Developer Relations in Open Source - Strategies, Challenges, and Growth w/ Joe Karlsson - EP. 27

Explore the world of Developer Relations (DevRel) with Tinybird's Developer Advocate, Joe Karlsson, in this podcast hosted by Matt Yonkovit. Discover key insights into DevRel, the role of Developer Advocates in open-source communities, strategies for content creation, and the importance of a strong online presence. Learn how to navigate challenges, align business goals with community needs, and understand the nuances between users and customers in open-source businesses. From dealing with layoffs to understanding vanity metrics, this podcast series provides a comprehensive understanding of the ever-evolving tech landscape.

✦ Joe Karlsson LinkedIn Profile: https://www.linkedin.com/in/joekarlsson/

 Chapters:
00:58 -  Navigating lay-offs in tech and advice for those going through it
03:24 - The challenges of measuring ROI in DevRel
07:10 - The difference between users and customers in Open Source Businesses
09:22 - The divide between community and business in open source and what to prioritize
15:21 - The open source flywheel and DevRel content strategy
21:09 - Small communities in open source and growing your audience
24:28 - How to Balance High-Pop and Solution-Focused Content
31:20 - Vanity metrics vs. The metrics that matter
32:39 - Balancing between different types of target audiences
35:31 - Joe Karlsson's first steps in DevRel
37:04 - What it takes to get a job in DevRel
39:58 - Interview tips for Dev Rel job positions
43:25 - Analyzing TinyBird's open source contributions
44:52 - Rapid fire questions 

Checkout our other interviews, clips, and videos: https://l.hosbp.com/Youtube

Don’t forget to visit the open-source business community at: https://opensourcebusiness.community/

Visit our primary sponsor, Scarf, for tools to help analyze your #opensource growth and adoption: https://about.scarf.sh/

Subscribe to the podcast on your favorite app:
Spotify: https://l.hosbp.com/Spotify
Apple: https://l.hosbp.com/Apple
Google: https://l.hosbp.com/Google
Buzzsprout: https://l.hosbp.com/Buzzsprout

Transcript

[00:00:00] Matt Yonkovit: Oh, all right. Woohoo. Welcome to a exciting episode of the Hacking Open Source Business podcast. But the business partner that I have here, Avi is not here, so it's just me. I am the solo host with Joe Karlsson joining me. Hi Joe. How you doing today? Hi. Hi. 
[00:00:19] Joe Karlsson: Thank you for coming to me and Avi will be missed.
But you will for sure make due for. This amazing little podcast. So thanks for having me on. I appreciate it. 
[00:00:31] Matt Yonkovit: No problems. No problems. It's always great to hear from you. A, a globally ranked DevRel personality, 
[00:00:40] Joe Karlsson: is there a ranking? 
[00:00:41] Matt Yonkovit: There's a ranking. You didn't know that you appear in ranking.
No, I didn't. Oh, yeah. So hold on a second. I'll find what rank you are. I didn't know if people know this. There is actually a website that ranks DevRel people. Wow. Yeah. Yeah. 
[00:00:58] Joe Karlsson: That's good. That, that's interesting. It, I have to say it like hits like my lizard part of my brain that validates my work a little bit.
Especially I, so audience members who may not know, I recently was like, laid off and recently just got hired on at a new dev role. But always getting laid off always makes you question your Like your ability to do your work, yeah. Yeah. And times are tough right now.
It's fine. I didn't take it personally, but 
[00:01:26] Matt Yonkovit: yeah. I think right now the world has changed so much. Just even lately we've had a lot of. Changes across the tech landscape where a lot of people have gotten laid off. A lot of, lot, a lot of people are reevaluating what they're doing.
I just chatted with someone also in the DevRel space yesterday who said that they had to postpone conferences because they, so many people were saying we can't travel because we're laying off people. And so they put a more auditorium on travel. So it's a crazy environment and Joe I'm glad you were able to find something and there's a lot of people out there who are still looking.
Any advice for them? 
[00:02:07] Joe Karlsson: Yeah, actually I was just thinking about bringing that up. I yeah. Obviously engineering's been hit really hard. DevRel is notoriously a expensive business unit that is sometimes hard to justify our value, and I've been thinking a lot about what I need to change professionally in order to like, not.
Risk getting laid off again. And I think that a lot has changed, right? The fat, like the hog is dried up a little bit. I feel like for me personally, the things I'm trying to do are trying to integrate myself more fully inside the organization and trying to partner more internally. And so that means like those sales, marketing, engineering and like trying to show your value.
Internally a lot better. So like telling that story. And the other thing too, I think is really focusing on quick wins as well. So like we need to be showing like returns or some movement of the needle somehow. And I think that we have to be less like squishy about the work we're doing too. So I'm just trying to be more like data driven and communicate better internally.
[00:03:09] Matt Yonkovit: That, I think that's always good. And you wanna up your presence, even when you're looking for a job, you still wanna be writing or be active on social. Oh yeah. You wanna show that you have a presence. I think that's really good to continue to do that. But also, yeah it's about bridging that gap between.
The developer side of things and the business side. And I think that actually brings a lot more relevance because one of the things that I'm really worried about from a DevRel perspective is it's hard to show ROI in a lot of cases, right? Yeah. From a business perspective, how do you tie what you're doing in the community to actual dollar outcomes that's driving the business forward?
There's a lot of trust that happens, but sometimes that's the first thing to go right. 
[00:03:48] Joe Karlsson: Yeah. And Matt, I'd be curious to hear your perspective on that too, but I'll give my 2 cents on it and I'd love to hear your thoughts on like how to show ROI and DevRel right now, or like in dev, like developer marketing.
But for me it's not about like I for sure track things like conversions, like how many people converting to free or paid tiers for my content, like number of hits and clicks and that sort of thing too. But I feel like the biggest thing for me is just Output consistency. Cuz I think I have a good hunch on what's going to do well or not do well, but at the end of the day, The stuff I can't predict is gonna do well and take off and stuff that I think is gonna do well.
