host intro

Hey there, welcome back to the show. I'm John Steele. This is the Funnel Vision Podcast, episode six. And I wanted to talk about a trend I've been noticing in marketing technology. So if you follow marketing technology the way that I do, you notice something that tends to happen a lot, which is founders, engineers will launch some sort of an app or tool. And it builds a large following among developers. GitHub, Product Hunt, these places where developers tend to pick up new tools and ways of building things.

And it develops a huge following in the open source community. So you know, sort of a free-to-use version of the software. It develops this huge following, and then the company uses that to gain attention from investors and try and grow. And their go-to move is to always broaden the search out. They they don't want to monetize their own community. they'd rather broaden it to a new type of buyer, a new sort of corner of the business world.

And I was really wondering why that tends to happen. So I w connected with my friend Janessa. Janessa Lantz is an independent marketing consultant now, but she was the VP of marketing for dbt Labs, a very technical company that launched a tool in open source that is sort of a data transformation tool. It sits inside of data warehouses and sort of helps you, you know, move data between different tools. Something that developers are really interested in if they're not interested in really like a full stack database or a full stack tool over top of their data. You know, if they want to build something like that from scratch, they have to transform the data a bit. And it's something that was very popular among developers in the open source community. So they really built this big community. And to hear or tell it, they it gave them a halo around their brand that that let them sit close to really behemoth companies and household names within the tech community, it really gave them a leg up.

So she's seen how an open source product can then pivot to trying to make more money, trying to grow, trying to draw investment, things like that. And she's seen that this process can play out sometimes. And so I wanted to talk to her about that a little bit. She also spent a few years at HubSpot, so we get into that a little bit, but I really wanted to talk to her as well about the changing positioning of engineering within companies. The line between technical and non-technical is really starting to blur a lot. And so I wanted to talk to her about how she sees that. And when we talk about how she's using AI within her own consulting business, it's a really wide-ranging conversation and I think you'll get a lot out of it. So let's get into it.

interview

John Steele (00:00) Janessa Lantz, welcome to the show.

Janessa Lantz (00:01) Thanks, John. Great to be here.

John Steele (00:03) Yeah, thanks a lot for coming on. So I wanted to talk to you about a number of things related to, you know, marketing. We're both in the marketing field, and I really was interested in, you know, how you've been able to market really technical tools to some more non-technical buyers and some of the the the sort of issues that may arise around that. But I wanted to start by just

John Steele (00:27) introducing your an independent consultant with your own consultancy. tell us a little bit about how you got started with that and you know how you've sort of differentiated yourself from all the other consultants that are out there.

Janessa Lantz (00:39) Ooh, good question. So I've been doing this for about two and a half years now, coming off of a five year stint at dbt Labs, which was in the developer tooling space. I did another stint with a data company before that. did some time at HubSpot and then really post dbt Labs, I've been working primarily with companies in infrastructure, developer tooling, data, that type of thing. So if I if I specialize in anything, it's really around the type of products. And so I've always

Janessa Lantz (01:07) been or recently you know the last 10 plus years of my career have been marketing technical products to a technical buyer

John Steele (01:15) Yeah, so with your consultancy now, you know, how do you find it's different working than working institutionally inside of a company like dbt Labs?

Janessa Lantz (01:24) It's wildly different and in some ways it is wildly the same. So

John Steele (01:30) Mm.

Janessa Lantz (01:30) I think that the the biggest thing is you you don't have the same kind of big team dynamic. And so if you love or or hate the consulting or fractional life, it usually comes down to how you feel about building with a a team over time, right? And because you don't have that when when you're

Janessa Lantz (01:48) By yourself. A lot of the problems look the same. Everybody needs help with messaging and positioning. Everybody feels like they could be more clearly articulating the value that their product offers. Everybody is thinking about

Janessa Lantz (02:03) AI visibility and how to, you know, market in a way that is appealing to both developers and less technical users who are now empowered with AI and agents. So a lot of the, you know, the the problems are are very similar. Some of them are new because of AI. But, you know, the work of marketing is always about

Janessa Lantz (02:24) knowing knowing who our customer is, who our buyer is, what is what concerns them, what are the problems and challenges that that they have and and how how do we build something that solves that for them. And so in that way the work of marketing is always the same.

John Steele (02:38) Gotcha, gotcha. And I you mentioned dbt Labs. I definitely wanna talk to you a bit about that in just a minute. But I wanted to ask about you mentioned you worked at HubSpot. and I I would love to know what that experience was like. I've been a you know a HubSpot shop.

John Steele (02:55) here. I have installed them for about 15 years in different capacities doing marketing work and I've always found the company to be and their and their products to be to be kind of fascinating and so I was curious what it was like to work there.

Janessa Lantz (03:08) Man, I have so many good things to say about HubSpot. Just a phenomenal company. I was there after they had gone public. They were around 2,000-ish people. the marketing team was maybe around 200 people. So this would have been, I think I started there in 20 the end of 2016, and I was there for about two and a half years. just a phenomenal company.

Janessa Lantz (03:31) company. Katie Burke was the was the chief people officer when I was there. I think she's at Harvey now. And I I really from the org that she built, I learned so much about how to do recruiting really well, how to run a good hiring process. I j I got to work with the the fountain both of the founders and the executive team on building like an executive communications program and and was there, you know, supported on the launch of their

Janessa Lantz (03:58) Cust their customer product. I think it was called Customer Hub. I don't know why I'm blinking on that now. So just a r a really remarkable organization.

Janessa Lantz (04:05) You know, and still have some fantastic friends from that company who who I I met there. So, you know,

John Steele (04:11) Mm-hmm.

