Startups don't need DevRel. A debate.

So, Jack, thanks for having me.

We play in that same space. And, I think I saw Stefan's tweet that I think you amplified Jack. And so I was like, wait a second. And I ended up writing a little blog post about why I think he's wrong. Even though I'm sure we'll have a cordial discussion about Deborah, because a lot of my friends are in growth marketing too.

We've also done an episode with Michael, the founder of Work OS, where he shares a lot of tips around crossing the enterprise chasm, landing your first enterprise deals, and making sure that you're ready for them. Thanks, Workhorse, for sponsoring the podcast and back to the show. Awesome. So, Stefan, do you wanna introduce us to the tweet that you, and and what you said in the tweet?

And so I would love to hear your you on that Dan I know you wrote a blog post about it. But you can you can kinda go into it like what you said, and then, we'll go into my opinion.

Like me might be a good kind of thing to constrain our space. And I think it's absolutely important for every software startup, frankly, every company to have their customers be advocates. I think that's a that's spot on. It's great. There's way more credibility when a customer says, Hey, WonderGraph solved my problem.

Right? Then if someone who's paid by WonderGraph says, hey, WonderGraph will solve your problem. I don't disagree with that. The issue I have is that customers are busy and they should be because they're probably using your tool to solve other problems, right? To solve their problems.

So, it's awesome if you can get a customer to write a blog post or a testimonial or contribute some open source code. But I think that you're going to, if you rely on your customers to do all of the things that a properly functioning DevRel department would do, you're going to be, I'm pleasantly surprised because they're not going to do some of the grind and the grunt work that needs to happen to make a Dev tool, company successful. Right? And that would be things like really detailed documentation, dietaxis. It has a great, like, example of, like, the 4 kinds of it.

Like, there's reference, there's tutorials, there's how tos, and then there's, I don't remember the 4th one is, but, you know, you can't necessarily expect your customer to do that much documentation. You know, maintaining open source libraries. They're complimentary to your product is something that I wouldn't expect customers to want to do, right? They're gonna file tickets and maybe they'll solve the bugs that are really painful to them, but they won't necessarily, build towards a cohesive vision or, you know, do long term large scale feature requests. Generally, what other things?

And then the last one is, you know, your customer may be sharing well, I'm sure there's more things, but, like, the last one I'm gonna mention, because I do want you to talk, Stefan, is your customers may be referring people to other, you know, to your product, but it's probably pretty unlikely that they're flying their devs out to speak at conferences or meetups or, you know, promoting your tool on their social networks. Right? Because, so that awareness piece is something that again, like a properly structured dev rel program can help.

So the tweet was you shouldn't have dev rows your customer should be a dev rose. I should have probably prefaced this saying in early stage startups like myself and others. You shouldn't have dev rose your customer should be your star. Your dev rose And I think we're gonna agree a lot more because, early on, I think it's imperative that people know about your product and they try it on. And a little background about Wonder Graph.

So we are about 14 people. Every single person is a developer. I am actually not even considered a developer anymore because all the sales and marketing things I do on the growth. I 100% agree with what you said. I think when you're a large organization and you have a lot of people, who's gonna write the documentation? You know? 14 people, we can do it. Our customers help out since it's open source. You're right. We open our whole offering is open source.

So we have to maintain that. And we do that now. But as we continue to scale, it's gonna be a little bit difficult. And right now, every single founder is focused on educating. You know, we build something or we're building something we tweet about it, or we talk to a customer.

And so currently what we're doing, and I think Dan, you're totally right is your customers are busy. And the thing is some customers, can't even talk about your product because they're in, you know, the government space or they have a strict NDA or, you know, they just don't wanna show their face on camera. And I totally agree with you that later on a dev rel will become a handy for that. But I think early on and the biggest problem that we had was like I think Jack said this and like we were just talking about this or I think you did was, sure that a guy with a lot of followers, he might not be a good dev row just because he has a lot of followers. And I totally agree with you because when we were hiring early on the first hire we actually wanted to make was a dev row.

And we raised, 3,000,000 in funding, which is a good seed round, but you need to stay conservative, especially everything that happened in 2023. And people were coming to us asking for a salary of like 290,000 for a dev row. And he's like, but look, I have a 100,000 followers. You know, I can tweet about this. And what I saw was we didn't.

We became the dev row team, and we decided that customers would become the dev row. But then we saw another company that raised a very similar amount to us. Same thing. They had a lot of followers and if you look at who's more ahead because the founders were a little bit more focused and not I think early on, it's really imperative that you don't hire that dev row. And there's an interesting point.