It's just absolute bombs. The key for me is just like to make things consistently. And then tracking secondarily, like all of the performance metrics you'd expect around, like how engaging your content is or whatever, I don't know what how are you, like how are you 
[00:04:39] Matt Yonkovit: tracking that these days?
So I think that there's a couple different things, right? So the first one is the closer you can get to actual users that are either entering the ecosystem because of your activities Yeah. Or you can tie directly to new people that you are reaching to. That's the important thing. Obviously the nurturing of it, the existing community is critically important.
But depending on the company, there's a lot of times when you have a community team that's focused on the nurture side. And from a DevRel perspective, a lot of the activities that I typically see are more externally focused to get people interested, to get them inspired, to have them try new things and bring new people into the ecosystem.
So I've been working quite a bit with trying to track. Which parts of my content are actually driving eyeballs to either the website or even signups or downloads. Yeah if I can track, like I, I was at FOSDEM this year and at FOSDEM we launched a new website and we had a couple hundred people who visit the website.
That's purely based on what I did. Because it's like a new website, no one else knows about it. I went to this conference. Yeah. I told people to go to it. They did? Yeah. But I'm also exploring using links that can be tracked. So whether you use Bitly or yeah.
Other things, I'm using our own tool Scarf. You can see how many people are clicking on those links. And so if you use those to shorten your links and as you start to provide content on social and everything else, you can start to track. Oh, these people are engaging with it because what I've also found is, Twitter and LinkedIn lie about the number of clicks that you get.
[00:06:18] Joe Karlsson: Yeah. I know they all have an incentive to lie a little bit and you're like, that doesn't feel right, but okay. 
[00:06:24] Matt Yonkovit: Sure. Yeah. Yeah. And I think that's where it's definitely a little different working it across the different social channels. Some do things better than others.
But I think that's really the ultimate thing is how do we prove that we're driving additional people to enter the community in the ecosystem We want to grow and expand the ecosystem and make sure those are there in the ecosystem are happy and content. Yeah. But when we talk about community, too many people think contributors first and not those users.
And that's what drives me nuts. 
[00:06:57] Joe Karlsson: Yes, I totally agree. I totally agree. Yeah. It feels like it's Yeah I a hundred percent agree. I totally agree with you on that. And I get the business like, wants to, like push on that too, but I feel like, you have to nurture, you have to nurture those users.
[00:07:10] Matt Yonkovit: Yeah, absolutely. And I think that, users and customers are not the same thing. By the way. That's a key thing, right? From a business perspective, any business that's a commercial entity that's trying to commercialize any sort of technology, whether it's open source or not, they want customers from a business perspective.
Whereas, from a DevRel or a community perspective, we're more on the users. Yeah, 
[00:07:34] Joe Karlsson: and are you defining that a free tier user versus a paid tier? 
[00:07:38] Matt Yonkovit: Yeah, paid, you can view any sort of user as a. Customer, if you want to view it that way, but ultimately there's this divide, right?
I think that's where the divide needs to be crossed over, where you've got the divide of the free, the open source people. Let's say, you're downloading, software, you're using it, you're not paying anything. You don't know who those people are necessarily.
No. That's where DevRel tends to start. And from a sales perspective, sales is I would just want to turn those people into paying customers. Yeah. And that's where that kind of alignment needs to happen and needs to get better at. 
[00:08:23] Joe Karlsson: I, yeah, I totally agree. I, so it's are you saying it's it's like folks who are like, I don't know, like lurkers or pre like pre-sign up or pre-trying it.
It's like just kind 
[00:08:35] Matt Yonkovit: of so think about you work at MongoDB, right? Yeah. So there were probably, let's just, I'll just throw out some ob obtuse number. Unique numbers. Yeah. There's a million people. Who use MongoDB? We'll just say there's a million people. Okay. Yeah.
There's 5,000 customers. I see there's 999,000, or 990 5,000 non-paying customers or users, right? Yeah. And so adding people to that is important. I actually gave a talk on this at FOSDEM at the community track. Let's see. I think I got a neat slide on this too.
Let's see. Ooh I look at this 
[00:09:20] Joe Karlsson: slide's the perfect medium for a 
[00:09:22] Matt Yonkovit: podcast. Yes. It's so we're also, do we do videos so people can watch on video too? Oh, good. There you go. I won't pick my nose. Good. You can, if you want it could, it generate some views, right? Yeah. We're all about the extra views.
Yeah. So let's see. Oh, geez. Look at this. What did you do? The window and. So here's my, here's one of the slides that I've got here. So hopefully you can see that maybe it's, I can. Okay. Anyways, whatever you can see it. Alrighty. So this is where there's like this kind of like divide, right?
So you've got active contributors, code contributors, slack users, all this stuff. And then you've got the business side, which is totally into the number of customers. The ARR, the expansion, like the, there's these two sets of metrics that are not necessarily aligned with one another. Yeah. And I like these two little graphic graphs at the top because it's like the business folks versus the woohoo we're gonna get crazy here folks. And not to say that DevRel, people can't be the serious. But yeah it, there is this kind of like difference.
Yeah. That 
[00:10:38] Joe Karlsson: people are, I agree. Yeah, I totally agree. And it's either like about helping other like marketing and sales understand how these help them or some, I've been an org sewer where you have to like just protect those numbers too, or have Prevent sales and marketing from smothering your community and growth and slack growth and like just community growth in general, which I've also had to like, put my foot down on due to, I feel like it's a little bit of both.
Yeah, we're advocating for the developers and we're like, and we're helping like internally sell that these like concepts of how this helps them and protecting them. 
[00:11:17] Matt Yonkovit: Yeah. And I think that, from a standpoint of how most. Business folks think that this is gonna work is, like we've got the influencer marketing so Joe, by the way, Joe, you are number 37.
Look at that. 
[00:11:36] Joe Karlsson: You're number. If I had to rank myself, I would not look at that. Look at it. I would not have put myself that high. 
[00:11:40] Matt Yonkovit: That is, you are number 37 on the DevRel developer advocate. Top 100 influencers in DevRel so just an fyi, you have a, you've gotta live up to the reputation.