Janessa Lantz (04:12) huge fan. And and I learned a ton there, obviously, about marketing. but I think that that some of the best things that I took away from my time at HubSpot were really just like how to how I want to show up as a leader. so I

John Steele (04:25) Mm-hmm.

Janessa Lantz (04:26) worked with Meghan Keaney Anderson, was was the VP I worked under, just a phenomenal product marketing leader. Kipp Bodnar, who's the

Janessa Lantz (04:33) the CMO who He just did such a good job.

Janessa Lantz (04:36) always listening to everybody on the team and hearing them out and then ultimately, you know, setting a course and leading leading people toward that. Just incredibly

John Steele (04:44) Mm-hmm.

Janessa Lantz (04:45) calm, chill, but impactful leader. And so seeing and working with these people, I think gave me a much better sense of the kind of leader that I want to be. and so I I was there in like my

Janessa Lantz (04:58) mid career years so like I I think like eight or so years of experience. And while I don't like it's always a people always have to think about like does a big what is the purpose of a big company name on your resume? and

John Steele (05:12) Right.

Janessa Lantz (05:12) for me, I loved HubSpot. They were transformative in the early years of my career and they were like a dream company to work for, but really

John Steele (05:22) Right.

Janessa Lantz (05:22) it was about the people that were there. and that's

John Steele (05:25) Mm.

Janessa Lantz (05:26) not true, right?

Janessa Lantz (05:26) You can work at a lot of companies that have great names

John Steele (05:28) Yeah.

Janessa Lantz (05:29) and they may or may not align with your personal values. so for me, I continue to look back at my HubSpot years and I'm just so grateful for it and learned a ton there and just continue to be a huge fan of the the company.

John Steele (05:41) And it it had to be, you know, kind of a shift, you know, going from from that to you worked at dbt Labs right after that. And so for people who don't know, dbt Labs, it they're a data transformation company, is that right? And it it's inside of a like a data warehouse. is

Janessa Lantz (06:00) Yeah. Yeah.

John Steele (06:01) that is that right? So so it's yeah, it's a so that's a pretty technical product. something that you know people that are

John Steele (06:08) in like IT CTOs and things are are concerned with. that's that's quite a shift from the the main customer of HubSpot, is it not?

Janessa Lantz (06:15) Yeah, so there there is some continuity there. before HubSpot I was a company called RJMetrics, which is where I met the founder of the founder and co-founders of dbt Labs. We were all at RJMetrics. RJMetrics got was like a full stack business intelligence tool, got completely beat up by the advent of cloud data warehouses. And so it felt like at the time, you know, this was twenty thirteen, twenty fourteen, that overnight people were just like, I don't want a full stack BI tool. Like just get my data.

Janessa Lantz (06:43) into the warehouse and and I'll take it from there. And so that's

John Steele (06:46) No.

Janessa Lantz (06:47) really where the company built a a data pipeline tool and then needed to do some data transformation on top of that. you know, just it

Janessa Lantz (06:55) changing dates and

John Steele (06:57) Mm-hmm.

Janessa Lantz (06:57) you know all the all the things currency formats, right? And so that's where the the first version of of dbt was built, dbt Core. And so it it was kind of a the certainly the transition from HubSpot to to dbt looks like a rather big jump. But I was also, you know, had had seen quite a bit of transformation in the data space

John Steele (07:19) Mm-hmm.

Janessa Lantz (07:19) and knew a lot of the people and had continued to do some consulting work with the

Janessa Lantz (07:23) With the early dbt team then Fishtown Analytics, they were you know friends of mine still are. And so it in that way it wasn't that big of a jump. And really, I think HubSpot was a time for me to dig deep into the communications and executive comms and really working with like what does it feel like to work with a like a much more executive stakeholder?

Janessa Lantz (07:45) Which ended up being incredibly valuable as I then, you know, built the the marketing team. So there was some continuity there. but at this point in my career, I'm like, I'm not a I am in no way a a martech marketer. that

John Steele (07:58) Mm-hmm.

Janessa Lantz (07:59) that was kind of the the brief thing, but I got so so many other great things from that experience.

John Steele (08:04) Yeah. And people who follow this podcast and have seen our previous episodes will know that we we did a whole episode with a founder who left HubSpot and was building their own thing and they would definitely have needed tools like tools like dbt Labs for that. And it sounds like you got in right at the beginning of that. And a lot of people are are

John Steele (08:22) pursuing that now. I've seen a lot of people online like building in public, like we're trying to build a new thing outside of you know full stack tool like you mentioned. So it's interesting to to have that. So you've you've said many times that you're you don't consider yourself that technical and I'm curious as to you know how you were able to bridge that gap of of maybe not being so technical and and trying to

John Steele (08:47) at least initially sell to a very technical sell a very technical product to a very technical audience.

Janessa Lantz (08:53) Yeah, yeah.

Janessa Lantz (08:54) The the thing that I always aim to do and and always encourage marketers on my team at dbt to do is to build deep relationships with the the people who have technical expertise. rather than try to fake it yourself. You have to know, you have to get good at knowing when to bring people in. And so you do need to be curious. I mean, I like when I was starting with dbt, I didn't have access to ChatGPT and Claude, and I did a ton of Google.

