[Weekend Drop] Developer Relations (with Sai Senthilkumar of Redpoint)

Redpoint invited me to riff on DevRel topics and 400 people signed up.

I was interviewed by Sai of Redpoint based on these blogposts:

The session was covered by Tom Tunguz, whose blog I love (https://tomtunguz.com/shawn-wang-offi...) and the feedback was wonderful!

Full video on YouTube https://www.youtube.com/watch?v=guK1XiLQbH8

Timestamps
  • [00:01:42] What is DevRel? 
  • [00:04:59] Where should DevRel report? 
  • [00:06:57] Getting Started with Early Stage DevRel 
  • [00:12:28] How to Structure DevRel Efforts 
  • [00:16:02] When to Hire First DevRel 
  • [00:18:23] Community and DevRel 
  • [00:25:41] How to Start a Community 
  • [00:29:47] Technical Community Builders 
  • [00:31:04] Social Media Managers 
  • [00:33:14] North Star Metric 
  • [00:39:20] Product DevRel 
  • [00:40:37] Finding Great DevRels 
  • [00:43:47] DevRel for Dev Platforms of non-DevTools companies 
  • [00:46:38] DevRel Tooling 

Transcript
 
[00:00:00] Sai Senthilkumar: My name is Sai and I'm at Redpoint investing primarily in B2B software with a focus on developer oriented business. I'm very excited to be chatting with Shawn Wang today about the importance of developer relations for any company selling to developers. 

You know, we find that several developer companies we work with today are hiring for diverse leaders and oftentimes it's function gets overlooked early. Or maybe not built out soon enough. So today we'll talk a little bit more about how to structure and measure our world-class Debra organization for any startup and why it's so important for a company's overall health. 

So I'm wanting you to be joined today by Shawn, who is the head of developer experience at Temporal. Shawn, do you want to briefly introduce yourself?  

[00:00:39] swyx: Yeah. Hi everyone. I am Shawn Swyx online as well. I guess my dev role ex experience starts at Netlify where I was the second DevRel hire. And we grew from about 30 ish people when I joined to about 250. 

And. I think something like 300,000 developers to 1.5 million. And then we, and then I left in 2020 to go to Amazon where he spent a year working at amplify and thinking about AWS level or branded Daryl. And we can talk about what it's like to work at. You know, a series B to C stage company. 

The rail versus a big company devil. And then I joined Temporal this year in in February to head up developer experience. And we're a series, a company focused on microservice orchestration, which is a bundle of words, but basically we're reinventing asynchronous programming. And if that doesn't hook your interest, I don't know what will, so I'm happy to talk more about that. 

[00:01:42] What is DevRel? 
 
[00:01:42] Sai Senthilkumar: Awesome. So is that Shawn is the, is the guy to speak with, in terms of structuring and starting out in Beverly also Shawn, I guess starting with the basics here, you know, many people wrote in asking for clarity around the devil row. So, so in your, in your mind, what is Deborah and the various roles and responsible. 

[00:02:04] swyx: In what is dev route and the various rules and responsibilities. Okay. There a very big question. So dev REL I think is essentially for a lot of people is essentially rebranded marketing. Developers. Don't like to be marketed to every time you hire a professional marketer and you get them to talk at developers, their eyes glaze over and they're turned off by your marketing buzzwords and your emphasis benefits over features because you refuse to talk about how things work because marketers don't know how things work. 

Cause they're not technical. You hire developer relations before. Developers want to be spoken to by other developers. And they want to be explained on how to use things, why, and not to be handheld too much to do some hand hand-holding, but not to do too much handholding that you restrict their creativity. 

Because I think some of the best DevRel programs have often just said, we can't wait to see what you build, which is a very cliched term in Debra. It's actually, it's pretty true. If you talk to the early Twilio, derails, they just held hackathons and they're like weird a communications layer. What can you come up with? 

And they are often impressed in a lot of their new products direction comes from the, the stuff that developers want to take their product in. And so Val is very much of a bottom line. Developer first marketing efforts. And I personally segments the growing sub specialties that devel into three set, three segments, which is community content and products. 

The reason I add products in there, which is not a very common thing to, to emphasize with Daryl is because developer relations has. Background or backstory as developer evangelism, which is kind of the old Microsoft slash Google name for it, which is essentially you hire professional influencers to travel the world and give talks. 

And it's very us to the rest of the world. Like I'm pre. The good word, which is very nice because a good talk and a good useful demo or a good you know, explanation is, is actually a very important, but there's not much of a two-way street. So, it's very, it's very like us coming out to them. And I think now people understand that they, once they're devils. 

Their PR company in front of developers. And they talked to them so much that we can actually use that product feedback to feedback into the development of the actual product itself. That's the vision. That's the, that's what a lot of people say that Debra is a two way street developer. Evangelism is a one-way street in practice. 

It's more like 99% outbound anyway. And 1% inbound. The reason being that no one has time for your product feedback. Everyone has their own product roadmap. You're not proud of the PM org. You're not part of the engineering org. Who are you that I have to listen to you. So people are still figuring this out, but I think the content and community pieces REL are a bit more developed. 

[00:04:59] Where should DevRel report? 
 
[00:04:59] Sai Senthilkumar: Totally. 100% agree. And when you mentioned this in the beginning, but I think it's important to flush out the difference between. Developer advocacy and Deborah and for marketing, I think oftentimes those, those two those two topics get confused with each other. So, so yeah, I, yeah, I think that is that's important for any startup that starting out. 

And this is a separate organization,  

[00:05:21] swyx: Hopefully, well, so there's there that kind of leads us into a natural segue of where does deferral report to, but you can have that conversation. I'm not sure that it's a solved question yet. I think a lot of people are converging on the consensus that dev roles should report into product. 