[00:11:54] Joe Karlsson: Yeah. I guess I, I guess so. I had no, there's some great people on this list too. That's 
[00:11:58] Matt Yonkovit: amazing. Yeah. So there you go. But so here's the thing, right? So as an influencer, you are this influencer over here, and the expectation from a company perspective is, You're going to generate a certain number of followers who are then going to watch you and of those who are watching you regularly and subscribing to your content, they're going to start trying and doing whatever you tell them to do. And then from that, the expectation is those will eventually turn into a sales perspective. Yeah. And so if you're, and how I tend to look at dev.
Perspective is my job from a developer advocate perspective, if that's my role is to create other external evangelists who will go out there and do what I do and help amplify the audience. Because Joe, even though you're number, you're voted number 37, you only have the ability to reach so many people, right?
Oh, totally. So you might, let's say you have an audience that's 10,000 people, that's still a drop in the bucket. Oh yeah. So if you can get 10 other people who also have an audience of 10,000. That's going to scale more. And so that's what this 
[00:13:05] Joe Karlsson: is. Anyways, actually, I've been thinking about this a lot too, cuz I totally agree.
I like, I don't have a massive following or anything, but if I wanted to, like my growth, personal growth goals are aligned with the business a little bit too. And I feel like a big part of like my personal strategy recently has been doing like partnerships, like jumping into existing communities too, and like helping facilitate those.
Yeah, partnerships 
[00:13:29] Matt Yonkovit: are a big deal. 
[00:13:30] Joe Karlsson: My personal goals for growing my audience are aligned with like the company and workforce goals for growing their audience and their sales or whatever too. And a big part of like how I've been doing that is through partnerships. Like just meeting with existing communities or exist.
I mean there's a million conferences or like product specific stuff or just programming language communities, but like getting involved with those I think is really in a meaningful way. I'm sure you and I know and everyone that's listening is like, has been to like, people on partnerships are just come in and just try to do a sales pitch or whatever, and that's like a massive turnoff.
I can think of keynotes, like paid keynotes have been to some big conferences such 
[00:14:08] Matt Yonkovit: Oh yeah. Pay keynotes are the worst. Yeah. 
[00:14:10] Joe Karlsson: Yeah. I totally agree. It just feels like a sales pitch and it's it shows like a lack of understanding of how to partner with those communities. 
[00:14:18] Matt Yonkovit: By the way like when we, when I was in charge of the conference at Procona, it's so we did the Procona live I banned.
Paid for keynotes. Yeah. I'm like, good for you. Boom. It's ridiculous. Nobody wants 'em. Yeah. So you 
[00:14:34] Joe Karlsson: kinda have to do it to protect the community. And I get it, like pays the bills and that's really helpful too. And and I know some companies do a great job of that too, of like getting a paid slot and Helping teach the community in a respectful, non-salesy way.
And by the way, there's I use the word salesy and there's nothing wrong with selling, like selling it, it's just, I think we have a problem with bad selling that ends up just coming across this unhelpful. 
[00:14:56] Matt Yonkovit: I'm gonna put on my, I have a sales hat by the way. It's my if you like 
[00:15:00] Joe Karlsson: people super villain.
That is good. That is, that's 
[00:15:04] Matt Yonkovit: my, this is my business hat. I wear this whenever I talk business. 
[00:15:09] Joe Karlsson: When we love salespeople, they do a really hard job, but we're like helping them out. It, it's, everyone wins. But yeah, it is, it sucks when it like comes down to just like pure bad sales tactics at a conference.
[00:15:21] Matt Yonkovit: And so I think that, not all sales are created the same and yeah, I think that depending on how you approach the community there, there's a organic kind of like sales pipeline and the ability to tap in. And a lot of people talk about the product growth flywheel. I don't know if you're familiar with the 
[00:15:36] Joe Karlsson: product growth.
I am not. No. It sounds. I like the, I like that 
word 
[00:15:39] Matt Yonkovit: flywheel. Okay, so here again we'll slide up this, shall we slide this up? Not that I am like, Trying to do a slideshow here, but when we talk about like this it's could be a little harder to read. It could be pictures worth a thousand words.
Yeah, I get it. Oh, wow. Look at 
[00:16:01] Joe Karlsson: that. I know, isn't that the words? He's making the screen bigger for a slide. But on Google Slides, all it does, it just makes the header smaller or makes the actual slider 
[00:16:10] Matt Yonkovit: opposite what you want. So anyway, so let's see. Okay, so here we go. This is my really shitty graphic.
Okay. So I am not a graphic designer, but this is the open source flywheel. But you could find it like, if I go search for product growth flywheel, I'll probably find a better example. Let's see, product growth. Fly. Ooh, 
[00:16:31] Joe Karlsson: a nice 
[00:16:31] Matt Yonkovit: clicky keyboard. Oh yes. Love, love the clicky keyboard. Let's see.
So images, let's pull up a, oh, look at this. We've got the product growth flywheel. They're all over the place. How about the HubSpot one? There we go. So let's let's go ahead. We'll open this in a new tab. Open it. New image in a tap. There we go. Look at that way. Beautiful. Look at that.
Isn't that beautiful? That's beautiful. So the idea is from a growth perspective, what you have to do is attract people, then engage them, and then delight them. And so as you do that, those strangers. Who are attracted will become prospects who will start to look at your stuff and then they'll become customers and then they'll eventually promote you, and then it will, become this kind of like cycle that continually propagates 
[00:17:17] Joe Karlsson: itself.
Yeah, I love this. I can I say too so I'm a content focused developer advocate. I make videos and podcasts, whatever. But my, I, my biggest thing I'd say to get people like, Closer to the center of the flywheel is to teach, not 
[00:17:31] Matt Yonkovit: sell. Yes. Helping will sell. Selling will not help. Yes. 
[00:17:36] Joe Karlsson: And can I give an example?
I love to imagine you're at home on Facebook and you have a friend, a realtor friend, and every day they're just posting a new listing. They're trying to sell on the market. Like I, I don't know about you Matt, but I'm gonna unfollow that person so quickly. And for me but imagine that same person, right?