Janessa Lantz (09:22) I did a ton of sitting with Drew Banin, who was one of the co-founders, and asking him just really dumb questions. Like, tell me, tell me again, how do APIs work? How do webhooks work? How do you know what is the difference between these types of data materializations? You know, the founder of dbt sat down and gave me this, you know, little little tutorial about like data warehouses and structured data using like a like a just a spreadsheet and saying, hey, how

Janessa Lantz (09:49) How do you think about everything in columns and rows? And so that that was incredibly helpful and certainly you have to be willing to lean into learning about technical details. And like my first week at dbt, it was very important to Tristan, the CEO, that

Janessa Lantz (10:06) I understand the workflow, what we call the analytics engineering workflow. And so I worked with an analytics engineer at the company and did some transformation on a Mailchimp data set, you know, loaded into

John Steele (10:18) Mm-hmm.

Janessa Lantz (10:19) the warehouse, did some basic transformation, committed it to GitHub, you know, did did the whole process and it was a pain. It was like very difficult.

John Steele (10:25) Mm-hmm.

Janessa Lantz (10:26) But wow, I'm really grateful for that today.

Janessa Lantz (10:29) Right. Because I actually made

John Steele (10:29) Yeah, absolutely.

Janessa Lantz (10:31) a a few, you know, a handful of GitHub commitments and commitments, GitHub commits in in you know twenty nineteen that

John Steele (10:37) Commit, yeah. Yeah.

Janessa Lantz (10:40) that that feel like I'm really happy to have had that experience because a lot of marketers in 2026, you know, are having that experience for the first time. So

Janessa Lantz (10:48) I would say overall lean in and be very uncomfortable, right?

John Steele (10:52) Mm-hmm.

Janessa Lantz (10:53) Ask a ton of questions, but like man, build relationships with the with the people who who are the audience that you're marketing to.

John Steele (11:01) Right, right.

Janessa Lantz (11:01) And,

Janessa Lantz (11:02) you know, we we constantly did this at dbt, like, is this tweet funny? 'Cause we were s it was still Twitter then, you know, is this tweet funny? Is

John Steele (11:08) Yeah.

Janessa Lantz (11:09) this meme funny? Is this is this right? Is this the r exactly

John Steele (11:12) Right.

Janessa Lantz (11:12) the right way to say it? And we would get, you know, people who were the audience from all over the the company giving feedback. The the third thing I would say as part of that is that you have to have a mix of technical expertise on the marketing team. So we

John Steele (11:27) Mm.

Janessa Lantz (11:27) did have people who were

Janessa Lantz (11:30) more of the audience that we were marketing to. So it's like some of the people, you know, were your classic marketers. They built their whole career in in marketing. Some of them were more of like the DevRel community type people. But you need that mix of people who who do just like know exactly how to how to speak to the the person.

John Steele (11:47) Right.

John Steele (11:47) Yeah, you you mentioned, you know, running messaging past more technical people within the company. I I I had a similar experience. I worked at a boutique engineering firm for a while, and man, is that important? Because if you you say the wrong thing or whatever, I mean you can get roasted by people online. You can look really unprofessional. You can, you know, so it's really it's

Janessa Lantz (12:07) Yeah, yeah.

John Steele (12:07) really important to to word things the right way. So, you know, for our non-technical listeners, people watching the video, maybe just quickly take

John Steele (12:15) them through how dbt's product works and sort of why the open source version of it was so exciting to these technical people.

Janessa Lantz (12:24) so previously, you know, a data analyst would have built, you know, a a SQL report, right? It could be as simple as saying like,

Janessa Lantz (12:32) What what is our what is our revenue this month?

John Steele (12:34) Mm-hmm.

Janessa Lantz (12:35) Well that seems really easy. But when you say revenue, is that pre-tax or post-tax? Right?

John Steele (12:41) Mm.

Janessa Lantz (12:42) When you say customer, well what about a churn customer? Right?

John Steele (12:45) Right.

Janessa Lantz (12:45) Or is it a product user? Is it somebody who's ever paid you a dollar? right, and so so almost every single question that you have

Janessa Lantz (12:53) you realize there's a qualifier. you know, the joke in the data space was always like, what do you mean by a day? Because literally,

John Steele (12:58) Mm-hmm.

Janessa Lantz (13:00) depending on time zone, what you call a day is different. And time zones are like one of the they're just a huge pain in the ass for anybody who works in data. Right. And so

John Steele (13:08) And and anyone

John Steele (13:09) that's worked within a HubSpot understands this because of like fields and objects and things within a HubSpot. You can

Janessa Lantz (13:12) Yes, yes, yes.

John Steele (13:14) customize all those things that go in there for your own company. So you will define things differently based on the use case and based on the company that you're doing it for. And you need

Janessa Lantz (13:24) Yeah, why why

John Steele (13:25) engineering to be able to do that.

Janessa Lantz (13:27) Why are the Salesforce why why does sales always have a different data set than the than the marketing team? some of that just has to do with how how Salesforce thinks about a day, right? How we how we mark time. Some of that has to do with when the different tools sync the data, and then some of it

John Steele (13:43) Mm-hmm.

Janessa Lantz (13:44) has to do, you know, with with how you describe the the data around that. And so what was possible because of SQL becoming a more powerful language is that you have

Janessa Lantz (13:54) All of these people who are data analysts who know SQL and understand the business, and

John Steele (14:00) Mm-hmm.

Janessa Lantz (14:01) then you had kind of all these people who were data engineers and knew Python, but often didn't have all of the business context. So suddenly the people who have the business context have the tools.

Janessa Lantz (14:11) To do the data transformation work, to say, here's exactly how we want to describe a customer, a business seg segment, a metric when we talk about you know, a conversion, you know, what what do we mean by that? What are these different events? So they had all of these tools that they didn't have to go to a data engineer and wait around for them to do it anymore.