I actually do wanna talk about the conferences. I think and I would love to get your opinion on this. So when we went to graph Q L, conf. There was a lot of big competitors. I'm talking like companies with 100 of 1,000,000 of funding and they just had a lot of money and they sent these dev rel peoples.

I showed up I baked pretzels that look like graph q l logos I wore an orange stew. I made myself a clown to get as much attention possible and when I looked at these large companies that did sell these dev rels they were just sitting there. They're on vacation. They're in San Francisco. They don't really care.

But when you send a founder, he's wearing a freaking space suit and dragging people to his booth. And so that's where it gets tricky for me because I I told my founder, I was like, when we get this big, I don't wanna hire anyone else to go because they won't have the same passion as I'm

But I would love to dive a little bit more into that because that's where I think it gets tricky because when you get big with any company, like, are the DevRel is doing what they should be doing? Are they educator? Are they doing the things that you said? Because if they are doing the things that you said, then that's a perfect DevRel team.

Like, at those early stage companies, I would say under 10, because I co founded a different startup, not FusionAuth, something else where I was, the technical cofounder and I

And, I a 100% agree that for a company that is hired, you know, that has raised a couple $1,000,000, which I agree sounds like a lot of money, but doesn't go as far as it choose to, you know, hiring anybody making 290 k a year or 250 k a year would be, I mean, I don't, I don't, I don't, I don't care if they invented GraphQL. Like that probably is not the right fit. Right? Because the company is too early and we've all heard stories or I've heard stories of people who are succeeding wildly at FAANG. Right?

And then they come into a startup and they don't have the infrastructure around them. They don't have the process. They don't have the resources and they have a hard time. They, they fail. So I think some of it is just startup nature. Right? The other thing I wanna kinda a couple of things I wanna say is there's a great video that I was listening to yesterday by he used to be, like, head of DevRel at New Relic. His name is Jonan. I'll dig it up and

And someone who wants to be kind of a utility player. Again, 14 people in a company, you can't have someone who's not a utility player, right? Like, everyone's gotta be pulling along, until you get bigger and then you can start to specialize. So, a lot of agreement there. I will say bigger companies, you know, I think the well, I will say Stefan that like I have run into this before, and the honest truth is no one cares about your baby like you do.

And I guarantee you when WonderGraph gets to a 100 people, you'll start to want people that fill specific roles that aren't going to be full of passion and willing to wear a suit and big pretzel. Right? Like, you I don't even know how you could assemble a team of a 100 founders. Right? Like, maybe you can and maybe people have done it, but that to me is, you need to just start having structure.

And that means that some people are gonna care less, unfortunately. That doesn't mean they can't do a good job. It just means that they're going to they're not gonna care like you. Probably because they don't own double digit percent of the company like you do too. Right?

Like, there's a lot of aspects to that. To kind of bring it back, I I think that for early stage companies, the founders should be in developer relations. And obviously, if customers can help out and provide additional content or additional, like, social proof, that that's fantastic. I'll pass it back to you to agree or disagree with that statement.

It's like kind of like a cult love that they love for the company and that's exactly the type of company that I want to build is that people go to conference and they wear. You know our logo as a badge of honor because we're creating some cool stuff in the graph go space. But I I mean Jack, should we just end the podcast? I guess we agree on everything.

Now they're doing that. And and it just kind of creates a little bit of untrust and the more followers he has, it can actually hurt your product because you're still trying to experiment. You're still trying to find it out. And if he's creating content on something that maybe isn't gonna be what wonder graph is in the future or it's not born towards a large roadmap, then I actually see it as more harmful than

Or whatever the UK equivalent is. I don't I don't know. But, you know, something that's gonna be huge, it might it might bring in a ton of traffic.

And so, I think, that is true of other if if you hire somebody who is interested in the space of DevRel, but not a big name, they're going to have more loyalty to you as a as as a company than someone that has a lot of other alternatives and kind of picks you, one among, you know, many people they'd be happy to have. You you hit

And both dev rels have actually left and it's because they obviously are not looking. They haven't found the p m f. They've shifted and things like that. And this is where it gets weird for me and I always fight with docs about this because like the more followers you have. Obviously, it's a little bit easier that there might be your ICP and those followers.

The people that your product can actually help and your dev rel can educate. And so what I noticed was that these 2 dev rels were working in GraphQL. And they completely went to a different industry. Now somebody's in databases but not graph databases or anything with databases, just databases. And so he went from GraphQL and hardcore GraphQL to databases.