And instead of like just posting their listings every day and trying to sell those, that way they're making like videos or blog posts talking about here's how the market's changed. Here's how my predictions the next five years for the local, I don't know, San Francisco market or whatever, wherever they're located.
Or Hey, I helped this person selling the house, renovate their bathroom and it increased the value of their home by $15,000. And here's the step by step of how they did that. That's like engaging content that's helping me. And it's something I might actually end up sharing and I'm something I'm like less likely to just unfollow or check out from.
[00:18:28] Matt Yonkovit: Yeah, and there, there's a balance there though. Even that stuff, if you do too much of it can cause issues. 
[00:18:34] Joe Karlsson: I totally agree. Yeah. If the, they're doing so much content posting that they're not even talking about any of their listings at all, that is also an issue. But even for me too like I sometimes I even use like, It, it's like a balance of content, like keeping people engaged, but also talking about it.
I know James Francos canceled now, but like back in the day he was like the selfie king and he said that he used to like, use his selfies to engage content, like users to follow his art and his poems and that sort of thing. But he was like balance his art that no one really cared about with his like, selfies, which people like super engaged with.
And I feel like it's a little bit with like businesses too, right? Like we need to like, keep the end goal of making money in mind. But we also like needed Kind of balance that with like more fun content or things that people are actually gonna find value from. 
[00:19:20] Matt Yonkovit: Yeah, and I think that's a really good point, is the balance is key.
Now, there's some arbitrary numbers that I have heard from quote unquote experts that say for every piece of like self-serving content, you should have four or five pieces of something else. Yeah. Now I've heard that three to one. I've heard that four to one. I've heard that five to one pick your poison, but the idea is yeah, art you should be sending more out that is not really specifically focused on your sales pitch than you, you are, trying to sell. 
[00:19:52] Joe Karlsson: Yeah, I think it's a good point and it depends on your audience and it depends on where you're at. And I think that involves testing to figure that out, which you should do, right? I don't know. I personally, I like my content strategy involves a lot of just like trying stuff out and seeing what works.
And yeah, if people are engaging in it, cool, keep like working in that direction, if not, Don't, yeah. And actually, and I was talking about this before we started recording too. I've been like, I have this soft idea. I've been like trying to work out, but about treating con, like open source content development, like an open source project.
Because Matt, like the, one of the biggest things I've seen people struggle with working with this flywheel and getting people engaged too, is it's a lot of c. Companies building features or content for things that no one asked for, which I'm sure you've seen a million times. I've seen a million times.
Yeah and some executives got an idea and they're like, this is build this. But like for us, we have the advantage of having a close relationship with a community of users who are like using our product or like using similar products. And going to them. And I don't just asking Hey, what do you guys need to see?
Or what are you guys struggling with? And I usually provide some like options for them to choose from. I'm thinking about this, or this. Which one do you care about? And to work in that direction. 
[00:21:09] Matt Yonkovit: Yeah. And what's hard though is not everyone is at the same spot. And I think where I've seen some people really struggle is when they're looking at like smaller like communities or communities that are less engaged.
You've still got to. Push the limits of what you're doing and try different things. But in a lot of cases you might not have the audience to necessarily ask. Absolutely. So it's a lot more trial and error early on. Yeah. And that's, that can be scary. I can tell you that, from my perspective the audience, the current company that I'm at, we're relatively small.
We're a startup, right? Yeah. And we've got a growing list of people who use our software, but it's not a huge, list. And our target audience is like founders and executives at these companies. If there's, let's say we've got how many open source companies are out there in the world?
Let's just say that there's 10,000. I don't know, I'm just gonna throw 10,000 out there. They might each have a hundred people, but there might only be like two executives at those companies. That really care about like the growth and the things that we're, talking about and pitching.
And it's not to say that everyone couldn't care about it, it's just that those who are like really willing to engage might be a smaller audience and Oh, 
[00:22:30] Joe Karlsson: totally. And founders tend to be more like super busy. People aren't just gonna spend Yeah. Five minutes answering a poll question on a random Slack channel.
They don't join three months 
[00:22:39] Matt Yonkovit: ago. Yeah. And I've talked to some people about this, which is funny because they're like, ah, like I, I talked with Gabe Weiss from Google yeah. Year and a half ago, and he was talking about how. Even like internally he'll do a video or he'll do content and it's oh, it's got, 2000 views.
And, the folks there are like, you only got 2000 views. What the hell's wrong with you? Like you sh all of our Google content has a million views. Why does yours only have 2000? And he's because the database audience is so much smaller, so different or, and so I think that it's hard to sometimes compare yourselves.
To other people in the space and then feel like maybe you're not doing a good enough job or maybe Yeah. Beat yourself up a little bit like that. Yeah, 
[00:23:25] Joe Karlsson: I totally agree. And it's hard to be like in the open to not compare yourself. I'm guilty of that too, and I agree. I'm, I just started a new role at Tiny Bird again, as a database company too, which I agree.
Definitely more niche. Yeah, it's a. And like it's got a, it's a new company, right? So it's got a smaller audience. It's new in the us so yeah, trying to grow that can be really hard too. And I agree. And I think that gets to a point too about like data, like how big your sample sizes start taking data seriously.
And it, again, this is more like an art than a science. I don't have any numbers around this, but for me personally, like small communities, it was. Like you can get insights into like page views and that sort of thing too. I don't think it really, you're not getting meaningful data to like actually glean anything too useful.
So like getting too stuck on like this blog post and perform well, that doesn't mean that strategy won't work or like abandoning it. It's like trying to take it into a grain of salt to maybe evaluating other things or trying it out for six months even. And seeing if it works or not. 
[00:24:28] Matt Yonkovit: Yeah.
And so I actually did. So when you talk about blog posts, this is an interesting topic and when you're looking at this content, right? So obviously different audiences, different sizes. Yeah. So when I was working at Procona, we had about 300,000 viewers of the blog every month. It's a pretty decent size.
Yeah. But what was, and that's good. Yeah. Yeah. What was funny was the blog there it would have some content that would get like 20,000 views in the first, like three days. And then there would be other ones that only got like a hundred views in the first, few days. Being the, crazy person that I am I actually we made a contest for all of our engineers and people who were writing content where we would give them a championship belt at the end of the year.