Janessa Lantz (14:31) They could do it themselves because SQL was a more powerful language and because warehouses were so much more powerful, and because you then have a tool like dbt that's saying this work can persist. So rather than building like a one-off query and saying, hey, data analysts, run the report so we can get the board numbers. anybody can anybody can have access to that and and run it whenever, and other analysts can collaborate on it, you know, and edit it and sort of like build knowledge together over time.

John Steele (14:58) Mm-hmm. Great. So I think I think that's gonna be really helpful for people to who may not be familiar with these types of tools to understand, you know, how much thought and energy goes into this if you don't have what you said, like a big giant tool on top of it controlling everything for you. If you wanna have more fine grained control of your data, you know, tools like this can be can be transformational, as you said.

John Steele (15:22) I wanted to talk about community a little bit. So let's see. So I guess I wanted to originally have you on because I had started with a client call I had. So I got approached by a company that had built this AI product that was open source, which means that

John Steele (15:38) you know, it was out there for the code was available for people to get on GitHub. They could, you know, take it, try and, you know, play with it themselves, things like that. And they got a big following and sort of a community groundswell of people that were interested in this product. And they got a ton of stars on GitHub and things like that and people talking about them. Well then they realized okay, you know, we really want to monetize this. We really wanna grow. And

John Steele (16:04) It became very clear to me in talking to them that they were a

John Steele (16:07) approaching they they didn't really want to approach their community really at all they were more moving towards more of like a marketing audience and I feel like I I see this happen a lot and I was interested in exploring why that might be because you know you see these people that build these really technical products they develop a community around it they develop a technical following a ton of people are really interested in what they've built and then once they go to grow you know that's not where they start they

John Steele (16:36) kind of want to we need to branch out to a new audience and so your experience at dbt was interesting to me because your product did develop a groundswell of support from technical users in a similar way and it started from this open source product is that correct

Janessa Lantz (16:51) Mm-hmm, mm-hmm.

John Steele (16:53) So

John Steele (16:53) so so talk a little bit about first of all for the non-technical audience, well, what what is open source? Explain it a little bit better than I did there, if you if you could. And then also just talk about the process for you guys of like building this community, you know, fostering it, and then also being able to build you know, a successful paid operation as well.

Janessa Lantz (17:15) Yeah, okay. So

Janessa Lantz (17:17) Open source is open source software is basically just software that is free and available and anybody can use it. it's usually very easy to run locally on your machine. And then where people run into problems is when they're saying, we want to deploy this across the organization, and you have to think about infrastructure and security and you know managing multiple users and governance and all this other fun stuff, and then that's that's

John Steele (17:41) Mm.

Janessa Lantz (17:42) when you you start to think about the the software around that.

Janessa Lantz (17:44) But basically open source software is just

Janessa Lantz (17:47) free. So like for dbt, you could just download it and use it. the thing that every marketer got tripped up on is they'd say where's the email list of all your open source users? You're like, we don't have

John Steele (17:58) Yeah.

Janessa Lantz (17:59) it, right? We don't know who these

John Steele (18:01) Yeah.

Janessa Lantz (18:02) these people are. and so it is a it is a very different motion and you often see, you know, much more of a community motion around open source companies for for this reason because the the software is often more

Janessa Lantz (18:13) difficult to use. And so what was unique, I I think, about the dbt community is that we were really we were really helping a new group of users work in a new way. And

John Steele (18:26) Yeah.

Janessa Lantz (18:26) so analysts already knew SQL, but what they didn't understand what were often the like software engineering best practices that could make them much more effective. And so we were talking

John Steele (18:38) Mm-hmm.

Janessa Lantz (18:39) about version control and they were learning

Janessa Lantz (18:41) GitHub and we're talking about the need for testing, right? So if you're talking to a developer, they're like, obviously, I'm gonna test my code. and an analyst

John Steele (18:48) Mm-hmm.

Janessa Lantz (18:49) is like, What do you mean? Write data tests like this is I've never thought about this before. modularity, all of these things were like concepts that were that were established for one audience, and we're suddenly saying, Hey, data analysts, you you should learn these too.

John Steele (19:07) Mm.

Janessa Lantz (19:08) And so,

Janessa Lantz (19:08) I think that that's just a use case for for community. Like everybody wants to have a community for their product. when when is it a good fit? I think is when you have a group of people who are l learning something new together. And really what we're talking about is an identity shift where these these analysts went from saying, I write SQL over here all by myself.

Janessa Lantz (19:29) To saying, I can be like, I can adopt these new ways of working and be far, far, far more impactful in my in my role. And so it comes with like thinking about your role in a different way, rather than like, I provide answers with data to almost more like I build I build systems and architecture that deliver answers for the the entire company.

Janessa Lantz (19:51) And so in that way it was a that was it was an enormous shift. And so the early version of community at dbt was really our co-founder Drew, he probably did most of the community presence. I think he sent some 30,000 Slack messages in the first two years of the company, some absurd number.

John Steele (20:06) Yeah.

Janessa Lantz (20:07) Like he was just answering questions, right? So it was almost like community of support. though lots of meetups and office hours and all that good stuff as well. and then eventually it became more about events and bringing people together. and then in

Janessa Lantz (20:19) By twenty twenty we raised Series A and then started moving into the commercial motion. And then it was more selling the the what was then called dbt Cloud product and saying how do we get enterprises to adopt this?

John Steele (20:32) So I why do you think that

John Steele (20:34) That transition, you you mentioned Series A funding round, and this is kind of where I see the this sort of transition happening a lot, which is people start to raise money or they start to want to grow in some some concrete ways. And that's when pivots start to happen. They start to go, we need to approach a new audience or a new type of buyer, or we need to get more interest from a smaller group that we haven't established as many inroads with, something like that.