And I think that's fine, but I'm also, I don't judge people where dev REL reports into marketing. I don't judge people went there for all reports into engineering or, you know, at Netlify we were a separate org. We were separate from products, separate from engineering, separate from marketing. And we reported up to the CEO. 

Okay. It matters basically as a founder or, you know, or based on company direction, how important your dead relatives, your liver function is how senior you are a different leader is and also how you measure them because at the, at the end of the day, even if your dev REL team reports into product, but the way you measure. 

Is entirely marketing metrics. Like how many hits did you get to a front page? Then they are our marketing team. It walks like marketing talks about marketing, their marketing, even though they technically, you put them in your org chart under the product team. So be very careful about just play playing at like making your devil or you know, part of product, but not actually, because that leads to some interesting tensions between. 

What it means to be a representative for your company and how that actually leads to tangible impact for your business. So yeah, I'll leave it at that.  

[00:06:57] Getting Started with Early Stage DevRel 
 
[00:06:57] Sai Senthilkumar: And it, even before you get getting into building, you know, Deborah, I guess like what, what are some tips that, you know, any early stage startup or founder that's in that develop that is selling to develop. 

What are some tips that they should know when getting started? Yeah, I guess what are some lessons learned from, from your time of  

[00:07:13] swyx: peripheral? How early stages are we talking? Let's say series  

[00:07:18] Sai Senthilkumar: a, you know, you've just raised some capital, you know, you're scaling to the series B, you're selling to developers, developers, easy. 

These are core bread and butter, you know, I guess what are some.  

[00:07:27] swyx: I think some amount of focus and regular cadence is important. So, so at, at a point of series, a like you have some form of products like your, you have been around enough that you, that you raise your seed and you, you convince people, okay. 

Invest in serious. Say. That's probably where you should start hiring you know, a different team like we hired to, to I, I came on in February and then we hired two DevRels this year and I think that's where you start trying to figure out what the public story is. You want to tell, because you have some core base of early adopters, but you need them to not just you know, not just grow into like, the early majority, which I love to use this model of crossing the chasm. 

And I like to say that my personal specialty is helping developer tools cross the chasm, which is a very. Profitable a specialty, if you can actually do that repeatedly. But yeah, so you need to start basically like kind of start building machine of like curious what here's a collateral that we here's the typical story, like our, our one sentence pitch, our five minute pitch, our 10 minute pitch. 

So 20 minutes a patient on a one hour pitch and you need to kind of market. Among different sets of audiences and figuring out what kinds of audiences like how to qualify these audiences immediately so that you can go down, you know, between two to three different talk tracks that are usually relevant. 

Because the reason I say two to three talk tracks is because I'm not a strong believer that there is one marketing message to rule them all. Because a lot of developers who was companies are horizontal companies and it can be used in multiple different ways. And the things, the words, the. Influencers, the concepts that appeal to different audiences vary based on their own background. 

So you need to figure out like what, what the, what that fit is. And then I think for me, like, the most important thing to establish at this early stage is the developer journey. Like what, yes, you probably have as CSA stage, you probably have a working products. You're probably building more features that are going to take you to the next level. 

And your products is pretty dense already. And probably a lot of people barely even scratched the surface. That's what everyone says, right? Like it's a little bit depressing as a deputy founder to build a bunch of features. And then like people just use you for like the bare minimum. That's, that's just how it is. 

A lot of people don't have time for anything more than the, the core. So first of all, your core value proposition needs to be really good, but second of all, you need to guide them through the rest of your product to see how awesome it is so that it can grow and usage. So that I termed the developer journey. 

What should people. Land on your landing page. And then they're like, okay, like my friend said, this is cool. I'm going to come check it out. But I only have five minutes. What are you going to give them in those five minutes? Are you going to dump your entire feature set? Are you going to give them a hello world that is trivial? 

That is that they could do in a weekend? Like what is the, what's the wow moment that you're really trying to deliver? And what's the open question that you're trying to leave in their minds. They're not happy with the five minutes and they want more. So I think that's something I think about a lot. And I, I draw a lot of inspiration from game design where it's a progressive review of features. 

And you don't give them that much apart from like, do this and then do that. And then, and then go here and then go there. And you have a very strong sense of like what level one level, two level three, it looks like. And so that's the journey throughout through to products for me, it's about what a core, what the core concept is. 

You know, doing a hello world or like, you know, deploying the, playing a basic app and then building, and then branching out into maybe like performance or security or deployments monitoring that kind of stuff. But that's for us, you know, and your, your message could be a bit different. Finally, I'll, I'll talk a little bit about specialties within the team. 

So , we are very focused on front end developers. So the way that. We'll thought our team was that we had different framework specialists. So I was the react specialist because I'm a well-known speaker in reacts. And we have view, we had angular, we had you know, sort of static site generators specialists for us at Temporal. 

We have SDKs in different languages. So we have a Java slash PHB guy, and then we have a gold person and I'm going to be the TypeScript person. And we'll, we'll have Python, ruby.net, whatever. You might want to split it up by, by that because there's a certain audience that each of these developer relations people speak to that maybe they're already known in the conference circuit. 

Like it's super easy for me to get into any JavaScript conference. And I think that's, that's a, that's a home ground or unfair advantage that you should play when building out a different team. It's a bit hard to hire like super well known people. So sometimes you need to have an eye for trajectory rather than. 

You know why rather than you know, where they are today because they Def Netlify definitely took a bet on me very early on when I was getting started my speaking career and it paid off for them. But sometimes it doesn't work out and sometimes you have developers where we kind of are not that productive, but I think that's where a really good dev manager can come in. 