If they had the highest ranked or the highest viewed content, but how we phrase this is there was two different ways to do this, which is the first 30 days worth of uses. And then versus the entire year's worth of views. From a consistent basis, because there's two types of content, in my opinion, when you're writing blogs.
The first one is really about making sure that you have that pop, that initial kind of oh my God, we need to go get this. It's the Hacker News effect. Yeah. Yeah. Like it, ah, we wanted to get this on the first page of Hacker News, and then you'll get 25,000 views. In one day. But then there's the other ones which are for SEO purposes or for people who are actually searching for a solution.
Yes. And you know what those trickle in? Yeah. But those trickle in ones are sometimes worth more than that. Hacker News pop. Yeah. Because when you know, like you have a topic like, oh, you know how to how to, beat the system on an AWS and reduce your bill by 50% or whatever. I'm just gonna throw that out there.
You throw that out there and maybe somebody at Hacker do goes, wow, that's awesome. What's that going to necessarily do for your company? Unless you are Corey Quinn and your whole company's about saving, money on aws. You're probably not gonna do a lot. But it's interesting.
It will get people to take note. They'll know who your company is. Yeah. They'll check out that content but they're not going to necessarily go to your database. Company. Yeah. For that. Yeah. But if you have a, how do you deploy this using no JS and JSUN objects, blah, blah, blah, and people are actually searching for that as a solution.
Then that is probably going to generate a higher conversion rate to actually become a user. Than that, that higher pop content. 
[00:27:00] Joe Karlsson: Yeah, it's like your product is showing itself a problem. Yeah. And I've seen a lot too, like an engineering team will change some process that makes their engineering process better, but it's not related to your product at all.
So people are coming for like your database company and they come through your like C I C D content. It's okay, that's cool and interesting and and it's, I think it, that's definitely good awareness play, especially if you're like looking to just grow your name a little bit. But you're right, it's not gonna lead to bottom dollar and you need both. You need both. I think you need both. It's, I definitely would focus more on the long tail content. I think in terms of if you can make a piece of content that is useful for the next three to five years, like that's a great return on investment and it's actually solving some real problems.
Yeah. And I was gonna say one other thing too about the the total views on content too is Sometimes the best content is gonna get lower views, but it's more niche and more targeted. So like you're solving like a more advanced problem that is like very specifically solved by your product.
And it's only gonna be targeting like a hundred people, but those a hundred people are gonna be like really good use cases 
[00:28:04] Matt Yonkovit: for 
Your product. So this is where I'm starting to evolve my thinking and this is like a difficult problem to solve, but it is not impossible. Yeah. And this is where I've been experimenting quite a bit, is.
Those views, I wanna know how many of those views are from people who actually tried or used my software. Yeah. And that conversion ratio, the guess is on the ones that have a hundred, the con, the conversion ratio from those who view versus the and try. Yeah. It's probably gonna be higher.
Yeah. And I've been experimenting with that internally yeah. On some of our content. And some of the other things we're doing which is cool. Because like obviously hey, it's great I got 10,000 views, but if those 10,000 views, none of them became customers or users or just read it and said yeah, this is in mildly interesting.
That's different than if you have a hundred people who read it and say that is spectacularly awesome. Yeah. 
[00:29:01] Joe Karlsson: That's a, that's been like a part of my like post tech weird market thing too is like focusing more on one oh or 2 0 1 type content. More like intermediate to advanced content.
I feel like most companies, like there's a ton of content for getting into whatever market segment you're in, but I feel like products are gonna be differentiated on their advanced features. So I feel like I'm like trying to lean into like more advanced subjects, more like recently like digging in a little bit further, which has been a shift personally for me.
[00:29:31] Matt Yonkovit: Yeah, and sometimes you've gotta make that, like shift, right? Yeah. Sometimes you have to look at that and I think that, so what I was talking about here, let me. So you got a slide? I have a, I have an image. I have an image.
I have an image for everything. Ok. I have an image for everything. So yeah, so this is my the, I just made a blog specifically to talk about like the reason you should be tracking, production users. So Joe this is a classic usage problem. And by the way, this is not those listening, this is not, me trying to promote my content.
This is just happens to be Joe like, like throwing me softballs and he is oh, what about this? And it's oh look, I have something on that. Let's talk about it. So for those who are listening and not watching, what I've done is I put up this graphic that I just did about how do you correlate.
Content metrics with your data. When you look at monthly activities, if you've got, 2000 unique page views and 5,000 people who clicked in and, a hundred people who downloaded or, 50 back links to the thing, like all of those are marketing metrics or metrics you probably track from a community or content perspective.
But the question is, which one of those lead to actual people downloading and trying the software? Yeah. And based on this, what you'd probably say is, Just on the number of unique views, blog three is the most popular blog. So obviously that is the one that you would say you want more of that content.
But in this case, blog one has way more downloads than blog three. So people who actually tried the software, more people did it to the one that appears to be like less popular and yeah. And that, 
[00:31:16] Joe Karlsson: that's not to say like blog three isn't valuable, it's just it's different.
They're solving different problems. Yeah, 
[00:31:20] Matt Yonkovit: exactly. It's different. Yep. It's different. And I think that's where, from a, an overall perspective, right? I think. We have to evolve our thinking. And one of the things that I've been really harping on is there is so much focus on metrics that are vanity stats.
It could be the number of, people who visited the blog. Yeah. It could be the number of people who joined your Slack channel. The number of GitHub stars. I hate GitHub stars, by the way. Hate you. I hate GitHub stars. But all of those are vanity metrics.
[00:31:57] Joe Karlsson: Yep. Yeah I agree. And it's not bad to track 'em, but like I think the key for me is what's your primary metrics and and why are you tracking them? Because I think what you're tracking and like you're gonna have the biggest impact on like, Where you're putting your time, I think affects the quality or like what you're doing with that content too.
And like you could, yeah, you could do something that's like really salesy and like really gonna push hard to convert people and it's just gonna turn people away or push people, sign up for GitHub stars and forget that your community isn't growing, but like it's getting a ton of hits and throwing, but It's not helping much.