John Steele (21:00) why why do you think, I mean maybe that's not exactly how it worked at dbt, but why do you think that that happens? Why do you think that you know technical buyers are not seen as the growth play out the gate for for a lot of products like this?

Janessa Lantz (21:12) Well the the challenge that that we have and a lot of open source companies have is that when you're talking to individual developers, they want to do it themselves.

Janessa Lantz (21:19) They like doing it themselves.

John Steele (21:21) Mm-hmm.

Janessa Lantz (21:22) I, as a recently empowered with AI person, also like to do it by myself, right? And I I feel much more in that camp today. I'm like, just let me do it. I don't have to wait for somebody. but you can see some of the ways that goes wrong. it often optimizes for a single player experience. it often is much faster, but less scalable. It can introduce security problems, right? All of

Janessa Lantz (21:46) The things that we're seeing with how AI creates speed is similar. And so what will happen is if you're selling to developers, they'll say, I don't really want to pay for the cloud product. I can do all that myself. and so then you need to sell to the CTO or the VP of data, and you're not having a conversation about technical capabilities, you're having a conversation about how do you onboard, how do you govern, how do you you

Janessa Lantz (22:12) think about rule based access control and who can touch what data sets.

John Steele (22:17) Mm-hmm.

Janessa Lantz (22:18) you know, you have your public company and you there are all of these rules about how your data is audited and and formed and you have to make that, you know, visible and accessible. so do you really just want like one or two developers hacking on this thing or do you want a whole, you know, hardened system a around it? And so it is it is a more like this problem is is more valuable, but it's also that you can't assume that a technical person

Janessa Lantz (22:46) is the person who actually has the change management power.

Janessa Lantz (22:50) to to make this larger change happen, right? So when you're saying you know, we want to get 200 analysts at JetBlue using this thing, well, that's a very different problem, right? Than

John Steele (23:02) Mm-hmm.

Janessa Lantz (23:02) saying, how do I, one or two developers, get started with with dbt. And so I think that's that's one of the that's one of the bigger reasons. I'm not sure if that's exactly what you're talking about with your client, where it's it sounds like they're almost talking about a different

Janessa Lantz (23:16) persona. but certainly, you know, when you're saying, hey, we just raised a bunch of money and we want to sell higher contract value accounts, well then you have to think about saying, not just how do we get one person adopting, but how do we get 200 people adopting?

John Steele (23:31) Sure.

John Steele (23:32) Sure. Well why do you think founders then and and people that that build apps and things like that, why do you think they they focus on technical adoption first then? If if building that community is not going to be a an exact, you know, one to one for their their sort of growth play? Is it is it more so that engineers within companies are like the early adopters and you sort of have to convince them first before you can even have a conversation higher up or like why do you

John Steele (24:01) think it so often starts there where they build this big technical following first?

Janessa Lantz (24:06) So this is this is the this is the heart of a marketer's love-hate relationship with open source software. Very, very, very raw feelings about this. So, what is just the freaking best thing about marketing and open as an open source company is that when developers love your product, they tell other developers. And so we just had this brand halo around everything that we did because there were thousands.

Janessa Lantz (24:34) Of companies using this product. And so

John Steele (24:37) Mm-hmm.

Janessa Lantz (24:38) our customers, you know, even as a Series A company, our customer base was very small. Our revenue numbers were very small, but we just had so many people online talking about how much they adored the product, showing up at conferences. There were one of my favorite metrics to track was the Snowflake Conference, which was a huge event for us. And I would always look what vendors.

Janessa Lantz (25:01) were showing up the most often in talks. It was always dbt. It was like Snowflake itself and then it was dbt or like AWS. Like that is bonkers, right? Like we were

John Steele (25:09) Mm-hmm. Not it, yeah.

Janessa Lantz (25:11) nobody. And so that is why you start with the developers and the open source. Because

Janessa Lantz (25:17) We could get conversations with anybody. People wanted to do co-marketing with us. Companies that were 10 times bigger than us were willing to give us the time of day because we had this incredible community. The first conference that we did, we had almost 2,000 people attend, right? Because not because we had a huge customer base, but we but because of this this open source halo. And so that in that way, it is just like a marketer's dream.

Janessa Lantz (25:42) And then the downsides, which is you are always marketing against a product that is free. And this was very painful for us, you know, when the whole

Janessa Lantz (25:51) With the whole tech pullback of like 2023, and suddenly people were tightening up budgets and funding was tightening, all of all of these downstream effects of the change of interest rates, and and and suddenly we're going into every single account, and we have to justify why you would pay for the thing when you could do it for free. So then you have to be able

John Steele (26:11) Mm-hmm.

Janessa Lantz (26:11) to quantify developer hours and security risks and get really good at telling a value story. But that's why. Like the momentum.

Janessa Lantz (26:20) out of the gate of a free product is just phenomenal. And I think

John Steele (26:24) Right.

Janessa Lantz (26:25) that in in this day of where distribution is like the hardest thing ever for

Janessa Lantz (26:31) all every product out there it is it is certainly something to to think about, you know, and I think you see the Lovables and the Replits, you know, they're giving a ton of value away for free because that's how you get people to kick the tires and and, you know, talk about how great it is. Some of those people

John Steele (26:47) And and you're right, you are seeing them everywhere. I mean, Lovable and Replit

John Steele (26:51) are getting mentioned everywhere. Every tech article you read is is mentions them,

Janessa Lantz (26:54) Yeah, yeah.