And up-level your individual contributors. 

[00:12:28] How to Structure DevRel Efforts 
 
[00:12:28] swyx: So. Yeah,  

[00:12:30] Sai Senthilkumar: no, that's a great point. And I think you know, we'll get, we'll get into this later when we, when we discussed how to actually measure the ethicacy of if you done an organization, but you, you have a great way of thinking about it. You know, I think in the seminar we said something to the extent of there are four pillars of developer relations. 

And I think that is that's a company like Hashi Corp, where they have. You know, there, there are much later stage that they're not a series a or series B startup. And I think they bring it into education, community events and audience, but you know, that's just not realistic for a series a or series B startup. 

That's just starting out. And I think you just have a great framework for this by just asking what kind of Deborah are you? I think you get, as you mentioned, you have those three emerging subspecialties and Beverly the community focused upon think focusing product focus. Just asking that very basic question in the beginning. 

I think we'll go along with.  

[00:13:18] swyx: I think so, too. So, I mean, since we're on zoom, can I screen share the visual? Just so we so that, so the audience members can actually see what we're talking about. So, what size talking about is this four pillars thing from the previous session with Adam Hashi Corp. 

And I mean, you know, there are a series what F or E company. And I w yeah, I would love to have an education team, you know, especially former teachers. A really good every time I see a former, you know, someone who actually has done teaching in like a public school setting or something crossover into developer education, they're amazing at it. 

A formal events team. Amazing, fantastic. You know, but I, you know, I think these require fairly large, large budgets and have a very long PR cycle. Whereas I think maybe a lot of earlier stage companies are more focused on community advocacy and audience, right. So my, the way I break it down is a little bit simpler. 

I and for the record, like I, I had seen this before, but like I just kinda made up, meet this up based on the people that I know. So like I have. Certain archetypes in my end, when I look at these things and that's kind of how I view it, right? Like at the end of the day, all everyone wants this monthly active developers. 

That's not screw around. Like that is the source of truth for where Deborah is going to feeds into the, the rest of the value equation for a startup. There are certain specialties for people. Like some people are not that great at content. You know, they, they, they don't produce like, especially inspiring blog posts or talks or anything, but look, they're charismatic as hell. 

You know, they, they, they just know everyone who's anyone. So their community devils, you know, whereas some people are just like thought leaders. They're just, you know, they just get chunk out like amazing and visually appealing and just like fantastic stuff. But then maybe, you know, and, and so maybe it should be applied. 

And then others, like they can, they're, they're sort of more quiet, but they are very inside. The understand the core user insight that like people, you know, when they aggregated a lot of feedback from people, they can really sort of channel that into a very strong sense of product feedback. So I liked this and I liked that for me, the most impactful time as a dev REL on. 

Was on product launches. So really focusing on making a launch successful by being the first beta user internally within the company RV products and just being brutally honest, like if you cannot pay someone to be brutally honest at you, then you don't have an intellectually honest culture. And then, but then also like making the demos and making the the collateral and the, the storytelling that will make your launch successful and a really good launch. 

You can't really emphasize it enough to, to really help you. So I'll, I'll leave it there. I don't want to like spam people with, with too much, but I can, I can talk a bit about intern about each of these specialties and we'll get  

[00:16:02] Sai Senthilkumar: him.  
 
[00:16:02] When to Hire First DevRel 
 
[00:16:02] Sai Senthilkumar: We'll get more into that when we, when we get into measuring measuring the ethicacy. 

You know, you mentioned you know, after series a maybe even after your seed, but it seems that there isn't a black and white answer in terms of, at what stage in a company's growth earnings, should it consider building a debt relative? But it, it, it seems that you know, it may where it may be different from company to company, but generally it's when you, you're starting, you have a product out in market and you're starting to get developers to start paying for it. 

[00:16:30] swyx: Was there a question in there? I mean, I agree,  

[00:16:33] Sai Senthilkumar: I guess for Temporal, when did, when did they hire you? And then when did you know, when did they start thinking about it? Was it after this year?  

[00:16:42] swyx: Yeah. Yeah, exactly. Right after series a, they they started reaching out to me and I, I said, no, at first. And then I said, yes. 

So that's the short history. Cause like I just joined Amazon and I was like, I was very, very ready to rest, invest for four years essentially. But like, you know, I think when you, when a rocket ship comes along and something that. Fundamentally feel at your core has been missing all this? Well, I think we should take the totally. 

[00:17:10] Sai Senthilkumar: And you were getting into this earlier before, but the question of who should that role report up there? It, I mean, ultimately it touches multiple facets of the business, product marketing, customer success. So I guess there isn't a black and white answer there as well.  

[00:17:24] swyx: There's no black and white, but I think most people agree on products. 

They just don't. You know, it's one thing to like, all right, move some people around on your org chart and just say, okay, you're under products now. And then job done. Right. But then your metrics haven't changed. You. Okay. Ours are all marketing. So what does it mean to belong in products, but then you measure them like, That you just, you're just treating them like a marketing team. 

And they're a very expensive affiliate marketing team for a lot of people. That's how it is like at you know, at, you know, Phil in previous company, we counted our Google UTM tag attribution, right? Like, oh, I drove 10,000 visits to our blog this month. And you can pay for that way, cheaper, elsewhere. 

You can pay for inorganic surge. You can pay for like, you know, there has to be something else that your devils are giving you because otherwise they're just very, very expensive affiliations.  

