Latest / The Chief Customer Officer Podcast / EP014 Killing the CSM Role: TAMs, Forward-Deployed Engineers, and more

Transcript

EP014 Killing the CSM Role: TAMs, Forward-Deployed Engineers, and more · The Chief Customer Officer Podcast

• Jeff Breunsbach: All right, welcome back to another episode of chief customer officer. This will come out on Thursday, April 23rd. Jay, how's it going? How's your weekend? Jay Nathan: Good weekend was great. I can't believe how chilly it is here in Charleston. I've got my sweatshirt on today because it's only going to be 72 degrees. No humidity. So we're spoiled. Jeff Breunsbach: We did like a mini staycation last week with ⁓ my brother-in-law and his kids. So we had four kids in the house. We rented a house that was near the beach down in Seabrook. ⁓ they were leaving their home and it was like 30 degrees and snowing, not even joking, in the middle of Michigan. ⁓ They came down here and it was like 80 degrees last week. They were just like in heaven. And my wife just sent them a text this morning because our son, our three year old is very temperature driven. Like if he's too hot or too cold, like it just throws off his whole, actually it throws off my whole day if he's too hot or too cold. So he, it's 55 degrees whatever. I'm getting him out for school. He's got pants and I gave him like a light jacket. He gets out in the car and he's like, I need my big puffer. And I'm like, dude. Jay Nathan: ⁓ perfect. That's funny. Jeff Breunsbach: We're in South Carolina, it's 55 degrees, it's not that, it's gonna be 75 today. He's like, I need my big puffer for school. So my wife sent a text to them, was like, I don't think Kayla would survive in cooler climates. Jay Nathan: Hahaha! ⁓ my gosh, that is a riot. Hey, you know, when there's no humidity, it's genuinely cold down here. That's the case I try to make at least. Jeff Breunsbach: I'll agree. Yeah, I'll agree. Well, before we jump in, one thing to mention, we've got, ⁓ I'm throwing together a little AI hackathon on Friday, this coming Friday. ⁓ I'll put the link in the show notes, but 12 p.m. Eastern, whole idea is hopefully to have five or six people actually come demo like real things that they've built ⁓ using AI, using, you know. some sort of new workflow or something that they feel like is ⁓ really like on the cutting edge of kind of the technology. So if you want to come listen for an hour, I'm probably not going to record it, at least this first one, just because I want them to hopefully be able to share something that ⁓ I want them to be able to share something that they could, you know, if for whatever reason, if they can't share, like if they can't share their screen and they have live data or something, like I'm trying to figure out some way that like we can do it. But ⁓ who knows? This might be one of many things that we do. and we'll certainly, I think, take some of these and write it in the newsletter. Jay Nathan: Well, you know. Yeah, absolutely. You know that everybody on that call is gonna have their little recorders with them, right? So it's gonna be recorded, whether you record it or not. So at least record it and get the transcript and then you can use it to write some of the articles. We can do that, you know? There we go, all right. Jeff Breunsbach: True. Alright, fair. All right, fair. But, so yeah, it'll be a Friday. ⁓ And hopefully, you know, if it's something that works, ⁓ if I can figure out a lightweight way to do that each month, we'll try and keep those things going. But I think it'd be fun to, I think we're all just looking for real examples of like things that people are doing. ⁓ this is kind of how we used to do some gang-grotain stuff. And it feels very natural to like, hey, let's just get some people in our room to share real insights. So, yeah. Jay Nathan: That's what people want right now. That's what people want right now. I we're seeing some interesting stuff. So we have a lot of calls with, ⁓ SAS companies and other, other, other non-SAS companies as well, but with folks who are trying to figure out how to leverage Claude to do more and some really interesting stuff starting to happen out there. And it's interesting because I had a, a product guy yesterday, tell me that I work with real closely at Pendo. Tell me like, In six to 12 months, no SMB will be buying any, any software. That's not like your core CRM or anything else. And I think we're probably already there because I think departmental users of these things are really starting to lean in and create solutions. actually created something this weekend. I needed a, just like a pipeline reporting mechanism for HubSpot. And I was like, am I really going to go in and configure HubSpot to do this? And the answer is no, I'm not. And what I did, well, by the way, HubSpot's pretty sweet in that you can do a lot of the configuration via cloud. So they've done a great job of sort of opening up the platform, making it quote unquote headless ⁓ so that everything can be accessed via APIs and command line and so on and so forth. But ⁓ I went in and created exactly what I needed for a dashboard. I know how I needed dashboard to run for. Jeff Breunsbach: Hmm Jay Nathan: pipeline management, right? I want to see how many opportunities are being created. I want to see how many deals we have in each stage, which ones are overdue, like a heat map of which deals are healthy and not healthy based on, you know, how long they've been in that stage or how long it's been since we've communicated with them last. But I had everything I need and I did it in literally a matter of minutes. I'll share this with you, Jeff. We probably won't share this with everybody because it's got Jeff Breunsbach: you Jay Nathan: Live data in it, ⁓ this is it, right? And by the way, I can go in and edit this thing and it saves directly back to HubSpot. So it's two-way integration and I don't own the HubSpot module that allows me to do this kind of reporting, by the way. So I've been able to do everything that I need to do and make it sort of interactive two-way ⁓ without. Jeff Breunsbach: Interesting. Jay Nathan: without actually buying anything additional