And then another one was also with GraphQL. And now she's doing something. I forgot what it was. I think it's an ORM. And so for me also I think it's tricky like who is this person's actual audience like I would rather hire a dev row in the dev row space of graph q l that has 2,000 followers.

But these are big enterprises that look at this guy's opinion and they're like wow this is fantastic versus a guy that has 50,000 followers but I don't know who they are. Are they a bunch of junior developers? Are they decision makers? Who are they?

Because there are devs and devs talk to devs. But, I don't think and this gets back to something we talked about before hop on the call, like, a lot of, early stage companies. I think less now. I think the the shine is off the rose a little bit, but a couple of years ago, people were like, you know, if you're a dev tools company, you hire DevRel and suddenly all the other problems go away. Right?

Like, it's like magic fairy dust. And, you know, I I just don't think that's true. Right? Like, DevRel is one other function, and I think it's critical. And the reason why I think it's critical and there's no such thing as like accounting REL or other things REL, I think I haven't actually run across anybody.

Right? Like, when I I did that startup, we were in the food space and we didn't have like food ambassadors or anything like that. The reason why I think DevRel is kind of different, developers relations still a little bit different is because the developers have kind of a couple of unique attributes. 1 is they are allergic to marketing. We've all heard people say that.

They are using things for free. They, the product choice, the process is a lot more involved for dev tools and at the same time, a lot stickier. Right? Which is why I think people are interested in building dev tools because if you use FusionAuth or use WonderGraph and it's valuable to you, you'll talk to people about it. It'll get embedded into your product and it will be, oh, you know, hopefully it's win for customers, but it will definitely be helpful to the company to be sticky along that, the products they're integrated with for a long time.

And so, that all is to say that, like, it's important to talk to devs and have them be a part of the audience that your company is addressing, but it's not, just because you will talk to somebody and you can maybe you have a megaphone and a 100000 followers or 10,000 followers and you can talk to a lot of people, that doesn't mean that it's gonna be right for everybody as you kind of said, Stefan, but it also mean doesn't mean that your product is actually going to be right for them. Right? And so or ready for them. So, I just want everyone listening to this podcast to kind of be aware that, like, just like hire hiring an awesome salesperson isn't magic fairy dust for your startup? Hiring an awesome Deborah is not gonna be magic fairy dust for your startup.

Like, it's all these pieces have to work together, which is why, by the way, shocker, startups are hard.

I think that you need to depend, but I think, Deborah, inherently, especially early stage startups, should be cross functional. And that means that kind of having them report to marketing or having them report to sales is really, going to short circuit some of their, the benefits that a company will get. Because honestly, I view my role at FusionAuth as as head of DevRel to be the voice of the community, which is people who are not paying us. But there are people who are well, let me rephrase that. They're not paying us in dollars.

They're paying us in attention. They're paying us in mindshare. They're paying us in bug fixes. They're paying us in suggestions. Now, you can't pay a mortgage with suggestions, but you can, it can vastly improve your product and vastly improve your chances of success. But I've helped out with sales calls. I've helped out with marketing. I've helped out with customer support. I don't know. Am I am I side stepping your question, Stefan? Because it was a good question.

Like, my job is to educate. And I was like, yes, it's to educate a potential user to transition into a potential buyer and to solve their problem for their organization or for themselves and eventually give you money because you're right. You can't pay a mortgage with PR contributions or, you know, get hub stars. Eventually, somebody needs to put a credit card account and pay. And what I've seen with other DevRels is, like, my friends will say that my goal is to educate.

I was like, yes, that's your goal. And you report to this person, but he reports to this person who needs to tell him, hey, this is how many impressions we made. The blog post, blog post bought this many visitors. This many went through this funnel and, like, this is what we see when we do our own, you know, advocate and it's like at the end of the day like and correct me if I'm wrong like your job is to educate people to be successful with your product. And another one we'll dive deep into is what I've seen with some large companies which I don't like and I can easily differentiate between a good DevRel and a bad DevRel because a good DevRel, he doesn't just push his product.

He says, well, you know, it could be good for this use case. It could be for that use case. And a specific example, I saw a, DevRel from a very large web hosting company, and they were asking, like, can you use this with Vite? And they were like, no. It's totally not possible. You need to use turbo repo. And, obviously, a very large company owns turbo repo. But then somebody on a comment that they're like, you absolutely can. Why why would you say that? And so that's kinda where I see.

I see it in more not examples reframes the question, which was that's why I thought Deveraux goes into sales because if it was really about education, he would have told him, you can use Vyte. You can use turbo. Here's the benefits, and here's the the cons of both. You should go with whatever you would like.

