Fostering Developer Communities with Jake Ward

Brian Rinaldi 0:05
Hello, everybody, I really appreciate you joining us for another episode of DevRel(ish). For those of you who are new to the show, this show is basically like we’re going to talk about all things related to being a developer relations professional or other kind of tangential jobs, tangential roles. That’s the issue in the dev rel. So my name is Brian Rinaldi. I’m a developer experience engineer at LaunchDarkly. And actually new for this episode. My is now like Aaron is now officially a colleague of mine at LaunchDarkly. Brian,

Erin Mikail Staples 0:39
you have to deal with my crap now. Like, I mean, we already probably talked every single day, but now it’s like you have to talk and work with me.

Brian Rinaldi 0:47
Yes, and no. I’m still still shocked.

Erin Mikail Staples 0:51
still shocked. Yeah, we’re still working on it. It’s been almost two weeks and he hasn’t gotten rid of me yet. So

Brian Rinaldi 1:01
yep. Now now. It’s just instead of just on Discord, I just have you on Discord and on Slack.

Erin Mikail Staples 1:07
All the channels. And

Brian Rinaldi 1:11
And okay, so joining us today, our guest is is Jake Ward. Jake is the founder of data protocol. Data protocols, also sponsors dev relish. So thank you for that. But, and, Jake, you’re coming to us from like the middle of our you near the water there in San Francisco. Where are you?

Jake Ward 1:31
I am, I’m down at the Ferry Building on the Embarcadero in San Francisco. We’re out here working with some partners to build new curriculum and courses. And I thought I’d spend the afternoon talking to you guys down by the number of the bay.

Brian Rinaldi 1:48
It did look familiar. I feel like that does look like you’re right by the water in San Francisco. So hey, bridge wise. Yeah, yeah. Awesome. Well, hopefully the weather is nice and doesn’t start raining on you. Rain that much. Oh,

Erin Mikail Staples 2:03
yeah. Like the most scenic background we’ve ever had a guest on the show.

Jake Ward 2:12
I appreciate that. As long as the music doesn’t interfere with the audio, we should be fun.

Brian Rinaldi 2:18
No, it’s okay. I’d like Latin music. It’s it’s perfectly good. You can barely hear soundtrack of our lives. So so. So Jake, first of all, we up till now we’ve had like everybody who’s come on is working like a straight DevRel role more or less. I mean, some people have been CTO, but your background is actually a little bit different. So first of all, you were like more of an entrepreneur. But you have a really interesting background. We were chatting about a little bit earlier, kind of when I was digging through like your early career. I’m like, Oh, this is this is different. You have like a really interesting career path. So tell us a little bit about about, you know, what got you where you are today in terms of like, working with Deveraux professionals and stuff like that.

Jake Ward 3:07
Ah, I’ll do my best to try and keep it to a tight five or six minutes as that when I was a much younger man, I worked on Capitol Hill for the senior senator from Maine Olympia Snowe. I’m also from Maine. So that was sort of a dream come true. I peaked early, as my mother likes to remind me on a regular basis. And in 2006, which seems like a lifetime ago, the very first net neutrality fight was underway. And Olympia was she sat on the Commerce Committee, which was considering the role and she was the only Republican that was pro net neutrality. And so the founders of Google used to come to our office on a pretty regular basis and teach her how the Internet worked. This was sort of a fundamental fight about how information would would move around the interwebs. Everybody’s probably heard the hilarious statement from Senator Ted Stevens about how the internet is not a dump truck. It’s a series of tubes. That that came from a hearing during that neutrality fight. And he’s actually responding to a question that Olympia asked the panel. So my first blush with tech was at a pretty high level. And from there, I was hooked. I spent the next few years working, doing sort of policy communications for advocacy for big tech companies like Google and Amazon and others. And then in 2012, I co founded the application developers Alliance, which was the first sort of trade association for software developers and 2012, which also seems like a lifetime ago, we were on the frontlines of a number of issues, right? Apple would just rip flash out of all their apps, privacy was becoming a real issue. We were getting ready to bang up against GDPR. In Europe, there was a lot of stuff going on. But more importantly, I was on the frontlines of having conversations with about 150,000 application developers in Europe in the United States, and I thought he’s gonna be on to something. This might be really important. This could be sort of the the future of where we’re going. And I remember saying once on stage that software engineers at the manufacturing class and digital age, and they should be treated as such, that this is a skill. This is something that needs to be appreciated and taught and trained and continually elevated as a craft. And so when I left the Alliance, I sought to help the people, you guys that are helping them, right, what is what is it that every developer relations person wants, they want more time with their developers, they want to give them what they need the way they want. They want to be able to measure it so they can defend their jobs, defend budget asks, and scaled programs. But mostly, they want to enable developers to build cool stuff. And when I’m sort of in the post, Cambridge Analytica months, I spent almost a year and a half on the road, talking to 1000s of developers and asking them, What do you need? What can I do to help? And they all said the same thing, tell me what you need me to do, whether it’s how to integrate your product, or how to comply with your rules, or how to meet the terms of service of our partnership, and I’ll do it, I don’t want to break this, I’m not trying to do the wrong thing. Just be more clear about what you want. And so we set out to build that solution. And data protocol enables any company to treat all of their developers as they are managed partners, white glove, everything you need, what you want, in your hands when you need it.

Brian Rinaldi 6:29
Very cool. So yeah, you took you started out in more of the policy, because I mean, typically on the show, it’s it’s been like, Okay, we start as a developer, and then we get into the community, but I actually really enjoy the idea that you came from dealing with, with policy issues. And by the way, I was gonna bring up that quote, if you didn’t, the series of tubes, because I do remember that exactly. We had T shirts.