John Steele (26:55) it feels like. So yeah, that's a that's a good point. You do you do see them a lot. but you you

Janessa Lantz (27:00) I just did a syn

Janessa Lantz (27:02) I just did a synt Kilo Code as well. They were since they were just acquired recently and and they did the same thing of like they gave credits away, they were everywhere, they're at conferences, they're saying just like try it, try it, try it, try it. You do

John Steele (27:14) Yeah.

Janessa Lantz (27:15) this thing, we give you credits, we give you credits because that's ultimately what you need is like you

John Steele (27:20) Mm-hmm.

Janessa Lantz (27:20) need you need users and you need them to then go online and and tell other people about it.

John Steele (27:26) Right. Yeah.

Janessa Lantz (27:27) so yeah.

Janessa Lantz (27:28) That's why.

John Steele (27:29) Yeah,

John Steele (27:29) you were you talked a little bit there about the the limits of of building community and and sort of what can happen there. you know, I know a lot of companies will sometimes move features that were in the open source to like the paid side and that can create tremendous like backlash from from those communities. how do you you know

John Steele (27:48) Being in marketing, how do you deal with stuff like that where there are changes that happen and you've you've angered the community? And obviously you you you don't want to upset the brand halo, but sometimes there are you know there are realities that have to happen within companies. So how how do you sort of balance that with marketing the the paid product?

Janessa Lantz (28:06) Yeah, I we certainly didn't always get this right at at dbt, though I think we had some some very good advisors from companies like MongoDB and HashiCorp and you know, companies had who had misstepped on this and and and done it right. I always like the the HubSpot model which which they were always quite intentional about was that new features are released at the enterprise and then they kind of move down so you actually make new features available for more people.

Janessa Lantz (28:34) And so

John Steele (28:34) Hmm.

Janessa Lantz (28:35) a core part of their philosophy was always that you you really want to like you you almost want to test a new feature.

Janessa Lantz (28:42) And then like at as it gets good, you know, you make it available to to everybody. so

John Steele (28:47) Mm-hmm.

Janessa Lantz (28:48) so that works, you know, I I it that that worked for them in some respects. Again, I'm talking about an approach to pricing that was like to you, it's like almost ten years old now.

John Steele (28:58) Mm-hmm.

Janessa Lantz (28:58) but but there are some things like that where you can think about what is made available when. you can think about releasing things as beta, right? So it's like you can use

John Steele (29:06) Mm-hmm, mm-hmm.

Janessa Lantz (29:08) it now, it's free now. It may not be free forever.

Janessa Lantz (29:11) at dbt there was a lot around like licensing, you know, and so you know, what what are we putting under what license? And so when the when the company released the new engine called Fusion, right, then that was under a a different kind of license. And so there there are ways around this, but in almost every case the approach is to say is to avoid taking something away.

Janessa Lantz (29:32) from your customers. That's what you

John Steele (29:33) Right. Yeah.

Janessa Lantz (29:34) don't want to do. And the easiest thing, honestly, is like put a beta tag on it. Right. And communicate.

John Steele (29:39) Right. Right.

Janessa Lantz (29:41) This is free for now. We don't know when it's going to stop being free. But like you can use it. And that's really what you see

Janessa Lantz (29:48) Anthropic and and OpenAI doing, right? So it's like,

John Steele (29:51) Right.

Janessa Lantz (29:52) hey, we're very clear Fable's not gonna be free forever, but but right now it

John Steele (29:55) Mm-hmm.

Janessa Lantz (29:56) is. And and they landed they landed that message, right? A little waffly with, you know,

John Steele (30:01) Yeah.

Janessa Lantz (30:02) what is included where, but ultimately I think that is the right approach is to say it's it's it's free for now.

John Steele (30:09) Yeah, so the line between technical and non-technical is obviously starting to blur. I wanna read you this. This is from a Gartner report. Gartner estimates that by 2026, citizen developers at large enterprises, that's people not in traditional IT roles.

John Steele (30:27) at large enterprises will outnumber professional software developers by a factor of four to one. According to Gartner, by 2026, developers outside formal IT departments will account for at least 80% of the user base for low-code development tools. Almost 60% of custom apps are now built outside the IT department. Of those, 30% are built by employees with limited or no technical skills.

John Steele (30:54) Obviously AI is a big bridge here, but you know, I I'm curious what that starts to look like within companies where, you know, what what sort of happens to the developers in this case? What you know, what what are how are their how is their role going to change? I mean, you you've been working with these people for many years, you know.

John Steele (31:17) How are they seeing this time? I mean, have you spoken to to anybody about that? Like how they see their job kind of changing now?

Janessa Lantz (31:22) You know, I man, we're all we're all in it right now.

John Steele (31:27) Okay.

Janessa Lantz (31:27) I I do think that in in in every role the expertise and judgment that we have all accumulated is incredibly valuable and is going to be, you know, for for some amount of time. But these distinctions around, you know, who is technical and who is non-technical,

Janessa Lantz (31:44) feel very strange to me these days. Because what do we mean by that? Right? Like,

Janessa Lantz (31:50) You know, I I

Janessa Lantz (31:52) spent a couple hundred bucks on Anthropic's API because I set up an API key and accidentally had my Claude configured to call the API key instead of my subscription. Right?

John Steele (32:02) Mm.

Janessa Lantz (32:03) It's like a stupid mistake that every single developer has some version of that story, right, with some tool. So it's

John Steele (32:09) Right.

Janessa Lantz (32:10) like, yeah, it's like I get to make developer level mistakes you know, as a as a non technical person. So

John Steele (32:17) Mm.

Janessa Lantz (32:17) if I'm a person with an agent

Janessa Lantz (32:19) Am I still non technical? Am I dangerously technical? And so

John Steele (32:23) Right.

Janessa Lantz (32:23) I think that we're all dealing with versions of that, like.