Like I would have 0, I have 0 credibility. Right? So, you need to maintain that credibility. And at the same time, you need to recognize that someone's paying your salary, and that is at the end of the day due to them, you know, a company being able to make sales. So, I do not think if you like black and white, you should get in a DevRel, right?

Like black and white decisions because there's that there's that constant thing. But to answer your, your proximate question, I would say I absolutely have recommended other solutions to people for OAuth problems when they talk to me at FusionAuth. And that is my own kind of personal line. I will, you know, talk about the problem, learn about someone's problem, and make a decision whether or not FusionAuth is a fit. An example that I think I mentioned on the interview with you, Jack, is, you know, someone has a really tight, tight tie into the AWS ecosystem and they don't need certain features and they're willing to put in some effort, I think Cognito is a great fit for an authentication solution.

I think a nice way to play this as a DevRel is to say, hey, whatever you do, you probably shouldn't be doing this. You should probably should be using some kind of tool. Right? Like whether you're using Wonder Graph for GraphQL or competitor A, B, or C, you probably what we do is kind of delicate enough or important enough that you should outsource it. And we think you should take a look at our stuff, but like obviously there are all the people in the market and I think that is a way to kind of thread that needle, right?

Cause you're still like, you're still propositioning somebody, proposition is the wrong word. You're still telling someone like buy is probably the better answer. Because to be honest with you, I've been a developer for a long time. Most of the time using open source library or buying something is a better answer unless it's really core to your product. But you're also say you're also kind of acknowledging that like your solution has its warts and has its strengths.

Right? And the other thing I wanna say is I I a 100% agree to a good role a good DevRel should raise the product raise the product that they're that's paying your salary. Right? And be like, hey, yeah. This is a possible solution I think could work, but they in your example, like, they absolutely should have said, oh, yeah, that worked with Vita, I think you said.

I'm not as familiar with that space, but like, again, to preserve that credibility. Otherwise, you're just a, highly paid public sales engineer.

I'll probably absolutely a 100% work for you. And that's fine because that's what their job is. But, like, you will short circuit your credibility and actually become less than an effective DevRel if your answer is always FusionAuth, Wonder Graph. Like, it just is not a good idea.

They're fantastic. And try them out. So I go when I try them for a subscription, which I actually did by the way. That's why I got this example. And then I later found out that my friend is actually an ambassador that gets money each person he brings on.

And so, like but he actually likes the product. And for me with software, what can be a little bit tricky is I don't know if they're actually using the vitamins or if they're actually using the product. And so when somebody recommends me something, but they're paid also by that company. And this doesn't also has to be DevRel. This could be tech influencers that are, like, tweeting, like, you need to host on Vercel, and then you find out Vercel is paying them money as a, I think as a sponsor.

For me, that bothers me a lot because it's like, your your incense incense I don't know what the word is, but you know what I'm trying to say to to get people to try out the product. And, like, for me also with my own friend, that was a little bit turned away because I was like, wait. But, like, would you have told me about the product if you weren't getting paid? And so that's where also it gets tricky with DevRel is because, like, would you even be talking about, you know, like, Wonder Graph or Fusion off if you didn't work for Fusion off or for WonderGraph?

To be like, hey, I take these all the time. I'm also gonna get 10% if you sign for this. Right? And please take that with a grain of salt. But I do recognize this to my mom. Right? Or whatever. Right? Like, and I think that that is a way for you to make a choice with your eyes open because I think part of the betrayal you felt, maybe betrayal is a strong term, part part of the sting was that you that it was secret. Right?

Like, if he'd said like, yeah. I believe in this so much. I'm, you know, repping these people and getting money for it. That's actually in some ways a different kind of, like, strong endorsement. Right?

Because, but Deborah, you know, I think that good Deborah's believe in the product, that they can, you know, they believe the company that their repping is good. They believe that they're, I mean, they're willing to put their public reputation to some extent behind the company, even if they can quit their job and then move on to a similar company. Like they're still associated with that, that company. So, I guess maybe a question back to you, Stefan, like if you left WonderGraph, right, you know, 5 years down the road, things have changed and your life has changed and you're like, this part of my this chapter of my life is over, will you rep WonderGraph as much as you're doing now? And I would be astonished if the answer is yes, but I'd be interested in hearing.

I would absolutely recommend Wonder Graph. But, like, let's say, you know, there's a different solution and it's better. I would still say, hey, I repped Wonder Graph, but there's also this solution. Check them out. And, you know, for me, I think what you said is important, the disclosure, and it's like, I don't see enough of that on Twitter.