Erin Mikail Staples 6:55
Love that hold on, where can I find this? There’s actually it’s like, under the blue sea or something is a book that I read in grad school. And it’s all about how they the policy that went behind actually building the infrastructure

Jake Ward 7:07
that we, you know, we call The Daily Show immediately, like, as he was saying, this ridiculous thing called The Daily Show, and we had T shirts made. And so when the final vote came out, about a week later, a bunch of people were sitting in the in the hearing room wearing T shirts. It’s a it’s not a dump truck. It’s certainly too

Brian Rinaldi 7:28
arrogant, too young for this. You may have missed this one you probably in like, in like, elementary or junior high when?

Erin Mikail Staples 7:38
Yeah, I was maybe an elementary, junior high. But I did take an internet policy class in grad school. And so not to be like, Y’all are my textbooks, but like y’all were in my books. Yeah, it’s probably

Jake Ward 7:49
November of 2006. I think it’s one that hearing was, and that was actually

Erin Mikail Staples 7:53
the later Middle School.

Jake Ward 7:56
Almost like the rest of the quote, is even more deranged in the beginning. Oh, really? He goes off on this thing about how his staff sent him an email last week, and it just came to him. And it’s because there’s so much stuff in the, in the system to too many things. It’s like, actually, it’s because he didn’t scroll out on your Blackberry. You old old man. I shouldn’t. Senator Stevens was a good public servant. And he’s dead now. So rest in peace. But the times they did pass him by?

Brian Rinaldi 8:27
Yes, I remember those times. I remember. I mean, not to kind of belabor this point. But But I remember years ago, I was, I had a job. And I was building websites that this day was we focused on like, stuff for doctors and medical stuff. It’s just a marketing firm. And then my boss was like, my monitor was acting up and she’s like, do you think it has a virus? I really had a tough time explaining to her why that wasn’t possible.

Jake Ward 8:58
fairly certain it does not.

Erin Mikail Staples 9:03
That’s just me with my parents, because they have no idea what I do for a living. I asked my mom, I did ask my mom. I was like, Mom, what do you think I do? She was like, gosh, like, How’s your new job? And I asked her, she’s like, you’re a person who fixes bugs. And I was like, actually, kind of maybe close. I actually create them more than

Jake Ward 9:25
that’s, that’s a hell of an evolutionary and she used to think you spam the internet.

Erin Mikail Staples 9:29
Yeah, we moved from spamming the Internet to fixing bugs. Then I asked her like, what kind of bugs and she was like, I don’t know. You still do it. People I don’t know online. And I was like, Alright, cool.

Jake Ward 9:39
That’s a bridge too far. But my policy is interesting place to start this conversation, right? Because the vast majority of policymakers particularly 10 years ago, but even today are certainly sexagenarians They like they’ve not I’m not a digital native, right. There wasn’t a computer in my house. I’m 45 there wasn’t a computer in my house until I was 14. You The difference between being able to make decisions that affect people at scale on these complex issues has a lot to do with your comfort and your your fluency in that space. And what I found pretty early in my time in politics, and even more in the years following is that the gears of innovation and policy don’t align, ever, like innovation spins like this, and policies supposed to spin like this. So if they ever catch, it’s almost a mistake. And that we have to do better for ourselves that as an industry, technology needs to be responsible, it needs to be proactive, it needs to imagine a world in which there are no guardrails, from a regulatory or legal standpoint that actually apply to the next thing, because policy only looks backwards. And that’s the other reason that I built data protocol was to give developers what they needed to not make mistakes, and to responsibly use data and products and build their dreams.

Brian Rinaldi 10:56
Right. So you know, and, and when you, we got to tour the product a little bit, but like, so to kind of get a better sense of like, for everybody of what you can do. So beta part of all like, basically, you can manage your community by like, creating videos and things like that, and being able to kind of cater content directly to to different parts of your community to kind of cultivate it anyway. Yeah,

Jake Ward 11:25
I mean, at the risk of turning this into a promo, which I’d really rather not do, because nobody wants to know that. The I believe that video is a better way to deliver information to an audience, I think that it’s more intimate, I think you can add more nuance to it. And so we built a platform that leads with video that is developer facing, I don’t want to get into the nuance of it. If anybody is listening wants to take a look, just go to data protocol.com And check it out. But there’s a Linux terminal in which we can ask questions and as a notes function, so we drive home sort of engagement, your hands stay on the keyboard, and the information you need is brought in and chyron and layered in a notes. So if you’re going down for dev doc, which is sort of standard for every point of engagement in this, when you’re when you’re building a community, right, the first thing that we usually lead with is developer documentation, because that’s the language in which we speak. But that’s so limited in what we can prioritize. And what we can add links and interaction are often just seen as noise on the page. Whereas video, you can do much more sort of nuanced planning of complex issues and let people choose their own adventure. And so we built this platform that is single sign on, fully attributable, so you have an account, we, our partners use us to build content, we write it produce it, hosted all of that. So a one stop shop, we don’t add any work to any of the developer relations, developer engagement teams, and you send them to us, we’ll send them right back to you. But when they’re here, we’ve got video, we’ve got resources got in our dev, Doc’s, all of it is attributable with metrics, the outcome being a higher level of engagement, you know, that they know, they have done what they need to, and you can give them updates and information. It’s the it’s sort of the Rosetta Stone of Community Relations, right? It’s at scale, it’s it’s feels like one on one, it feels like this, it’s a synchronous, right? It’s there when you need it. It’s a joke. That’s what StackOverflow would look like, if you didn’t have to yell at people first. It’s, it’s just, it’s a better way to deliver information.