from HubSpot. Don't tell them this because I don't want to pay any more for HubSpot than I'm paying right now because it's not cheap. So don't nobody tell my secret, but that's what I'm doing. It's working pretty well. Jeff Breunsbach: ⁓ That's cool. when you, like talk through a little bit more about how you like did this exactly. this through Cloud Co-Work or Cloud Code that you built this? Jay Nathan: Okay. Yeah. So it's, it's Claude code. So, ⁓ initially, initially I did it just through a chat and I had the MCP go pull out all of our deals out of, by the way, you could do this for new rules too, right? It doesn't have to be new business opportunities. had the MCP server go pull all the deals and create a dashboard. And then I sort of coached it on what I wanted. And then I said, okay, I need to move this into an actual application that I could just use every day now because it's Jeff Breunsbach: Yep. Jay Nathan: Exactly what I need. I don't want to have to go to quad and have quad connect to the MCP server, refresh everything. I don't want it to be static. other words. So, ⁓ so I did that. It built me an application that is running locally on my machine. Now the challenge with that is how do you, got to get it out to the rest of your team. Right. And this is where having an application layer that you can actually actively build on that secure and is. hardened for whatever certifications you have as a company ⁓ from a security posture perspective, but ready to go so that you can deploy applications like this constantly on top of your core applications. I think that's going to be sort of like the core discipline and capability we have to have as a company. And I think that's going to be the same, probably no matter whether you're an SMB or maybe mid-market or even enterprise level. organization, you're going to want to be able to build stuff on the fly because it helps you understand what you're going to, what you need. Right? Like this evolved over the course of a weekend. It started with a prompt and then it evolved. Right. Jeff Breunsbach: Yeah, well, the thing that I think the narrative that you keep seeing from people out there is like this narrative of like, you know, if businesses start building internal applications, then you basically are going to create like this internal tech debt where you like you're going to have to continually like upgrade these things and make sure they're running properly. And so ⁓ and I think that sure, that's largely true. But I think that also just means the person, let's just say like normally a rev ops person or somebody who is doing that type of dashboarding right at a company. That just means that part of their work now, I don't know if it exactly changes anything other than where those things are deployed. Like you mentioned, right? Like it actually maybe just changes and says, hey, instead of you having to go drag and drop fields and mess around in HubSpot directly trying to build a dashboard or whatever, this person now is just spending their time in cloud code building that and trying to coach it and get it to what it needs. Right? So I don't know if it necessarily changes that like it still seems like these kind of core systems are in place, but now it's just that you're doing the core thing and still storing data in that, in the, in a specific way that you need that has controls. And like you said, has the knowledge that they've learned of the ecosystem over years. They've got partners, they have a ton of things that are going for them, but they're not great at dashboarding, right? Like that's not their core component or like they're not great at something. And so like now you don't have to Jay Nathan: Yeah. Jeff Breunsbach: essentially trudge your way through a part of their platform that maybe isn't a core focus or hasn't been built on. Now you can move that out to a cloud coworking or cloud code and basically say, Hey, I can create my own world using that data and the repository and the things and the controls and all the things that I need. ⁓ but now it just shifts to maybe where that operations like that, ⁓ kind of rev ops team and just kind of maybe shifts where some of their work lives. Jay Nathan: Absolutely. And actually as you were talking about that, it made me wonder like what is going to happen to all these BI tools that are out there? Because I mean, look, I mean, none of this stuff is going away, right? Looker, Tableau, they're not going away, but do they begin to lose some of their effectiveness and luster when I can just build custom code, create custom code to write anything, right? And it's just a prompt away. ⁓ I do think there's a, Jeff Breunsbach: Yeah, that's a question. Jay Nathan: There's so in our world, we think about this idea of like a center of excellence inside of a company, which is sort of a I think I talked about this a little bit last week on the podcast, but this this idea of having a central place where you can actually build a capability and expertise around a thing, right? Like deploying AI applications, deploying, you know, quad for whatever your needs are. And I think that's going to be important from a security and technology. Jeff Breunsbach: Yeah. Jay Nathan: You know, perspective, make sure the right platforms are in place, make sure that we're learning quickly. We're sharing and disseminating that knowledge across the organization. As we do learn, ⁓ one of the crazy stats that I read this week, which is actually not that surprising. ⁓ I can't remember if I mentioned this last week or not is that like something north of 45 % of all code being written by cloud code right now has significant security vulnerabilities in it. 45%. And you see that like there's some. some news about some platforms that were hacked last week that are like being heavily dependent on by Claude. ⁓ But this is the challenge, right? Is deploying technology, you're still deploying technology. And if you're putting sensitive data in this technology, then you have to be really, really careful to ensure that you're doing it in a way that you're not gonna lose that information or have the. Jeff Breunsbach: Yeah. Jay