[00:18:23] Community and DevRel 
 
[00:18:23] Sai Senthilkumar: Yeah, that's a great way of putting it. Awesome. You know, I, I want to make sure we hit on community, the importance of community here, a strong developer relations or John UI communities on the side. 

You also have a community at Temporal, I guess. Why, why invest in community, you know, in the first place you have a great post here, but I, you know, the audience would love  

[00:18:41] swyx: to hear. So I guess someone will share it. The post. So I'll give, I'll give my community credentials for people who don't know. So I got started in the New York tech scene where I was very much, in-person just getting to know everybody at every single meetup. 

And then I moved online where I was the moderator for the R slash react to a subreddit. And we grew that subreddit from 20,000, 30,000 people to 220,000 people before I left. And then I started small society from scratch. And now that was about. A year and a half ago, and now we're at about 11,000 people. 

And then I wrote a book where I run a paid community for the book. That's about 2000 people. And then for Temporal, I run the community events, forums and stuff like that. And when I say I run, I mean, with a lot of help from actual other people who actually know way more than me. So, I've just like kind of the coordinator it's, it's pretty funny. 

Cause like at Temporal. One of the least technical people there. It's just very unusual position for me anyway. So that's my community backgrounds. And obviously I, I have a lot of friends in the community manager space Rosie, Sherry, or the orbit, love people comScore and yeah, Mac written from from, from all these, all these community funders. 

Like it's a very hot topic right now. Okay. Why community? Basically it's the. It's the, it's your users that are identifying with your company and a lot of community forums, a lot of company, community forums, like glorified support channels, right? Like, it's just, Hey you know, you don't like answering questions on stack overflow. 

All right. I'll come to your custom discourse deployments and ask questions to you, but it's basically stack overflow. And what you really want is your users talking to each other, what you really want is your users hiring each other where you really want is your users building on top of you because they believe that you're, you're here to stay and they expect that their association with you will outlast their current employment. 

The problem with this is good luck fitting any of this in an OKR that is measured quarterly because community is a relationship based. Long-term goal rather than a short term transaction. But I mean, I think people are understanding that at least for developer tools, that's the strength of the community is very important for crossing the chasm. 

And I think a lot of people realize that when they cross the castle. Early adopters don't need community. They just need the tool to work. They just need to, like, they don't even need like that great of a docs. They don't need like a functioning, a vibrant job market for that technology. They just need to understand that the tool works and they can use it to solve problems that they have. 

But once you get into the mass market, the majority, then people start deciding on like, Hey, I picked a. The TensorFlow community, because that community is really great at helping beginners. And it's got, it's got plugins and resources for everything that I could possibly want, that those are not core to the technology, but there are a lot of the reasons why people pick technologies, which is there, there's a strong community. 

And I think to invest in that is a very, very long-term game, but it's a moat as well because everything else about your technology can be cloned or out competed, but a community is very, very hard to.  

Yeah,  

[00:21:50] Sai Senthilkumar: and I think you, you have a great point there. That community is, is a feature that cannot be copied and it itself is a moat for you and helps you gain these network economies. 

I think you elegantly structure that in your, in your post.  

[00:22:03] swyx: There is a, there's a VC out there. I'm, I'm really, I'm really sad that I'm blanking on her name because she wrote a really great book. I refer to her in my own blog posts on technical community builders. So, so please go check her out, but I'll give you the thesis of the blog post, which is previously. 

Traditionally community. It was like the community manager at now. The five was kind of like Lewis on the totem pole. You know what I mean? Like, they would be going through the forums and seeing like, okay, like this person asked the question three days ago, it hasn't been answered. Please can anyone in the company help answer this question? 

That is such a pathetic, like, I mean, you know, they were great, but like that, that just shows that the company. Value community, right? It went when, when that is, that kind of behavior is, is around. But then also the community experience or someone's interaction with a company is very fragmented. Like on the forums, I'll be interacting with the support team. 

In, in, in conferences, I'll be interacting with the dev team in webinars. I'd be interacting with the marketing team in sales conversations. I'll be interacting with. It's very disparate thing and there's no sort of central. I guess CRM to manage like the interactions, but also I think more, more, more, more to the point who in the C-suite or who in the VP suite is responsible for community and who's managing that journey through, you know, through your company. 

I think as I think. The community goes from periphery to core. Then a lot of things start deriving their insights from community. Like instead of having community be the afterthought of like, Hey, we launched this product from that's now go support it in the community forums, like actually go engage community users in the forums and help to drive product insights. 

Or, you know, or beta test with like a super user community where there's just some, which is a topic I really like as well. Like the, the idea that every, every company, like I'm, I'm a part of a few super-user programs like GitHub stars Stripe community experts. I am blanking on some of the others, but we're looking into establishing our own as well, because it was very, very helpful for basically not just like recognizing some experts within the community, but also giving them preferential access. 

As a ways to say thank you. And, and I think, I think it's just a very, it's a win-win proposition all around for your most engaged users. Yeah.  

[00:24:26] Sai Senthilkumar: And Shawn, you're wearing multiple hats control your community, fellow Corps experience this, this community eventually become a separate role within, within Deborah. 

Let's say you get to that stage on that.  

[00:24:38] swyx: I haven't seen it personally myself. But yeah, I've, I've seen some companies basically have a head of community or VP of community or chief chief community officer is also a title that I've seen thrown around. But it takes time, you know, it takes it takes the right person to lead that. 

And so I don't have a. Either way of like when the right time is, cause I haven't been through that myself. But I would love to see it. I would love someone at the highest level to say I represent the user and, and not, and this is very, by the way, this is very true for open source dev tools. I represent the. 