Brian Rinaldi 13:26
So and So you let me ask you like, because it is actually something we debate in general is, you know, is I tend to favor like, personally, I tend to be more of a writer. So like, I’m not great at creating videos, even though I do this all the time. So, but like, what, but I do feel like there is a generation of developers, particularly the kind of ones that are entering the market, right, you know, in the past five years or so, that really tends to go straight into YouTube or something, you know, to basically find out like, oh, I don’t google it, I actually go to YouTube, and I like search for a video about it. That’s me. That’s, that’s Aaron. See? And this was the benefit of me like I’m generation. So the same show

Erin Mikail Staples 14:15
is gonna say I’m a partial Googler. But like, Okay, I started probably 25th team was like learning to code for money, like got my first leg, and it was email templates. And that was I do remember Shopify was releasing, it was mostly in the Shopify ecosystem and like, but like they had, it was like their docks, and like the YouTube videos were helpful.

Jake Ward 14:40
So I don’t think the I don’t think the mechanics have changed much when a developer is sitting on a line, right? And they encounter a problem. The first thing they do is turn to the person next to them and say, Have you ever seen this before? Or do you know how to do this? Right? That is universally true. It’s true of my team. It’s true of yours with the ink resend remote work where we’re not near people we use I mean, instant messenger was invented. So the developers could talk to each other to fix problems. That’s, that’s where that came from inside Microsoft. Right. But the next thing that happened, and I was incredibly fortunate when I launched the Alliance, Joel Spolsky, the founder of Stack Overflow was the chairman of my board. And so I spent a lot of time with Joel, who wasn’t there at Microsoft when they were doing all of this stuff. And he built stack because you wanted to take that, that value out of asking an expert or asking somebody who’d been there before and scale it. And I remember him saying, developers aren’t gonna turn to anybody and ask a question that isn’t a developer that hasn’t seen that problem before. The only way, the only way you can outsource that to a larger community is to do it at scale, which is how they got the 20 million monthly users, right. But the the problem with that is it brings all of their usual humanity bullshit, where people are mean to each other, and they they talk down to each other, and there’s all that noise, and only people liked it. And you can see that immediately when chat GPT and copilot took off, their numbers dropped by 50%, because people didn’t want to get burned before they got the answer they needed. They could just put it into chat GPT. That’s all sort of technical stuff. Like have you seen this? Is there a workaround? What would you do here? stuff, when we’re trying to build is much more on the solving big problems? Right? Looking at challenges at scale? How do I solve for GDPR? How do I navigate this compliance issue with metta quest? What do I need to know about web three, if I’m going to use smart wallets at Circle, things like that, that we build for our partners? And all of that, it leads with video because 83% of developers according to slash data, say that they prefer video be the primary function for delivering that information, which all makes sense. But YouTube alone isn’t enough. Right? There needs to be the next piece, there needs to be a level of interactivity where you don’t just get an answer you remember it? There’s proficiency. We know we can ask you questions, so that I sleep better at night is a platform that I’m just giving you all this data, because you’ve demonstrated you read the privacy policy, you know, your terms of services, you’ve run through all the traps, we’re good, usually better as a developer, because you know, I’ve got your back any resources, you need any support you need. It’s right here, right? We’ve curated for you, we’ve put it in milestones, rather than just sort of spraying it over your dev docs. There’s still a reason there’s a there’s a need for developer documentation. And this goes back to the community piece that you have to focus on giving developers at scale answers when they need it. But if you’re command effing your way through 40,000 words, there’s a pretty good chance you’re gonna miss something.

Erin Mikail Staples 17:34
Yeah, and I think like one of the things I think about a lot, I think, in my own learning journey, like where I have personally a lot of gaps, like, I know my gaps like I can, if you give me a problem, and you tell me to figure something out, I’ll figure it out. But you’re like, why did you do it that way? I’m like, right. Like, and I kind of stare at you for a second because a lot of it has been stumbling and figuring out what works. Um, that being said, like one of the things that I think is really interesting is we’re getting the the tailored made stuff. One of the demos that we used to give all the time, my old organization is how do you train GPT based on a GPT model on your developer documentation, which is fine to a certain extent. And then it breaks and it and if no one is sitting there actually running the code. Find it? Yeah.

Jake Ward 18:26
Yeah, I had this conversation the other day with. So I’ve known I’ve been doing this for a minute. I’ve known a number of real life 10x coders, right, like people who, what they do, their brain is just wired that way. There’s no problem they can’t solve. The frustration they have, with people asking them all day long. How do I solve this? How do I fix this? What does this look like? is like an eight. The frustration they have when people don’t ask is a 10. By the I can solve this for you at 90 seconds just asked me the question. But imagine you could do that at scale, asynchronous on demand, where the question can be anticipated the answer can be relayed, and that information is there. Whenever somebody needs it. That’s the dream. So that you’re not taking those experts off the field. You’re you’re putting them in a studio for 45 minutes, once every six months, and you get all of the answers you need for forever. To your point, Brian, about how some many developers are getting better at doing video. There’s no question that’s true. I had a meeting this morning with one of the OG video content creators out here who talked about what I what an amazing moment it was when they started putting the lighting in his office so that he can record videos five, six years ago, and that’s important. I think that the quality of the video matters, but I think that this shouldn’t look like it was filmed in somebody’s basement. That that the point of this the manufacturing class in the digital age, we should respect the craft. And when we shoot we do it in a studio with green screen and a fold whole crew gaffers and grips and DPS and the whole deal. Because this stuff supposed to live on for a long time. And the quality of it, I think needs to be commensurate with the with the need. And these guys need help.

