’t the specific talks about getting started with Cassandra or getting started with DataStax’s cloud products. It’s about, “there’s a better way to manage your data.” And that gets developers every day to come to our sites, to our webinars, to our events. And don’t be selfish. Be willing to focus on that. There’s always a lot of pressure on the idea that the better way is you have to buy our specific product.
The developers will get there. If you’re aligned with helping them understand this new way of doing things and they get that and they get that you’re authentic around that, then sure. You’re going to be the person they trust to help them get whatever specific product you’re trying to get to. But that’s the part I would tell myself, that I tell myself every day, and what I would have told myself five years ago, which is what’s that point of view? What is the better way that you’re advocating for?
Paige: Awesome. And with that, we’re going to rap. Huge thank you to Chris, Ed and Ali for joining us, for sharing your insights and to our audience, to everyone out there listening, whether it’s live or not live.