[00:32:32] Matt Yonkovit: Yeah. It's really about connecting the dots with what matters to the business and what the strategy is, yeah. It, there's a lot of companies whose strategy is Fortune 500 companies. That's who we wanna sign up for. And, there's other companies who are like, we just want, like individual users.
Those two have completely different strategies and ways to measure the effectiveness of what you're doing. And you need to figure that out and then align content and everything else towards that. Yeah, 
[00:33:05] Joe Karlsson: I totally agree. Yeah. I, developers at Fortune 500 companies are like the the golden goose for everybody.
Everyone trying to like, sell new products or whatever, and it's super hard to reach 'em. They're expensive. It is. They're inundated with everybody trying to get their attention. But yeah, but I get it. Some makes sense. I get why those are the big, those are the big whales you want to go after.
Yeah. But yeah, it's, you have to be very careful that, and balance it with other prospects. 
[00:33:33] Matt Yonkovit: Yeah. And I think that one of the big, interesting things is, as you start to explore, any one of those, you risk alienating the others. Yeah. And that could be okay. Yeah, that's 
[00:33:46] Joe Karlsson: true.
Interesting. Yeah. I totally agree. I totally agree. Yeah. It, and it just depends. It depends. Yeah, I think, yeah, I totally agree. I totally agree. I know when I was at Mongo B, we like, I've always felt like I had to walk this like fine line cause like we wanted to show like Mongo b I think people's perception of it was like, it's not for banks or governments and we tried to show like we are enterprisey and we're less just like a fun point for developer, but we also still wanted to court our existing.
Like indie developers working on startups or just hacking at home. And that's a, it's a fine line. It's a fine line. You're like talking to two different audiences. And sometimes, if you have a team that's big enough, you can divide and conquer and start talking to both those audiences, both at all at once.
But if you're like a solo dev row content developer at a company like. You may not have the bandwidth to do that, and you kinda have to pick. Yeah. I 
[00:34:40] Matt Yonkovit: think you focus on what matters the most to the company where your target is, if you're gonna focus on banking for instance.
Yeah. Especially in the database space, there are certain things from a compliance perspective, a security perspective, that are very like, specific to those sort of customers, which may not matter as much to your websites or, gaming companies or fill in the blank companies.
And so I think it's just making sure that you're aligned to that. 
[00:35:10] Joe Karlsson: Yeah, I totally agree. And for me, that goes back to my new, my, my point about if you're Endeavor or your new company or like you're nervous about your role in Dev I think it's more important than ever to make sure that you are aligning with the o like the bus, what the business cares about.
And like talking to that audience and like making adjustments if you need to. Yeah. So 
[00:35:31] Matt Yonkovit: Joe, how did you get into DevRel in the beginning? 
[00:35:34] Joe Karlsson: Actually, yeah, it's a, I'll give you the quick version. So I I taught at a boot camp in Honolulu, Hawaii for a couple years, which I loved. And it was one of the top 10 best developer boot camps.
It was really fun. I ended up I was working like 80, 90 hours a week. Ended up burning out and then I started working for a large e-commerce company that sells lots of TVs and refrigerators on Black Friday here in the us and became one of the front end technical leads for that site. But I really missed teaching.
So I started to do live coding, YouTube videos, blogging, speaking at conferences, that sort of thing. And I was trying to just spread the word about how cool this big blue E-commerce company was, and like the cool work we were doing to everybody and my boss got mad at me cause I was spending more time speaking at tech conferences and writing content than doing my job as a front end technical lead, which totally fine, I get it.
But it was just like this, like burning like feeling in my gut that I need to like, go out and just like talk about these things I was learning and working on. So I decided to find a job where I could do that full-time and that's where I started. I got. Picked up into Dev Rail. It helped to have been building an audience prior to getting to Dev Rail.
And if anyone's listening to this and like thinking about getting into Dev rail too, I like, I think you can get in with no experience. I don't think you need to be good at public speaking to be good at Dev Rail, but for me it did help to have a little bit of community building and Plan a couple hackathons, do some writing, get some traction at some conferences, that sort of thing to get started with it.
[00:37:04] Matt Yonkovit: ,What do you think is the personality in, in, from a DevRel perspective that people are most like likely to succeed with? 
[00:37:11] Joe Karlsson: I think the perception is it's like people like you and me who are like, good in front of a camera and can like speak good. Or not afraid to. I suck at that, but that's okay.
I know I. Me too, honestly. And but I have so many friends in DevRel who have 20 followers on Twitter or what, like on YouTube? Don't have a YouTube or like online presence at all? I think it's a misconception because people see us and I think that they think that you need to be like loud to do that.
But I, you think you can be successful in other ways too? I think there's tons of people behind the scenes who are making amazing content that is moving the needle but aren't as it. As much of a face, 
[00:37:54] Matt Yonkovit: what I mean? Yeah. I think it's the difference between DevRel and Influencer, or influencer DevRel.
I agree. Agree. Yeah. It's like that idea of you have the person who's the face, but then you have the people who are actually doing all the work. Yeah. Actually producing the content, and when you're a smaller company, are doing a little of everything, of course.
But I think a, as you get bigger, there are people who are gonna write technical documentation tutorials, give, training classes, et cetera. Yeah, 
[00:38:20] Joe Karlsson: I totally agree. And like I think those people should be in there. I think if you are looking to get into DevRel today, I like if I'm evaluating like a junior DevRel person I'd definitely like to see some sort of effort into either content creation.
I don't, it doesn't have to be big, but I wanna see like you writing or like doing some sort of community building or planning some sort of events in the tech space or something or involved with an existing open source project. Like I want to see some sort of Interest in that area if you're gonna be coming into Dev Rel.
And again, not to say that you don't have to, I think lots of dev rail folks have come in without having that experience, and I think it's still possible. But especially in this labor market, I think it'd be good to have a little bit to show to come in. Yeah. What would you what do you look for if you were gonna hire a junior dev person tomorrow, what kind of 
[00:39:13] Matt Yonkovit: things would you be looking for?