Like, they'll be recommending solutions, and then you find out on Reddit or something that they're actually sponsors of them. And for me, like, I'm not upset that you're getting paid. I'm actually happy. Like, it's fantastic. You know, a company is paying you money to talk about them, but you should at least disclose it, which I think is the key problem that a lot of companies don't disclose that they're actually getting paid behind, you know, under the table, we'll say, from this company to talk about them.

And it's like, I I'm not upset that you're getting paid. I I if you just told me about that, I could have taken that into my, you know, cognitive analysis when I'm deciding a vendor and I'll be like, okay. Well, Dan told me about fusion off, but also Dan works for fusion. But, you know, I trust his opinion. He actually told me that he, you know, he works for them.

Let's give them a try. And then I try if I know and they actually solved my problem, then I'm okay with it. But versus is, like, you recommend fusion off like crazy. I go and I try it out. And then I find out on LinkedIn that you actually work for FusionAuth. I'd be like, wait. Does does he does he even know what my problem is? But, I think I went a little bit fur further from that.

But whenever I comment on an identity related thing, I'm almost well, I I should be I should be a 100% about this. I'd probably say I'm, like, 98%. I lead with, hey, I work for a competitor or I work with FusionAuth, and it's in my profile. And I believe this very strongly. And again, this gets back to this is an interesting thing, like, like there's this question of social capital, and we all know social capital is valuable, but you can either try to maintain it or you can mine it.

And there's, MLMs are example of something multilevel marketing. Right? I don't even know the names of some of them because I'm not really in that world, but there are they're out there. And that's a question a situation where you're basically mining your social capital to exchange for personal capital or financial capital. Now, everyone's financial situation is different.

If that's the way that you can, like, put food on your table for your kids or for yourself, then I'm not gonna say don't do it, but you have to realize that there's a cost. And what those people are doing when they're repping stuff on Twitter without disclosing things, they are, minding their social capital. And that I think has short term benefits and long term costs.

And if there's no bias to it, especially early on like when we have a customer case study we just released and like I felt like almost like tears coming to my eyes because like damn like you know we built this and this guy's talking so passionately about it. And like he has no ties other than he's paying us like that's but he's paying us like we're not paying him that to me. It was like. Okay, maybe you don't need a dev row very early on.

Sometimes in some places, in some spaces, it's easier than others. Right? Like, email is something that is paintable for everybody and is you know, everyone can benefit from better email, GraphQL servers or or auth servers or maybe a slightly smaller market. But the other thing is that I think we when you think about DevRel, especially as an early stage because there's early stage shut up, you're so interested in people like, just just try us, just find out about us. Our biggest competitor is no one knows about us.

But there's like top of funnel activities, where it's like an awareness thing. And then there's like lower down fund where people are like, know about you, but they're trying you out. And I think that's where early stage Jabber all can add a ton of value. Right? Because if people back to the 100,000 Twitter follower example, if people find out about you, but you don't have any documentation and they're reading through they're going through the WonderGraph source code.

They'll try to figure out how to do this thing. They're gonna that that's that's friction. That's sand in the wheels, where they won't be successful as you want them to be. So, lots of people think about DevRel. They think about those dev advocates speaking on the stage or, you know, tweeting or whatever other kind of social media action they can take.

But there's a ton of DevRel focused activities that are further down the developer journey where people can, you know, succeed or fail or churn or bail because they can't find the answer they're looking for, or the developer experiences of sign up is horrible or developer experience of using it is horrible.

So we release, like, 2 or 3 blog posts and SEO. It's a 6 months game. So a blog post you wrote now, and 6 months will finally pick up and people will read it. And before you know it, that post is bringing you 5,000. And, the best tip I ever got in marketing was that, 80% of your success will come from 20% of your effort. So our 20% is Twitter blog posts and YouTube. We tried everything else, but this is where our success comes from and this is what we double down on. Yeah,

And then my final thought, which I've mentioned before is DevRel needs, structure and they need a product that works, and they're not magical fairy dust that will automatically solve all your product and market fit problems.

But fusionauth. Io, and we have a set of vendor neutral articles as well as our documentation. So, if you wanna just learn more about the identity space in general and what identity proofing is or, the intricacies of OAuth, etcetera. That's all there. And we don't put the word I I've I've I've worked very hard to not put the word fusion auth in any of those articles because those are all supposed to be kind of, you know, useful for anybody who wants to learn about the space.