Brian Rinaldi 20:13
So, so one of the things like with within DevRel that I’ve I’ve struggled with when creating video content is like, what do you do? How do you fix things? It’s a constantly moving environment. Yeah. I mean, that’s, that’s the trouble, like, we create YouTube videos or whatever I mean, at work, and it’s like, okay, well, the UI is change, this has changed. Like, I can’t really go back and easily edit that into those videos. What? What is, what do you all do? Like? How do you sell that?

Jake Ward 20:40
That’s a great question. So one of our one of our early early problems, right? In our in our path? How do we scale this? And then what happens if something changes because something is going to change? What we found is that by breaking videos into components, and then having the graphics beside it, we can just change the graphic, right? So instead of saying in this the developer documentation, it does these things, you say, in this developer documentation, you’re gonna find the answers you’re looking for. And then on the side are links to the dev doc with the with the bullets. So to swap that out, it’s not a problem at all, you change the link and change the animation, you’re up and running in an hour. Sometimes. And this happened pretty recently, something changes that you have to say out loud, increasingly, we’re able to use AI to change that, right? That we can go in and grab enough video, audio, from this conversation from whatever we can get you to say a new sentence. And with the inflection writings and your approval, we just dropped that in over it. We did that last week. You’ll never know the difference, right? It’s yeah,

Erin Mikail Staples 21:45
having used some of those tools, like I always find the inflections really hard to nail. Are you using a model that’s in you can be like Aaron no propriety edit box. But like, you’re using a model that’s pretty already existing or have you treated fine tuned it like how are is that kind of worked off of?

Jake Ward 22:02
We have to train it to do better. We’re able to get like one sentence right now. Okay, right. We couldn’t do. Yeah, we couldn’t do 20 seconds yet. I would certainly played around with the idea of using avatars. They all are pretty deadly eyes, which I don’t love. Yeah.

Erin Mikail Staples 22:19
I also think it looks so creepy. Like, yeah, yeah. And the

Jake Ward 22:23
I hate the protocols, not a learning management system. Right. But it’s got a lot of the bells and whistles of a really good one. And one of the things that I like about an LMS is being able to measure proficiency, being able to pull that data out being able to integrate across things, what are the things that I hate is how frequently they in order to solve this problem, they abandon using a person and putting graphics on the screen and choose animating something and then doing a voiceover. I think that’s just shy of a voiceover PowerPoint in terms of just not worth my time. And so we we avoid that entirely. But if we wanted to do for example of what we call a TLDR. So it’s just a 92nd. Update. We’re deprecating this API, you need to go to this link. This is really important. And we needed to do it really quickly. We could run an animation and pull down and an AI generated voiceover or I could do it, who cares? And then it would just lay over the top of something else. The short answer is, we’re getting better at it. We’re already pretty good at it. It is part of what we offer our clients in terms of what they get for their money, as we’ll update things as needed.

Brian Rinaldi 23:30
Yeah, I hadn’t even thought I have tried things like the script early on, and if you know descript, but it wouldn’t I think it’s gotten better because that was one of the ones where you could just go in and fix. But it wasn’t, wasn’t perfect. But I think it’s gotten a heck of a lot better. One of the things that funny enough, because I do so I when I record videos, when I tend to have a script, and so you’re not looking at the camera, and they had an update recently that it’ll actually make it look like you’re looking at the camera. Like oh, okay, well, that might go try it again. Yeah,

Erin Mikail Staples 24:07
is that that’s under script, right? Like that’s, yeah, that’s cool. I do have like, it’s kind of annoying. Like I just moved my camera. So it does cut off like it does interface part of the screen. But this is also like, what happens when you have a way too big monitor and like I can sacrifice little real estate for I’ve

Brian Rinaldi 24:24
seen Aaron screenshare and it’s like it’s, it’s obnoxious.

Erin Mikail Staples 24:30
It’s like how many windows do you have open in the night, of course, leave all my windows open. So it’s like,

Brian Rinaldi 24:36
imagine you live in this tiny little New York apartment with this monitor that takes up half the apartment or something

Jake Ward 24:41
to TV. She’s done. It doubles as a TV

Erin Mikail Staples 24:43
the TV. It’s basically the TV. And so and it’s so funny like we have our we have our desks like my husband, I have our desk battleship style, and then he has a big monitor too. So it’s like this like kid would see each other. Yeah, we can Oh no, I can’t see him at all. And it’s so he gets so annoyed because we also will do up in like if one person drops in a huddle or not, like, you’ll start to hear the other person in the background recording. And there’s times I’ve been like, that was such a great take, but I can hear my husband back.

Jake Ward 25:12
Yeah, it’s funny. We, we, we put everything on a teleprompter, like part of the other thing is like, as LaunchDarkly you want to do this thing, we get the outcome. Our content team has instructional designers, they lead you through discovery, we write the script, you approve it. You come into studio, it’s on a teleprompter. So all of this right. It’s a very personal experience. And then post production gets it the animation comes in. It’s, I joke that it’s the best thing on the internet, but I’m not really joking. It’s really good. But it’s the point is that it’s supposed to be commensurate with the effort. Yeah,

Brian Rinaldi 25:46
yeah. Yeah, I mean, you know, these are the tools I was describing, obviously, are for much more like, just kind of like, Oh, I’m gonna ad hoc throw up a video about but it’s coming.

Jake Ward 25:56
It’s coming. It’s really interesting stuff. Definitely.