Number one thing that I go look for is, can I find them online? 
[00:39:17] Joe Karlsson: Yeah, I always get a Google first of course. 
[00:39:20] Matt Yonkovit: Exactly. So I've hired a couple of dev people with very limited experience just based on what I could find in GitHub and yeah, YouTube, like I've said to recruiters like, I'm gonna hire that person right there without even interviewing em, just based on what I've been able to see.
And some of these people only have three or four things out there. Yeah, but if you could find them and you're like, wow, I really like the passion or how they presented something, it really goes a long way to showcasing that. So there's that. But by the way, did you ever think about so I know like you, you were in between roles.
Did you ever think about not doing DevRel as your next role? Of course. Yeah. 
[00:40:02] Joe Karlsson: I actually, I started applying for just doing like database or front end work again. I did. I, cause I have to say the job market did make me a little nervous and I wasn't sure what was gonna happen. I actually ended up doing the exact opposite.
I ended up hyper-focusing on data and database related developer content. I ended up getting five job offers. Oh, wow. Okay. Yeah, I know. I feel super lucky. It turns out getting laid off means you have. You could full-time work on interviewing and meeting and networking with people. Doesn't 
[00:40:35] Matt Yonkovit: make it any less scary, but, 
[00:40:36] Joe Karlsson: no. And once things like started moving, I've realized like there's a still a massive need in this space. And the other thing I've learned too with this like tech turn downturn is like the big companies aren't hiring, but smaller companies are hiring like crazy right now. A lot of smaller companies are like using this time to like build some amazing teams really quickly. And I didn't have any issues with it. And also, non-technical, like non-software based companies like e-commerce companies or healthcare companies, of course are still hiring like crazy right now. But I did at the beginning and then I quickly saw that there's still a massive need for this weird niche I'm in.
And then I just moved full-time into looking in there and there was a million opportunities which. It shocked me. It shocked me. It's good. I come from a place of I have an established network. I have an existing audience I could lean on. I wasn't starting from scratch on this. I had a lot of experience in the space.
And it may, I may have a different experience if I was like a brand new, like college graduate coming out in the space and looking to get into it. But I, for me it was a really, It made me feel way more comfortable. Fair 
[00:41:45] Matt Yonkovit: enough, fair enough. So you joined Tiny Bird though recently. Like you said and coming into that though, you didn't necessarily know their technology, so how do you come up to speed quick on something like that?
Yeah, actually, 
[00:41:59] Joe Karlsson: I. This is this round of interviewing, I did a lot more due diligence on evaluating the products. So I didn't do this until the third or fourth round of interviews for the company I was at. But I did go through, at least they're getting started guide for each of them. So signing up for their, cloud service or whatever.
And then going through whatever they're recommending as they're getting started service. Just evaluate is this like how mature is this? How seriously are they taking developer content? How smooth was the process? How many button clicks did it take for me to get something working? And the easier and shorter time it took me to do that, the more impressed I was with those companies.
But it having even just going through an initial setup, it doesn't. I still don't consider myself an expert. I'm only on week two of this new job at Tiny Bird, which by the way, has been amazing. But it is, it gives you at least enough, like jargon or like words to use when you're interviewing to make you sound like more about the product than you do.
Or at least for me, having, like being able to articulate plans of how I want to get up to speed or what I think I might need to get done, or at least come up with some high level ideas for how you would improve the product, which usually. Impressed interviewers to have, like bringing to the table some ideas or thinking about it or how, like how you fit into their market play and how you like understand the product.
Yeah. I don't know. It's hard, but I definitely like at least poke around a little bit if you're in, if you're like, In the interviewing mode 
[00:43:25] Matt Yonkovit: right now? Oh yeah. Yeah. So what's interesting about Tiny Bird is there's a lot of open source projects that they've contributed to. And I know they're pretty big in the Click House community.
[00:43:34] Joe Karlsson: Yeah. It's built on top of Click House. 
[00:43:36] Matt Yonkovit: Yeah. So there, there's a lot of like overlap there. And so that, that'll be interesting to see, what what you come up with there. 
[00:43:44] Joe Karlsson: Yeah, I am too. Like we're still working on positioning and figuring out what we're doing here.
I think part of my play here too is to just figure out like what is gonna work and do some experimenting on kind of marketing. Tiny or tiny bird, for those who don't know, is based, it's a big company in Spain. It's was founded in Spain. There's a big audience out there too, and like it's growing here in the us.
But is our marketing play gonna be different in the us? What are the needs of users who are interested in here and we're trying some ab tests to see what kind of content might be working and working in just seeing what's working or not right now, like for example, we're trying to we're working on like realtime analytics versus realtime database.
Is that a different market segment? Is one clicking more? Does one open up more like opportunities for use cases and If so, are those use cases gonna be a good fit for Actually becoming paid users eventually. But it, we're like, we're a new team. They're like, they're expanding a lot.
So it's just like a lot of us just we're gonna throw some dart to the wall for a little bit, see what happens, and then try to iterate and move quickly on it. Fair 
[00:44:52] Matt Yonkovit: enough. 
Fair enough. Yep. Joe, we've got a little bit of time left before we need to send you on your way to your next meeting.
So we're gonna go through some rapid fire questions. Okay, great. I have 137 questions. Let's see how many we can get through. 
[00:45:09] Joe Karlsson: All right. I'm very good at interviewing right now, so this is great. 
[00:45:11] Matt Yonkovit: Oh, good. Yeah. Practice. You're in practice. Yeah. Yeah. What's your first Linux distribution?
[00:45:18] Joe Karlsson: Oh, like the first one I ever used? Yeah. My dad, actually, my dad was a software engineer, so we always grew up with a with Linux, and I think the first one may have been either Mint or in Ubuntu. I don't even know if it was pre and Buntu, but the one I'm most comfortable with and have bit used for 15 years is un 
[00:45:35] Matt Yonkovit: Ubuntu.
Okay. What's your most used open source tool nowadays? Oh. 
[00:45:44] Joe Karlsson: There's so many. I feel like Visual Studio Code is still huge. And then, this is a weird one. I use a tool called RamBox, which is like a, it like puts all your messaging apps all in one window, which is okay. Really nice. I end up using that one a ton these days.