Who never pays us a single cent, but contributes so much in tangible value in terms of code in terms of content and sort of like just buzz around the tool. And sometimes that, that person is just not represented in meetings, you know, that, that free user. So, yeah. I love, I love for that to happen right now. 

A lot of that falls on their roles laps and that's a, that's a job that we need to just formally recognize that we do. And I think that's. Yeah. Awesome.  

[00:25:41] How to Start a Community 
 
[00:25:41] Sai Senthilkumar: So the, the $1 million question here, I guess, what are your recommendations for kick-starting a community, especially you guys have done a great job, you know, and you just pay in a year or so. 

Like what  

[00:25:54] swyx: are some recommended. So, first of all, I don't think we've, I think we can do better. Let's just leave it at that. I, you know, I'm sorry that it's just never happy. And I want to acknowledge so many in the Q and a James light. Yes. I think that's Lisa. She I haven't personally connected with her, but I have, I think, I think her posts for like drawing, like sort of community led growth, I think is a, is a very core insight that I think a lot of founders should have a good think about if not actually implement. 

So. So recommendations, you said, right? Two things, like, I think obviously there's, there's this, there's this standard stuff, like have a slack, have a discord by the way, slack versus discord. If you're open source and maybe sort of more Indy go discord, if you're more enterprise Eagle slack, that's just how it is. 

I don't love slack either, but that's just how it is. And then So the two recommendations are this like one. Events are underrated for community. So you saw how into Hashi court, four pillar thing events where different team and community. But I think at a smaller stage, those are the same thing. 

People need gathering points, and this is about the temperature of. Your community as well. You need to manage your temperature. It's not always hot, but there's hot and cold and you need to make sure that the hot wet hot moments are really good. And when I'm not kidding around when I talk about temperature, so, for the sort of communication, immediate theorists in the room, you should refer to Marshall McLuhan who talks about hot media and cold media. 

And I think it's very, very true that there, there are a lot of community and. Forums which are cold. And then there is some community community engagement methods that are hot and you probably want a variation of these things. So slack, so slack is one of those interesting ways in which a cold can sort of upgrade to a warm, but a hot you know, event would be a, would be a meetup, right? 

Like a conference. And that's something that I think really helps. Netlify kicked it up, kick it off. Like we launched the conference where we weren't sure if we could sell up, sell up the seats. But we did. And then, you know, we went from one conference a year, it's a three conferences a year and that just really helps kick it off because once people have shown up and they start talking to each other and they have no choice, but they're kind of, they, they're kind of committed to like being in a room with you. 

One to two days and like sitting through a bunch of vendor presentations it's such a really powerful thing. And, and, oh, by the way, getting your users onstage to talk about how proud they are of using your product. Amazing. Right? So events are super under four, and you want to, if you want to get good at community and get good at events, because then people have a reason to, to come show up and join you. 

But also there there's a bunch of people who are just. Participate. So you need to be decent at a cold outreach as well. What is cold or what is you know, basically building an issue, distribution channel it's basically the stuff that you, you imagine like a mailing list like a Twitter presence, like a YouTube presence. 

I think those are the main channels that I focus on. And obviously blog posts that, that drive. The second part, which I really like to talk about community, which is underrated, I think is content content is the minimum viable community in the sense that all right, you know, you have your products and it's kind of like it exists and people can use it, but really good blog posts would drive people from hacker news straight onto your community. 

And they want to talk about your blog posts and give them a place to talk about it. Right. So, that's kind of how I drive my own. Writing community. That's how I think if you look at like basically the playbooks of the independent creators out there, they're writers for. Or there a YouTuber service, and then you have a community attached to them and the community aggregates around their writing or a content, but then they spread out from there. 

But at least it's a reason to show up that is not about asking for support. And once, once they see a place to promote themselves or actually, you know, get into related conversations that are about what you stand for, rather than just about using you then you have. Vibrant like a self-sustaining community where people just show up because they know that's where they're going to get good feedback or responses. 

[00:29:47] Technical Community Builders 
 
[00:29:47] Sai Senthilkumar: Yeah, that's awesome. And I guess the point here, in terms of finding someone to drive your community, you have a great post you're on, on that technical community builder would love to just double click there and what you mean by that  

[00:30:04] swyx: finding content, right. I mean, you know, at a bare minimum, just curation is really nice. 

So if you think about like the big newsletters in the tech industry, like software lead weekly you know, pointer there, there's a bunch of these like basically aggregation blog posts, but curated by someone fairly knowledgeable. And that can be enough. Like you don't have to write the content, you can just aggregate it and that's, that's still a form of content creation. 

So maybe someone to start there is, is good enough because writing original content that that's a hit after hit is, is a fairly tall ask. Like I write a blog post a week and I only get one hit a year. That's how bad it is. Right? Like that's how bad I am. But, but also like, that's just how brutal the content industry is. 

And at the end, at the end of the day, like what all you're trying to do is provide a space for people to come to, to connect with other professionals who are solving the same. That's right. Like, Yeah, we  

[00:31:00] Sai Senthilkumar: had a question come in here. I'm going to weave it in because it's relevant to the community section. 

It goes. 

[00:31:04] Social Media Managers 
 
[00:31:04] Sai Senthilkumar: What, what if you have a great team of devs and Deborah ELLs, but they don't necessarily want to run social for the company. So the company developer handle is run by an actual social team that doesn't necessarily have the developer chops. And so any, any tips for the not dev teams per se, the social teams running dev communities, or should you just, you know, maybe look otherwise elsewhere and get a technical community builder? 