Brian Rinaldi 26:01
I mean, I, you know, it isn’t it is amazing how quickly AI has transformed like, I think even like, for deverill. Honestly, I think it’s almost, it’s been more useful in many ways than I would say, as a pure developer, like, because when I’m coding, I’ll use copilot, and it’s, it’s helpful and stuff, but, but there were tasks in Deborah, like, for instance, chopping up this video later, and getting highlights, right, like, that used to be a nightmare. And because it was just, it was like so much ours that at the end of the day, I’m like, Oh, is this really worth the effort? And, and I kind of gave up on it. And now I just some push it into, like an AI thing. And it’s like, oh, here’s all your highlights tucked up perfectly for you. With you know, the Yeah,

Erin Mikail Staples 26:51
we definitely. I love it for as I will say like some kind of like what you were saying earlier, Jake is like the, the pre produced like super static, like, basically glorified PowerPoint type YouTube videos. I’m equally annoyed. And I’m like, I’m not gonna watch this, like I won’t. So I’ve actually found myself following a lot of Twitch streamers, because I like the live interaction, I have a question I can actually ask and learn a ton. And I’ve actually learned a ton that way. But like that being said, it’s like, one thing that I did at my last role is a lot of we would I would practice work through something on twitch or work through something like a team call or a zoom call demo. And then I would take the recording, and I would stick it in transcript. And there’s my tutorial blog post.

Jake Ward 27:34
Yeah, yeah. I mean, that’s the so to get us back on topic, because I’m pretty sure I do this whole thing. If you were gonna start a community right now, like you got a new product. It’s SAS, but it’s developer. First. You’re you’re selling to developers in mass, right? Or they’re your customer, they’re your users, but they’re not the customers. You’re you’re building scale on the backside of it net new, you’re starting today, what’s the first thing that you do? Right, the first thing that you do is build a funnel that leads with video. So you can introduce yourself and show people how to use your product for your, you know, to the point of this is my blog post, you’d put it in video. And that’s how you’d run it. You’d probably use Twitch, maybe YouTube, the truth is neither of those do with data protocol does, because we have the attribution engine. But that’s just a slightly different model, right. But the piece that we would, we would naturally skip over but we shouldn’t, is the metrics. And not just because it’s how we keep our jobs. But it’s also because that’s how you learn to get better at it. You can’t ask your your community to do your job for you. You have to ask them how to do their job better, and then adjust accordingly. We don’t have those metrics.

Brian Rinaldi 28:45
Yeah, yeah. I mean, I think I’ve having used like YouTube a lot. Like, it’s very hard to figure those metrics out from from what YouTube gives you. So all I know is like, I know the watch time and I know raw numbers of how many viewers and things like that. But like, I can’t figure out like, for instance, it’s hard to figure out why they stopped watching. Or you

Jake Ward 29:08
have no, you have no indication of why they. So actually, this is a is a thing that my team and I have been doing for the last couple of weeks. Typical past products, you spend a lot of time and energy and money, you go to events, marketing, thing, pieces, whatever, you’re driving people to the top of your funnel, you’re showing them your product, and you’re trying to convert them to adoption, right. But in that space, all you have is ostensibly Google Analytics, how long were they on your site? What did they click on? That’s it, you don’t have any attributable network metrics. If they don’t adopt and register. It’s a black box, you have no input as to why they were there, why they left how they got there, you have none of that information, which means that when you actually do the conversion from adoption all the way down through to like cost of service, lifetime value generated annual etc, etc. Your ROI number isn’t right. It’s wrong because it doesn’t account for any of the bounces. Pre adoption that you know is happening, but you don’t know why. So you can’t improve your uptake. It’s just trial and error. Like, what if? What if I went fishing in a different pond? What if I lead with this? What if I showed price earlier? What if I, and you find the thing that works well enough that you’re not going to change it? So one of like, one of the things that I tried really hard to do is solve that problem to create an attribution funnel earlier in the process of hold on Secretary.

Erin Mikail Staples 30:30
This is the the very San Francisco part,

Jake Ward 30:37
create, create a process whereby you have sort of an SSO earlier in that, in that pre adoption funnel, so you could ask questions, how did you get here? What are you looking for? We know the developers adopt based on do what I needed to do? Will it keep doing what I needed to do? How much does it cost? And do I have support? As I go through this process? If you can answer those four questions to their satisfaction, they’re gonna adopt your product. But how are you? Like, how do you lay out a website in order to answer those questions, not knowing why they got there in the first place. Right,

Brian Rinaldi 31:10
right. Yeah. And I think, you know, to that point, it’s an I’ve seen it kind of brought up a bit lately. As you know, we kind of hit peak div row, three COVID, right, like, yeah, maybe even during COVID. Like, we’re just every company seems to be hiring devils, like, like crazy. And then now we’re in kind of a post peak, a bit of a downturn for DevRel, in general. And I think one of the things that we’ve talked about a lot on the show, one of the things that has that DevRel folks have to kind of adopt is this, I need to buy into some more metrics. Like we used to be like, Oh, it’s hard to measure what we do. And I agree, it is hard to measure what we do it is, but like, but at the same time, like we can’t, ultimately those, we need to find metrics that can validate that we’re doing a good job that we’re having an impact on. I mean, at the end of the day, DevRel is about selling more software, really.

Jake Ward 32:20
Exactly. One side or the other, or both, but you can’t be a blackbox on both sides, you’re either growing the product or reducing breakage, or both, you can’t do neither.

Erin Mikail Staples 32:31
Right. And, Jake, that brings me to another question. And like how does because like we see roles like community, and I think community roles are probably another role that you don’t see title and Community Manager, head of community, community events person, those titles aren’t as much around anymore. I’d be curious to get your take is like, is this still a standalone independent role from dev rel? Or is it more of a customer service function these days? Or like, is it a dev role function that they should have?