But I'm trying to think of an open source tool. 
[00:46:03] Matt Yonkovit: Or maybe the open source tool you couldn't live without, like you just used it on a regular basis. Have you ever heard of Curl Before? Have you heard of gi? Of course I've heard of Curl. 
[00:46:14] Joe Karlsson: I use those quite a bit too. 
[00:46:15] Matt Yonkovit: Yeah. Who hasn't heard of Curl?
[00:46:18] Joe Karlsson: Oh, actually, you know what I have been using and I'm most excited about, it's nothing to do with data databases. I am, I bought a new house last January and I got really into smart home stuff. Ah. But I got into a open source project called Home Assistant. Which is I, it's amazing. I'm obsessed with it.
I think it's great. I think it's, yeah, it's amazing. So anyway, if anyone's into home, like smart home automation, check out a home assistant, it's amazing. 
[00:46:43] Matt Yonkovit: We, I just did a interview with Patrick McFadden from data Stacks. Oh. And he was talking about home assistant as well, because he's into that and he actually used it to tell when his laundry had stopped because the kids would never take the laundry out of the washer.
Yeah. And so it nags the kids. 
[00:47:00] Joe Karlsson: I literally just did that automation last weekend. I put a vibration sensor on the back of the dryer. And then I said it, so if you actually knock it, it won't go off. It has to be like, it tracks the cycles and then it waits a couple minutes after it finishes. All right.
[00:47:13] Matt Yonkovit: So yeah. Yeah. So when you were way back, when you were in high school, college, whatever, what was your first job? 
[00:47:19] Joe Karlsson: In college I did it tech support, which shout out to those folks. They Okay. Really underpaid and have a h horrible, hard job. And then my first internship in college was for the Toro company the, do like the lawnmowers or the lawnmower and snowblower company?
Yeah. Which was It wasn't great. It was fine. I ended up we were, they were migrating to Window seven at the time, so I helped with that migration and doing testing on the the software to make sure it would survive the window seven migration. So that was a couple years ago though.
[00:47:53] Matt Yonkovit: Okay. If you weren't doing tech work, what would you be doing? 
[00:47:57] Joe Karlsson: Oh, this is easy. I'd be barber. I would be a barber. I'd be a barber. Okay. That's, I'd be barber. Yeah, it's like you work with your hands. It can never be outsourced. You just get to hang out. It's feels like kind of artistic a little bit.
You get to chat with hang out with people. Yeah, and it's just like steady. It's a good honorable profession that I think is like kind of fun and would be 
[00:48:19] Matt Yonkovit: You ever think about becoming the open source 
[00:48:21] Joe Karlsson: barber? Oh my God. That'd be amazing. 
[00:48:24] Matt Yonkovit: Like you could like, you could do a show where you cut people's hair while you talk about 
[00:48:29] Joe Karlsson: open source.
I was imagining like a floy attached to a 3D printer or something. 
[00:48:33] Matt Yonkovit: Oh, that'd be awesome too, right? That could be great. If you remember the Flos. Good. Good on you. They were a hot pandemic item. Yeah, that's right. Just say it. There were a hot pandemic item. Yes. So Floy, everyone look it up if you don't know what we're talking about.
What's the last book you read? 
[00:48:55] Joe Karlsson: Oh boy. I have to say, I think I forgot how to read over the Pandemic. What? My attention span became a lot harder. But the last book I actually read, 
[00:49:03] Matt Yonkovit: or audio book, you could do an 
[00:49:04] Joe Karlsson: audio book if you That's true. I am reading the People's History of the United States on podcast.
But the last i, last book I read was a book club. I ran on the Business value developer relations. Oh. Which was always great. I've actually run that book club a couple times. I love doing it if I have a team of juniors on my team and just like having a team discussion to align on there you go.
General terminology and like industry practices and just discussing what works and doesn't work for our specific business. 
[00:49:32] Matt Yonkovit: Okay so Joe you go to conferences, right? Yeah. You've spoken at conferences. I assume. Yes. Yes, of course. Because you're, you can't make it to the list of top 100 DevRel people and not do that.
Realistically when you're go, when you're there and they're like, okay, Joe, you're gonna go up on stage, you're gonna keynote this 10,000 person conference. What music should we play? As you walk up to the stage, 
[00:49:56] Joe Karlsson: what is this? Oh. Yeah, I know like your John Cino, like walkup music. Yes. I've actually thought about this.
This is, I there's a couple wrecking Ball by Miley Ray Cyrus. Okay. Miley Cyrus be really fun, like a wrecking ball. The other one I thought would be funny is Licious. Okay. Yeah, I thought that'd be really fun. But honestly, there's a song by Oasis, which I'm not a big Oasis guy, but it's called, I dunno if I can swear on this podcast, but it's called F-Word in the Bushes.
But it has the hardest intro and no one's really heard it. Cause I'm really like, want like a John Cena dang it, you know what you want, like a strong heard of. That it has that same energy, but no one really knows it. I don't know it, go check it out f fucking in the bushes if you're interested.
[00:50:43] Matt Yonkovit: Hey, hey, you can go for the shock value, right? You might as well. 
[00:50:49] Joe Karlsson: It doesn't even swear it, it just like hits hard. It's just If I was walking out of a sports stadium and everyone was screaming I just wanna yeah. It hits like that. 
[00:50:59] Matt Yonkovit: All right. Joe, thanks for hanging out with me today.
We only got through what, a dozen questions, not even that many. So there's still 120 something left to go. We'll have to come on again and we'll finish the list up. Yeah, we'll finish that list. Joe, thanks for hanging out with me 
[00:51:12] Joe Karlsson: today. Likewise, Matt, this has been amazing and thanks to everyone.
If everyone wants more, follow me on Twitter at Joe Carlson one, or if you like weird little videos on TikTok and two things there too. And if you're interested in setting up a real-time database, check out Tiny Bird. It's amazing. 
[00:51:28] Matt Yonkovit: All righty. All right, don't forget to subscribe everybody.
See you next time. Don't forget.