[00:31:31] swyx: Oh, okay. Social. I think it's okay. So, by the way, for those who don't know, like the post for my, my block wasn't think community building was actually making a case for why your community managers should be technical rather than non-technical obviously that's a. More expensive, higher, but I think it does pay off because people, it really helps when developers understand that the person they're sitting across from actually gets what they're trying to do and can connect them with the resources or the people that they need to get stuff done, or, you know, just have mutual shared interest in. 

But social media, if you have a social professional, a professional social person I haven't worked with those. I'll be honest. So I don't know what it's like. I mean, I know I'm, I'm friends with Rob or gala boob as he, as he's known on Reddit and Twitter. And he's a professional social media guy, and he's funny as hell. 

And people follow him for that. If that's the brand that you're building more power to you, but sometimes that's not the brand of your building. And. And you need to make a conscious choice of like, is this, is this like, yeah, you're getting a lot of internet points for your social media person's output, but is there anything relevant to what you're trying to build as a business? 

And if it's not, you gotta cut it. If it is then great. So I, I don't want to say like, so, I'm, I'm an investor in a company called Swyx. And they're amazing at me marketing but that's just the founders who love developer memes. It's not a separate social team. So I, I, haven't worked with a separate social team to, to understand how to work best with them to, to really give you a good feedback there. 

[00:32:55] Sai Senthilkumar: Super base is great. I love that. I love it reads as well. I guess. Okay. So let's say you have a couple of folks in your Deborah or, you know, you're starting to build a community you're, you're driving that forward. The question that we, that that was most asked, you know, before the seminar is just measuring your devil efforts and then your, you already flopped that slide,  

[00:33:14] North Star Metric 
 
[00:33:14] Sai Senthilkumar: Shawn, and you, you shared your north star metric, but now people continue to talk about how hard it is to measure Deborah also, you know, would love to hammer. 

The point of just why, you know, monthly active developers is, should be your north star metric. We would love and will.  

[00:33:28] swyx: It's it's, it's the one that keeps you honest, right? Because your usage at the end of the day, and there's, you know, you couldn't men measure monthly active deployments or monthly active cluster. 

And by the way, if you're hosting platforms this is a very relevant question. If someone deploys on you and just hosts on you and, but never touches, never logs in. But you know, they're, they're still getting value from your site. They're still paying you is that a monthly active developer? Right. 

But the definition of monthly active developer is up for grabs as well. What most people who I talk to who are in the hosting business agree on is that if they're not developing their sites then they're not active. So, so be more brutally honest with yourself. Like, are you delivering enough features or are you getting the kind of developers who are growing usage and if they are, they should be constantly in your dashboard kind of tweaking stuff. 

Like it's kind of like in how, in how consumer apps, like, you know, the people want. To build a recurring habit of coming back to you at least once a week, if not daily. And for developers that small, probably like at least monthly, if not weekly. So there should be some kind of recurring behavior, but monthly active developers, basically, you want to land within more and more different logos, and then you want to expand within those different logos and monthly active offers is like the one metric that kind of covers that, right? 

So you're not, you don't just want to do evangelism, but then you also want that. W make sure that people don't turn off and you also want to make sure that people are growing within within the, within the companies that they landed. So, I think that kind of covers all of it, you know? It's especially hard, especially for open source because developers are very touchy about telemetry. 

Yeah. As long as the contract is clear and you're, you were steadfastly just only measuring activity and not identity. I think people are fine with that. Okay. So, the one thing, one thing I will say as well, is that like, that, that is obviously the, the point where you know, sales can take over and, and start looking for leads and, and, and all that. 

So that most directly translates to money. The reason I do not. Stare at like, I don't know what my med is on a month to month or even a quarterly to quality basis. I just, I just need to know if it's like up. So it's just like up, it's fine. Like I will have good bites. I'll have bad months. Some of it will be driven by product launches. 

Just by the seasonality or whatever. It's very hard to control. It's a lagging indicator. And so you need to compliment it with a leading indicator. In fact, a bunch of leading indicators, and that's the way I set it up. Right. You have one lagging indicator, that's the source of truth. And then a bunch of leading indicators that should correlate in the long run with the lagging indicator. 

So the leading indicators depend on your. You're a devil efforts ranging from community activity to your contents, you know, views and your, your distribution channel growth, like emails, subscribers, subscriptions, and YouTube, and all that. So your product metrics on bunch days and and NPS scores. 

Well, and I, and I guess  

[00:36:22] Sai Senthilkumar: the point of what are some metrics that you shouldn't necessarily care about as much the vanity metrics, the bad metrics per se.  

[00:36:30] swyx: Yeah, I have a, I think I might get in trouble for saying some of this cause officially I'll get up. Stars is one of our cars for as many for Temporal, but I don't like it. 

It's it's so it looks so sad to beg for a star. Like people should want to start you anyway. And then when people start you, they don't, they're not using you at all. They're just starting you cause they're throwing you a bone. So it's just so meaningless that. Basically the only time get off stars convert into real value is when someone is just completely naive about how this works. 

And they're just like, wow, it's 20,000 stars. It looks amazing. And, and also there's also like the trending page, right? Like if you get on there, some people do discover projects through the trending page and get up. What else have I said, oh, I have. At conferences to scan badges we're going to cube con and like you have to pay for like the device thingy to like scan a badge for customers. 

I don't think those ever work. First of all, it's like cheapens the interaction between you and them. They're just like, okay, I'm just a business card or to you. And then second of all falling up is super hard and very leaky. And those aren't good leads, you know, maybe the leads a week, maybe I'm weak, you know, And if we can, we can do all Glinka Glengarry Glen Ross there, but. 

I, I don't, I don't think, I don't think the transactional interaction of like, let me scan into a database on my first interaction with you works very well. So I'll just, I'll just put it as my 2 cents. I, I did at one time and I have a permanently negative impression of it. What else did I say? 