Jake Ward 33:00
I think that’s a really good question there. And I so I, I came up when Twilio was doing its thing, right? Like, you know, Keith, Casey, James Harden, and, and Robin, those guys were, and Linda Smith sat on my board at the alliance and was an advisor early days at the data protocol. She’s incredible. They invented this level of hand to hand door to door Dev Rel, they were in hackathons, 1520 a week, right? And they were 10 people. They didn’t care. They were just showing people how easy this was. Because they knew that if you prime the top people, the same phenomenon people were going to ask how’d you do that? And they just tell them and then it’s one more click one more click one more clip. And that growth, you know, they saturated the market. But even now, they’re reducing Devereaux across the board at Twilio, because you can’t grow it anymore, while they’re increasing engineering support, because you can break a lot. The question about community management, I think it’s entirely about how complicated your product is, and how important or how frequently you need to engage to do it. Well, I think so. Zoom, I think is a really good indicator of where this is going. This is a product that’s been around for 20 plus years that nobody had ever used before the world came to a screeching halt. And then it became like this rocket ship of for stock that hired a bunch of people, then they had Oh, no, we, we might actually have a privacy problem. We’ve got some other feature stuff. We’re not ready for this from an engineering standpoint. So they doubled down on the number of engineers and did a really nice job riding the ship incredibly quickly. And now they’ve got this new API that they want to roll out that puts all kinds of community functions and features on their on their platform, which is cool. It’s really cool. So they’re hiring DevRel folks to go tell that story, but they’re using sort of outdated modes of communication to do it. Not because they don’t know what they’re doing because they’re limited in terms of what they can do. ploy, but you can’t manage this on a discord channel. You can’t just run a Slack channel and keep people happy. Common room is doing an incredible job. Orban is doing an incredible job putting people in a position where you can have more conversation, at least with the the highest indicators of whether or not this is working. But it has to be scaled at an asynchronous level, in order to be effective, you have to do a better job one on one or at least virtually one on one, if you’re going to drive that level of engagement. I don’t know that anybody’s doing it well yet.

Brian Rinaldi 35:34
That’s really, in an errand actually came from we were at orbit, right?

Erin Mikail Staples 35:39
Yeah, it was the orbit there.

Jake Ward 35:41
I mean, it’s such a good idea. They’re there they in common room are way down the road on doing some really interesting stuff. But the question about community management is like, one to many. I think we’re done with that. I think it has to feel one to one. I think

Erin Mikail Staples 36:01
the the biggest takeaway I have from someone who worked at orbit who’s been in those community roles, who I mean, I started my first job in this space is a head of open source community. And it was an accidental, like, Oh, you’re not a product manager anymore. You’re now in head of open source. Good luck, figure it out. And I remember Googling, like, what does a head of open source do? Like what is head of open source community do like, what should I be responsible for? And I think, and, you know, I got the question recently, someone asked me, like, Why did you move to Dev Rel from community like, blah, blah, blah. And I was like, first of all, I actually like the building part of it. Like I am technical, I like to build the part. And that’s fun for me. But secondly, I have maybe a spicy, take that, like, you can have the best numbers, the best metrics, the best everything in the business, but at some point doesn’t matter. Like, are you engaging with the right people? Are you engaging with the stakeholder? And are you actually solving a problem? Like, looking back at my career orbit? Like, specifically, the worst feeling was, is like, knowing the engineering something was broken? Or we were sunsetting something or, and being like, I can’t do anything. This?

Jake Ward 37:07
What am I, my one of my original board members Harbor, one of my original board members of the alliance, Don Dodge, who was he was a you guys might know Don is he’s a legend. In dev advocacy. He was at Napster. He was a Microsoft Word for Donald Lewin. Anyway, when he was on my board, he was a developer advocate at Google. And this was again 2012. I’d never heard the title before. What’s the difference between Dev advocates, evangelist, and he’s like, I don’t go talk to developers about Google, I talked to Google about developers, my job is to take what they need, and bring him in house and to advocate for them inside this company. That’s really cool. That’s also sort of what community management needs to look like, which is that you are, you are asking for a show of hands, and then you’re going out and shaking a bunch of moments. And so you’re saying, let’s talk to me, what do you need? Why is this breaking? What are you seeing? And 234? Those that’s a trend, it’s time to go talk to engineering. And then you work through documentation, you look at product features, you can improve. We talk about this all the time, you can prove your product, your process and your program, with metrics with anecdotes are not enough. You need hardcore metrics. And we’ve just been terrible collecting memory for 20 years.

Brian Rinaldi 38:20
Yeah, I think you know, well, to go back to that point I was talking about earlier, first of all. Okay, I have two thoughts. Number one is, I think there was a difference between Dev evangelism and Dev advocacy. Now, it’s kind of like the roll is like, to me, it’s kind of both at the same time. There’s nobody who’s specifically that one way anymore, I think, I don’t know if you all agree. But I, I’ve never talked to anybody who has who can say, Oh, I’m just there to kind of bring this stuff back. It’s always like, I’m there to be the megaphone. And also to bring the stuff back.

Jake Ward 38:58
Google’s a different animal to I think that that Oh, yeah. I think that might be a Google specific thing. Oh, yeah.