Janessa Lantz (32:26) We can all reach a little bit deeper into the technical stack, right? So I'm here and I

John Steele (32:31) Mm-hmm.

Janessa Lantz (32:31) can go here. but certainly, you know, a dbt user of 2020 can almost be a software engineer today. And a software, you know, what can what is possible for a software engineer today? well, now they can manage, you know, 50 agents at a time and, you know, be tw 20x more productive. So I think that we're

John Steele (32:49) Mm-hmm.

Janessa Lantz (32:49) all sort of pushing the technical boundaries of our role.

John Steele (32:53) Right.

John Steele (32:54) Do do you think it goes to security at all? Do you do you 'cause I mean I know that's a big concern for people as AI rolls out and I know

Janessa Lantz (33:01) I don't think we care about security anymore. I think

Janessa Lantz (33:04) we did. I I think that the the SOC 2 vendors haven't caught up to AI security, but but I don't I don't think that's a concern for us anymore. It seems like we've just sort of moved on.

John Steele (33:14) We're in a we're in such an experimental phase with AI that people are just like, forget it. Like we'll just like

Janessa Lantz (33:18) Yes.

John Steele (33:19) we'll just take the risks and see what happens.

Janessa Lantz (33:21) Yeah,

Janessa Lantz (33:21) it's like you have you have you need two factor authentication for everything and we're like I don't know, just MCP

John Steele (33:24) Right.

Janessa Lantz (33:27) access, like go go crazy. so I

John Steele (33:29) Yeah.

Janessa Lantz (33:30) don't know. I'm sure somebody out there is very worried about security. I feel like overall security is not I would say everybody cares about speed today and we've sort of been willing to like, you know, cr cross our fingers on security for the time being.

John Steele (33:44) Yeah, yeah. And and sort of in a in a related question I would I would ask, you know, where do you see the value of of technical skills for marketers today? So like do you think there's value in like maybe not learning how to code because like you can get AI to code for you really better than you would learn how to do? But I mean, do you think there's value in like understanding how code works? Do you think there's value in understanding how these types of data warehouses, why they're

John Steele (34:10) i interesting and important to companies and things like that. I mean, sort of the inputs and outputs of it all. I mean, d do you think that's something that marketers should should focus on?

Janessa Lantz (34:19) Do

Janessa Lantz (34:19) I I really do think it matters and and I really wish somebody would make a a you know engineering or computer science for marketers program.

John Steele (34:28) Yeah.

Janessa Lantz (34:28) there there are a few ways where where this shows up. Like one of them is the there there are some concepts I picked up for my dbt dates that have just been enormously valuable, and one of them is this idea of how data moves through the pipeline. So you have your your raw table.

Janessa Lantz (34:44) Which is data exactly as it's pulled from HubSpot, no changes, right? All you're trying to do is move the data from point A to point B. Because there are actually a lot of things that can go wrong there. So you just want to be able to look at this and say it exactly matches that. And then you move into a staging table, which is saying, okay, I've now structured the messy HubSpot data into a way so my time zones are always consistent, my fields are named as I want them to be. A human can a human can read this and make sense of this data.

Janessa Lantz (35:14) You're still only on one source. And then as you get into your analytics tables that are used by the business, now you're combining sources, right? So now you're saying, here's my HubSpot data combined with my Google AdWords data, so I can now measure exactly which ad campaign is resulting in pipeline. So you're combining data sources and able to provide these richer answers to the business.

Janessa Lantz (35:38) Turns out this is a really good model for how to work with agents. So AI gets very confused very quickly, though context windows are growing. And so

John Steele (35:47) Mm-hmm.

Janessa Lantz (35:48) what what I found to be incredibly effective is to structure the inputs, whether that is what is your product marketing messaging, you know, what are the events we want to be at and the speakers and the themes and topics that we talk about to like structure this in some way so that you can kind of let the agent

Janessa Lantz (36:06) move through it in a more step-by-step process. Right. So that's just one example. But there are just a million examples like that. A client I'm working with right now called AirTop, they're they're in this place where like they started as a browser automation tool for developers. So an n8n integration was very much like

John Steele (36:24) Mm.

Janessa Lantz (36:24) hey you you're using n8n or Make now you can combine this with AirTop and you can search behind like behind the authenticated web. You can

Janessa Lantz (36:34) handle captchas you can do all this stuff and increasingly they are they are like moving up that stack of of marketing to a less technical audience but still

John Steele (36:44) Mm-hmm.

Janessa Lantz (36:45) you have to understand like proxy browsers

John Steele (36:49) Right.

Janessa Lantz (36:49) Right.

Janessa Lantz (36:49) And so and so sometimes you can abstract this stuff away. Like if you're a Lovable user, they'll abstract away like Supabase or Resend, right? So you may not

John Steele (36:58) Mm-hmm.

Janessa Lantz (36:59) know exactly what the infrastructure is that is powering your things. Sometimes you might just need to understand, like, hey, why why is why can't Claude get why can't Claude scan LinkedIn? Right?

Janessa Lantz (37:14) And and and you know, a tool like AirTop can. What what what are the what are the differences there? And so you have to understand like, well, what are what are the technical limitations? if you're building with an Anthropic API key, it's probably good to understand when Claude is going to call your API key versus call your subscription so that you don't end up, you know, blowing a couple hundred bucks that you don't need to. Right? And so I I I there are there are so many examples like that, and I think

Janessa Lantz (37:42) it's never been more useful to understand like the fundamentals and I truly wish somebody would make an engineering for for marketers. But there's probably I think there's also a lot of benefit in anything that's like systems thinking.