Oh, GitHub Google analytics, UTM tag. Yeah. So that one, that one I have been measured by as well. And I mean, yeah, I, I, the, the problem with that one is that it, it biases you towards writing a superficial Papa fund, like super top of funnel that may not have content that may not actually have anything to do with the business itself. 

Like as long as it's funny, as long as it's controversial, as long as it's It buys you spices, you twist the wrong thing, like in the worst ways of like how it, how the Twitter algorithm or the Facebook algorithm biases you towards clickbait. And then finally NPS I don't love MPS, even though I did mention NPS just now. 

I know, I understand that it's an industry standard and understand that you know, everyone understands that it's a flood metric, but it's so flawed and particularly developers. Implement NPS. Like I've been on the side where I've been asked to implement an APSI. Everyone knows how this works, and everybody knows that you don't care. 

You throw out the seven, you count the eight nines and tens and you, you know, so like what are you really measuring? What insights are you really getting? Like, I much rather have narrative insight rather than an arm's length. Like, am I doing, you know, approval rating essentially, of, of my, of my job, because I don't think that actually gives me much actionable insight apart from things that trending up or down, which I already have with my reactive. 

[00:39:20] Product DevRel 
 
[00:39:20] swyx: Yeah, did, this  

is  

[00:39:22] Sai Senthilkumar: a question that came in by the way, next it's relevant from an average conversation, but let's say you're never out several reports up to product. What are, what are some good metrics there that you should, that you should care about?  

[00:39:33] swyx: Okay. Yeah. So this is this is the thing I haven't personally figured out that much. 

So what, what I wrote is. If I that makes sense for running devel, which is coming community content, and then product is like a third wheel. Product is one of the things that we're still figuring it out. How do you think products should be measured? 

[00:40:00] Sai Senthilkumar: I think.  
 
[00:40:03] swyx: So there's a lot of philosophy of what product is, which varies from company to company. And that's an unsolved question. And then there's the philosophy of metrics, how you think you should measure the thing. And then we're, we're doing so many bad shots here then, then you're like, how should reporting into product should, should be measured, right? 

Like this is like three tiers down of unexamined core beliefs that you, that we haven't really figured out yet. So, that's why I keep it to tangible. Related to the activities that we're doing feeding into the, the one lagging metrics, which keeps us honest because we can monetize it.  

[00:40:37] Sai Senthilkumar: Awesome. 
 
[00:40:37] Finding Great DevRels 
 
[00:40:37] Sai Senthilkumar: You know, I, and I know we only have 10 minutes left here. I do want to touch on the last point, which is just hiring you know, Temporal, they locked out on, on finding you what were some other folks like you, you know, if you were, you were just starting to hire folks for your devil organization, where would you say. 

[00:40:54] swyx: Yeah, this is a tough one. So yeah. By the way, I get maybe three calls a week from founders asking about how to hire differs DevRel which is like, it's so amazing to me cause I, I kind of lucked into the job and then, you know, I randomly now have a reputation for it, but it's so amazing to me that this is something that's in demand because like, it seems like a fun job that everyone should want. 

What I usually say is true. Think of the ideal user that you have, that you would most like to clone and hire that person? Because what a devil is essentially going to end up doing is cloning themselves by everything that they do in their community and the, every content that they write. They're essentially going to put out signals that, Hey, this is a place that I believe in. 

And if you're like me, you should join me. And that's obviously, you know, had its problems with diversity, but it was a little we'll, we'll sort of leave that as a separate topic. It's hard because when I say. Typically you're the users that you would most like to clone is also your most ideal customer. 

So you don't want to poach your Deno from your customer? Sometimes oftentimes that ends up happening anyway, because usually the customer, you know, the person is a champion at the customer they implemented successfully and they're off to the races and then customer. The person probably believes in your company more than they believe in the company that they work at. 

So they leave to join you. That's how this all happened for us. And for offers there for hire. So yeah, I mean, that's very natural and very organic and if it happens great for you if you have, if you're struggling to get that, and I completely understand, one thing I like to say actually is maybe you don't like, especially if you're like pre series a. 

So I talked to a lot of like seed stage founders, cause I'm an angel investor. If you're pre series a like. You may want to make sure that at least your founder you know, your CEO CTO can be effective. . If you can't do it, why you, what do you expect that? So you can hire someone else who, who, who can do as good a job or better like yesterday, because they could probably do a better job if they're like an experienced speaker, expensed their phone or whatever. 

But if you can not be a PR, if you personally, as the founder, as, as like someone in the C suite, Can I do a good job of evangelizing for your company, then you haven't figured it out. You'd, haven't figured out like the, the core message to pass on to your detriment, to, to execute on. So make sure that you you've, you've done the job first because you're also part of the default team as a founder, like it or not. 

Right. You're the most credible one too. And the one with the most power to fix anything that comes up. So, maybe don't like, if you're struggling to hire your first role, maybe don't hire like someone who's already an established level who comes from a different community, because if you try that and expect them to transfer their audience to you, like NPM installed. 

It may not work because they don't come from that same mindset. They don't have, they haven't experienced the same problems as you. I much rather take an existing engineer who you know, has some skill that at writing posts and stuff and, and train them up because I think that their roles are more made than bored. 

And that's, that's kind of, I.  

[00:43:47] DevRel for Dev Platforms of non-DevTools companies 
 
[00:43:47] Sai Senthilkumar: Awesome. Well, Shawn, I know we have like another five or six minutes. I want to hit on some questions that came in through the audience. There, there was a good one here about how deck row is different when the product is not specifically for developers like Salesforce, HubSpot, or where it's enterprise stats, but you still want a developer community using the APX. 