Brian Rinaldi 39:05
And I mean, honestly, Google’s revise their demo program to everybody has, like, you know, they kind of we go through phases I’ve been, I’ve been doing this long enough to be like, I’ve been through various phases of Dev Rel, you know, to the point where, like, sometimes I get like, they’re like, oh, you know, what’s the difference between this title and that title? And I’m like, so slight change in focus, but it’s really very similar kind of roll. But, you know, I do think we, when we were in that kind of peak phase, it was easy to get away with, like, because everything was growing. So when everything was growing, everybody’s like, Oh, you know, it’s okay. Like we we’d love some metrics, but it’s okay. If we don’t, if we don’t have them, because we know whatever you’re doing seems to be working. And then now it’s like, now when when things are tight, it’s like, well, we have to justify our exist. And, and that’s just our existence like, maybe they’re not looking to like lay you off. But they may say, hey, you know what? We’re going to pull your events budget because we can’t approve that you’re, you’re going to work events is actually working? Yeah, exactly. Yeah, correlation isn’t

Jake Ward 40:13
enough. When times get tight causation is the only thing that budget makers are looking forward. If I do this, will it do that? Because predictability and smoothing out revenue is the key for any company of any size, those those spikes and drops? that’ll that’ll put you out of business.

Brian Rinaldi 40:30
Exactly. Right. Yeah. So like, you know, and figuring out what those metrics are? I think it’s been, it’s been the topic of a number of our conversations here, you know, when we talk to them, like, there’s different theories on what those metrics are, which, which ones like make the most sense, you know, and

Jake Ward 40:49
what, so one of my big things when we started doing data cuts inside the protocol, was that I didn’t want to do sampling, right, that sentiment had to be better than representative. Because I 1000 years ago, when I was in college, like, I know what response bias looks like, you’re gonna get squeaky wheel stuff, I hate your product, I love your product. My service was terrible, you know, errands the best. Those are the only two options that people that are engaging. But if you could ask everybody, then you’re going to get a real representation of where you stand. Could you do better? Should you do better? Would it even be worth it? If you did better? If you’ve got 82% approval rating? Should you chase the other 18? Or is that just dead money? Right, there’s a there’s some of that math that can only be can only be fed into the machine of where you put bodies and resources and time. If you have those answers,

Brian Rinaldi 41:44
right. Yeah, absolutely. And I think I think that, that we just need to figure out, you know, in each company is kind of different, what those answers are, like, what the numbers are, and, you know, I always when I when I brought this up in the past, like, you know, not recently not in this role, but in past roles. It was like, well, because it’s fuzzy, like, I can’t prove that that number is perfect. And so like, then I’m, so I’m just not going to use that number. And it’s like, well, a certain point, you just have to find a number that like, okay, it’s, I mean, to your point, like, yeah, you have those and I can’t I don’t necessarily have everybody’s answers, I only have the squeaky wheel answers. But I need some metric in like that, if I’m gonna say, you know, that way, I can say, Well, okay, we’re at 2% is not good enough for us. So we’re gonna see if we can do things to make to get it to closer to 90. And, you know, sometimes you find that metric and you like, you dig a little bit and you put some effort into, and you find out Yeah, it was, I can’t move that metric. There’s nothing I can do. And then you find a different metric. But it’s not it’s not a matter of like, I don’t have metrics. You know, because it’s difficult, I’m going to just be like, Hey, I can’t measure height. We can’t measure dev rel. So you know, it’s just not.

Erin Mikail Staples 43:00
I think one of the things that separates having kind of been in this role and a couple of different and like, seen a lot of different folks in the space, one of the things that separates people in general and community that I really admire and look up to versus folks that I’m like, What are you doing is like the ability to a confidently say, I don’t know what it doesn’t matter, or be like, is this a metric I should actually care about? And like, see, I think, you know, data can be manipulated database could be modified. And like data, you can say, you can make it paint a different picture that it is how you frame the story is how you tell the story. And it’s knowing what is statistically statistically significant to you and your goals. And I think that’s something we don’t want to have the confidence or like, we don’t we sometimes get uncomfortable saying that like, thinking about what is it pacifically? Significant? That’s a hard word to say today.

Jake Ward 43:50
If you were, again, back to the net new, like you’re starting from the beginning, you would lead with something that puts a face on your product, rather than letting it you know, Product. Product, like growth is important. Person lead growth isn’t the thing that really moves the needle, like, do I trust this person to give me what I need? Does the product do what I want to federate better, etc. So that’s net new, but the other piece is you build the pipeline for metrics. And so for us, like, I begged the product team over and over again to build all this stuff. And they’ve done an incredible job. But the questions are, we ask questions in each video, sentiment overall, is this good? Did you like this? Do you want more of this? You know, stuff like that, but also milestone driven? Are you taking this at this point in your lifecycle, so that we can set parameters around where people are getting information, how valuable it is, we ask proficiency questions, and then we have on platform metrics of what other courses did you take, how did you correlate this answer with that answer? What you get in terms of data cuts is if you come in from VMware, and you did this, this this, this you are more likely to have a high proficiency with this subject. So we can tell VMware Don’t send them through all this and you’re less likely to break your product. That’s a, I don’t care if it’s 20 people or 20,000. This, the significance of the statistic is less important in that instance and the outcome, which is a reduction of breakage. But if you’re talking about do people like this, it needs to be either incredibly representative of a statistically significant sample, or comprehensive and give you a number you can live with that is directly attributed to churn and burn, in terms of people bouncing off your product. Because if 100 People are saying, I love this, but your breakage is 50% People are leaving, they’re lying, right? Or you’re getting the wrong people. Or there’s a milestone after the course, where they are being forced out of your program. And you need to identify that or you’re gonna keep leaking developer parts.