Janessa Lantz (37:56) Right. So so understanding the architecture and and the basics, it's it's all relevant, right? Like how you set up your Claude project, environments, right? We're we're all learning about this stuff now.

John Steele (38:09) Right. No, that's a that's a really great point. And I think, yeah, the the fundamentals of of how, even how I mean, it's so interesting, like the companies that I've worked at, the engineering department is set up like completely differently in just the way that they approach tasks, the way that they approach getting things done. Like they they s they organize themselves differently. you know, the order of operations is different. So I think even understanding that and understanding what you can learn from that can be valuable for marketers to know.

Janessa Lantz (38:34) Yeah, and there's

Janessa Lantz (38:35) such a shift. Like I think one of the biggest shifts in marketing right now is we're all going from like working on an asset to working on a system, right?

John Steele (38:43) Mm-hmm.

Janessa Lantz (38:43) Completing a task to completing the system that makes the task. And so like the joke about developers for for many, many years was like, they would they would spend, you know, 10 hours building an automation to do something that they could have done in a minute. But like we're

John Steele (38:59) Mm-hmm.

Janessa Lantz (38:59) all making those decisions now. And if this is something you're

Janessa Lantz (39:02) going to do again.

Janessa Lantz (39:03) Then you need to think about how to systematize it. And so it's no longer like we should be, you know, editing an individual blog post. It's like, how do you edit the skill that is producing that blog post? It's no longer, you know, how do you get insights from customer calls? It's like, what is the system that's

John Steele (39:21) Right. Right.

Janessa Lantz (39:22) constantly parsing that data and surfacing that intelligence to you? And so all of that is that's just like a way of thinking.

John Steele (39:31) Mm-hmm.

Janessa Lantz (39:31) that

Janessa Lantz (39:31) developers have that a lot of marketers don't have. And I think it's a reason, you know, I'm seeing this at a at a handful of companies right now where they're saying, you know, our first marketer is our first marketer is just a heavy user of the product who wants to talk about it. That's a great first marketer for a

John Steele (39:47) Mm-hmm. Yeah.

Janessa Lantz (39:49) DevTools company, right? And there's a reason I think that that developers can can move into the marketing sphere. And I think that marketers have a ton to learn.

Janessa Lantz (39:58) from how developers think about particularly building building systems and repeatability.

John Steele (40:03) All right. And our last question, the question that we ask every guest we get on the show. What is an integration, automation, or tool that you can't live without?

Janessa Lantz (40:13) I think right now my current favorite is probably the Granola MCP

John Steele (40:19) Mm-hmm.

Janessa Lantz (40:19) with Claude. I just, it's such a dream. Like putting together client proposals. It's like reference this call, make me the proposal. And so that whole workflow is just like so easy today. And yeah. I mean MCPs in general. my goodness, life-changing. but

John Steele (40:35) Yeah. Really, really cool stuff.

Janessa Lantz (40:39) it's it's incredible. But the Granola

John Steele (40:41) Mm-hmm.

Janessa Lantz (40:42) MCP in particular.

Janessa Lantz (40:43) is just like w wildly useful and

John Steele (40:46) Why why

John Steele (40:46) that one? Why explain explain to the people what Granola does.

Janessa Lantz (40:51) sorry.

Janessa Lantz (40:51) do you use Granola? Are you a Granola user?

John Steele (40:53) I I've used it before, yeah.

Janessa Lantz (40:54) Okay. It's it's a call recording tool.

John Steele (40:57) Mm.

Janessa Lantz (40:57) it's like a it's like a note taker, but it plays really well with like your own notes. So

John Steele (41:02) Mm-hmm.

Janessa Lantz (41:02) it's not just it, which is I mean, okay.

Janessa Lantz (41:06) AI is always better when we give it context. And I think the way Granola works, where you can like type in your own notes and it's recording. And so your own notes kind of inform the way the summary takes shape, just make it a very valuable product in general. And I've been a happy Granola user for many years, but I wasn't a paying customer until I needed the MCP. And now

John Steele (41:24) Mm-hmm.

Janessa Lantz (41:25) you can just you have all of your call, your whole call transcript that you can pull directly into Claude. And so for me, it's like I'm putting together a client proposal. We already talked about.

Janessa Lantz (41:36) rates, we talked about start date, we talked about deliverables, all that stuff is in the call notes. So now you know just go and get it and and put it into a Google Doc. and it's

John Steele (41:46) Mm-hmm.

Janessa Lantz (41:47) just such a simple workflow. I mean the the the magical parts of AI are just like still doing the kind of like the most tedious parts of our job. So that's why

John Steele (41:55) Yeah.

Janessa Lantz (41:55) that's why I love the Granola MCP.

John Steele (41:57) Excellent.

John Steele (41:58) Excellent. Well Janessa, thank you very much for joining us today. I really appreciate it. And I think people are gonna love this conversation. everybody's worked with engineers before and and probably will need to again. And so I think it's fascinating to think about how marketers can can do that better. So

Janessa Lantz (42:13) I mean I hope so. I I don't

Janessa Lantz (42:15) I don't think AI is going to take their jobs just yet. So we

John Steele (42:18) No, I don't think so either.

Janessa Lantz (42:19) we hope that we don't have to we don't have to do it or the the

John Steele (42:22) Yeah.

Janessa Lantz (42:23) agents.

John Steele (42:24) most certainly. I would definitely spend way too much on an API as well, really quickly. So I hope they I hope they I hope they stay and help us out with that. All right. Thanks very

Janessa Lantz (42:28) Yeah, yeah. Thanks, John. Yeah. Bye.

John Steele (42:33) much for being here. Have a nice day.