How, how is demo different there? How would you think.  

[00:44:12] swyx: I haven't worked on that stuff. So please, please take this with a grain of salt. I've seen Spotify DevRels at work, which is fun. Did you know Spotify has their roads? And yeah, I, I understand that. Sometimes you just have a developer that notion has a devil. 

Yeah, there, there, sometimes it's not the core thrust, but you're still in developer community. I think that this one is more straightforward in the sense that like, Hey, come build a business on top of us, right? Like here are the existing users, here's the monetization. Hopefully there is monetization. 

Usually a lot of platforms develop a preference when they launched. They don't have any money to division figured out, which is super annoying. That's what happened to the Alexa by the way. A ton of crap because they never figured out monetization. And and yeah, I mean, basically sell it as a way to. 

An audience that a captive audience that is behind a walled garden that they could not reach otherwise. And here are all the APIs. Here's how seriously we support it. You know, you don't want to be Twitter where you deprecate APS every so often so that no one believes in you anymore. And you know, you want to host hackathons. 

You want to, you want to show examples of early success. You want to show, you want to give your developers. An idea of what is most demanded by your community of of your non-technical users. And then, you know, you want to show the re the revenue potential. Honestly, I think Shopify does an amazing job at this, right? 

You don't, you don't think of, I mean, I think more Shopify has like a dev focused company, but actually they're not like they're selling it. They're selling. It basically rails e-commerce rails to non-technical e-commerce users and the developers come along for the ride and build out the Shopify ecosystem. 

Right. That's a very successful dev ecosystem because Shopify has monetization figured out. So people gravitate to where the money is, which is why, you know, all right, let's apply this to the very app that we're sitting in right now. Zoom launched a dev community with no monetization, right? There's a whole, there's a bunch of plugins. 

Maybe you can use them. No one, no one has any idea. What they are you know, that's, that's not going to be that successful, even though zoom has a massive, like billions of users, audience it's hard to build a company on top of it, like slack launch the fund Companies when there are certain sizes, they launched like a creator fund or like a startup funds. 

So the cave come build on top of us, we'll fund you. And like, you know, we'll turn you into like a legit startup. I don't think there's, there's been like maybe one successful slack based startup, but that's it, you know? So yeah, you gotta really figure out monetization built into the. Yeah, totally.  

[00:46:38] DevRel Tooling 
 
[00:46:38] Sai Senthilkumar: Another good question that came through with some of the tool recommendations here technology that devil's stack, I we've actually been following a company called orbit.love that I believe you're familiar with as well. 

And you know, we're, we're calling it almost a deck as a Debra L CRM that effectively acts as the operating system that rolls in from agents and any other good ones or is orbit the one like, yeah. Well, what were your tool stack be if you were to recommend that.  

[00:47:03] swyx: So it's funny. Cause basically I knew Josh before he started, orbit. And so I'm reasonably friendly with them and recently actually sat them down for like an hour and just. Just like unloaded on them on like, all the ways that I don't think their product is good enough. So it's, it's so fun because like, when you have to actually use the thing versus like show a pretty dashboard, like it's, it's a world of difference. 

So orbits orbits at the pretty dashboard stage right now they need to basically become the definitive source of how every Silicon valley company measures their developer community. They're not there yet because they present with too many numbers. Okay. So, common room is the other one. 

Have you seen it? Have you seen anybody?  

[00:47:41] Sai Senthilkumar: That's  
 
[00:47:41] swyx: a good one. They seem neck and neck. I think common room has has more focus on the integrations aspects. Whereas orbit is maybe more on just general open-source community. I'm not sure. I mean, in the common room seems to have very good enterprise sales for whatever reason. 

It's super weird. I don't, I don't understand it, but yeah, both of these things are okay or good. Like, I don't think they're amazing because they haven't changed my life as a dev community person. Yeah, that's kind of what it's kind of where I leave it, you know, and I don't, I don't have a formal system, I think for, for sure other people are going to have a formalized CRM. 

We run ours in notion. It's, it's manually attracts not automatically integrated. I would love for it to be automatically integrated, but also it's not going to materially change my life because I know who my most active users are. I know, you know, I know, I know who is like, you know, super active and I know who's falling off, like, I don't need a machine to tell me that yet. 

When I may be you know, a much bigger company, maybe like C stage S a CVC probably I'll want to have that so that I can at least have a shared understanding of reality with my team. But right now, you know, there's three of us. It's fine. You guys are doing  

[00:48:49] Sai Senthilkumar: an awesome job and, and, you know, we're, we're big fans of what you're building over there and the community you're building over there. 

I think we're coming up on time, Sean. Unfortunately, I I'm gonna, I'm gonna kick it off. It's obvious.  

[00:49:01] Travis Bryant: Shawn we've been running these COVID office hour sessions for since we all went into lockdown and I think this was the most engaged audience. We've had of any one of these. And it was the first time we had in the Q and a, a request for how to reach out to you directly. 

So what's the best way to to connect with.  

[00:49:19] swyx: Oh, sure. You can find me on Twitter at Swyx. I'll drop it in the chat and you can find my sites where I have like a newsletter and more blog posts, all the blog posts you could possibly want. And Swyx.  

[00:49:31] Travis Bryant: Yeah, great. And what we'll be doing in the recap of this, along with the recording, number of you asked for that, and we'll be publishing that in our recap blog post, as well as we'll just make it an index of all of Shawn's blog posts so that you have an easy way to pop to those are the ones that he referenced as well as the the graphic that he put up there.



2021 Swyx