Brian Rinaldi 45:49
Because they’re a really, really good point. So we are running towards the part of the show, Jake, where we, you know, people come for the devil, and all the tips and stuff like that, but, but you know, people stay for the pickles. And so we’re at the point in the show, we’re rely on our on our resident pickling expert, Aaron. Yeah, our pickle Fact of the Day. This

Erin Mikail Staples 46:20
is actually just like the most important part and gives me an excuse to keep Googling and learning things. But today’s fact is from the Pacific Islands, where natives pickled their foods in holes in the ground line with banana leaves and use them as food reserves in case of storms. So it’s like fermented stuff in there. These are some of the most valuable food items in that part of the world. And they’re used as part of the courting process proving that you can provide to a woman in the courting process. You cannot marry or enter into marriage without showing your parents that pickle pits. So, okay,

Brian Rinaldi 46:59
okay. Yeah. All right. I’m

Erin Mikail Staples 47:02
very disappointed. I’m honestly you know, I can see my husband tuning in in this chat every so often, but I’m very disappointed where the hell was my pickle pit? Like?

Jake Ward 47:11
Erin, I actually I came with my own pickle facts.

Erin Mikail Staples 47:13
Oh, I’m ready for this.

Brian Rinaldi 47:14
Nobody’s done that yet to hear that your.

Jake Ward 47:18
Your the fact might be a stretch of pickled perspective. I am not a huge fan of actual pickles. Like I will occasionally eat them except on a Chicago style hot dog. In which case they are amazing. And they have a requirement. Yeah, drag it to the garden. Have you ever had a pickled fiddlehead?

Erin Mikail Staples 47:40
Oh, no, I don’t even know what a fiddlehead is. So fiddlehead

Jake Ward 47:43
is a it’s a green. It has sort of a bowl that turns over at the top so it looks like the end of a fiddle form. And it grows wild in vain it has. You boil them you can steam them and we’ve seen them they tend to come apart. They’re delicious. They have sort of a almost bitter it’s sort of like this tiny taste to it when you pickle them and chop them up and drop them in salads. They’re incredible. Again, I don’t even like to claim but so fascinating.

Erin Mikail Staples 48:13
I’m like this is very intimate. There.

Brian Rinaldi 48:15
I just Googled what a fiddlehead is in. Yeah, okay. I don’t think I’ve ever fiddled there’s no way you like, Rachel knows. Pro

Jake Ward 48:30
they grew in the backyard of my my childhood home.

Brian Rinaldi 48:34
Okay, yeah, Aaron, you have a new task.

Erin Mikail Staples 48:36
I have a new task fiddleheads but I’m reading this a young fiddlehead Fern cooked in Indonesia. It’s cooked in a rich coconut sauce with chili pepper. Lemongrass Tumeric I’m like That sounds amazing.

Brian Rinaldi 48:51
That sounds so good. Sounds so good. Yeah, we shouldn’t be talking food I’m getting hungry already.

Erin Mikail Staples 48:57
I know I’m like now I’m I need to jump off and like eat after this is like my next task is like holy cow thank you for that pickle fact I’m amazed you’re very well I need to I need to adventure New England is only a jump in a hop skip and a jump in northern for me so

Jake Ward 49:10
yeah, they’ve got to Vermont for sure. Probably New Hampshire but but in my backyard in Maine where I grew up we had

Erin Mikail Staples 49:19
two yields last class

Brian Rinaldi 49:23
yeah be interesting I think you know maybe if we made this such a thing that maybe future guests will bring the Kofax with them

Erin Mikail Staples 49:33
and you want to be guest

Brian Rinaldi 49:35
you’re now the second Jake you’re now the second person who came on who’s not a fan of pickles so you know the last person just flat out like I don’t really like anything pickled and then I was Aaron me really like shut that down. She’s like this interview is over.

Erin Mikail Staples 49:56
The other thing is like honestly, like we let him have it. We let him Should we let him get away with other fermented things? And he, we came to the conclusion that there are other fermented things, which is technically pickled. So yeah,

Jake Ward 50:08
so I mean, I like things that are pickled. It’s pickles that I don’t super love. And it’s mostly dill pickles. Like, you know, big standard dill pickles are not my fav most of

Erin Mikail Staples 50:19
them are like, honestly, they don’t like unless they’re done like with the good like I liked like the good like spicy you. Yeah. Yeah, yeah, put some flavor like otherwise like, traditional elastics don’t always do it for me.

Brian Rinaldi 50:31
Yeah, that’s why I gotta try and Aaron pickle at some point.

Erin Mikail Staples 50:35
This spicy. Yeah, they’re good minor. I like spicy. And like, I put a lot of mustard seed in it.

Jake Ward 50:40
I guess. I like it. I need that. Yeah.

Erin Mikail Staples 50:43
Crunchy.

Brian Rinaldi 50:46
Well, this was this was. This was such a fun conversation and really interesting, Jake. I really enjoyed learning about your background, because I mean, it’s just so you come with a very different perspective than really anybody we’ve had on the show up till now. So really appreciate that.

Jake Ward 51:04
Yeah, the last a couple times in your career to end up in the seat. But I’m happy to be here.

Brian Rinaldi 51:12
All right, Jake, and Aaron less I will say goodbye. We will be back next month. We’ll have details on CFE.dev soon. We also have an upcoming episode of too full to stack later this month. About next Jas and in episode. Sorry, a presentation by Rozelle Scarlett coming up in early December about what five which I know nothing about. So I’m really interested to learn about that. So hopefully we’ll see you all there. Thank you all for attending. Thanks, Jake, for being our guests. And we’ll see you next time. Bye.

Jake Ward 51:49
Thank you. Thank you guys love the show. Thank you