Nathan: have that information fell into the hands of people that you don't want it to. that's engineering, right? At some level, that's still engineering, whether that gets built into a platform of some sort, right? Where those concerns are taken care of for the developer, quote unquote, the Claude in user, Claude code in user or not is a question, but that's why you need some kind of sanction around this at the organizational level to be able to put those guardrails in place so that Jeff Breunsbach: Yeah. Jay Nathan: you don't get yourself in trouble with these tools. It was interesting stat. Jeff Breunsbach: Yeah. ⁓ one of the other things, yeah, that is one of the other things that, ⁓ I think is, is pretty interesting too is it feels like, and maybe this is just more bias of people who I talk to, but it feels like, and this is not news or new maybe, but, ⁓ it feels like Claude and least in the business environment, let's just say like the world that we live in, it feels like Claude has like, again, kind of taken a, another jump or another leap up where like, it seems like that's the tool of choice that I hear most often versus open AI and chat GPT and even codecs like you hear less than maybe Claude code and co-work like and I'm curious your opinion because I think there's two reasons for this. One is I actually think it's a marketing move. I think they did a great job of naming these things in a way that like were super simple for people to understand. Like there's chat, there's co-work, there's code and basically they there's a place for everybody in the business. know chat is your kind of least technical person that just needs an answer directly quickly. The co-work is like a more of an interface for somebody who maybe doesn't know code, but knows the business and can like float between both. And then code is for somebody who's, you know, more driven in engineering or coding. And, but then like you think about like open AI, it's like open AI and chat GPT and all. then it's like, what's chat GPT versus codecs. I don't know. It just feels like there's like a marketing move that they made where they made all these things very simply and, laid them out in a way that was like picking up steam in the business world that like made it super simple for that to happen. And then two, don't know, I haven't read about this, but it just feels like their model development has been better at, I guess, core functions inside of the business. feels like they've somehow gotten that better. So I don't know if you've got opinions on either of those two things, but that's just like my general layman's persons, like outside view. Jay Nathan: The interesting thing about that is that lot of people would say that cloud code was misnamed to begin with because it was sort of presented as this technical tool when in reality it is, right? the power that it provides is you don't need to be technical to actually get a ton of value out of cloud code. So that's why they actually created Cowork. ⁓ Jeff Breunsbach: Yeah. So thought. Jay Nathan: what's gonna happen with these three pieces, chat, coworking, code, they're gonna converge at some point and it's just gonna be clawed, right? I could be wrong on that. But I mean, that's what it's sort of moving toward. Chat is the simplest way to get started. interest ⁓ in people who listen to this probably already have this kind of laid out in their head, but the chat GPT is optimized around B2C. Jeff Breunsbach: Yeah. Jay Nathan: Open AI is sort of, they're going for scale and volume. think my Jeff Breunsbach: Yeah, largest tan ever, every person on earth. Jay Nathan: Yeah, right. Everybody. And they are either close to or already had a billion users. Although I think they, you know, their growth has slowed a little bit. And I think a lot of that's due to Claude. Claude has really now we do live in a tech forward space where Claude is more prevalent because sometimes like I'll ask people, you know, about, you know, do they use Claude? And they're like, what is Claude like Claude? What Claude code? What is that? Right. And it's like, we talk about it every day all day long. but I don't know that, ⁓ I think Claude's early investment in code, Anthropics early investment in Claude code really put them ahead in the B2B world. And, it just got really good, really, really fast. Open AI, you know, from what I hear, you know, people say on different podcasts, sort of spreading themselves thin across a lot of different Jeff Breunsbach: Yeah. Yeah. Jay Nathan: opportunities in the market. They're trying to figure out everybody's trying to figure out what's going to work because like, don't forget all of whether it's Claude or whether it's anthropic, whether it's open AI, whether it's grok, these companies are spending ungodly amounts of money on capex, which is basically build out costs to be able to, build these services for us. And you can already see Jeff Breunsbach: Yeah. Jay Nathan: So I think OpenAI just raised the biggest investment round ever in the world in history of anything. It's like $120, $130 billion. So they're running off of investment, whereas you see Claude beginning to run off of the flywheel of customer success. Not in the sense that we say, but customers are getting value out of it. They're buying more of it. Jeff Breunsbach: Yeah. Jay Nathan: They're switching to it. ⁓ But so it's going to be interesting to see how this plays out. Is open AI going to end up being the Netscape of AI? And if you don't know what that reference is, Netscape was the number one browser at one point. Then they decided to rewrite it and, and completely blew it up. But like at some point these services have to be paid for. Jeff Breunsbach: Yeah. Remember Netscape. Jay Nathan: Right. And there's going to be so much, so much investment in these data centers to service all these needs. And it's not going down. I promise you that, like just go listen to Jensen Huang talk for an hour. He's on all kinds of podcasts right now. All in, he was on a couple of weeks ago. He's on Dworkish right now, which I started listening to this week. I mean, just listen to this guy talk and you will understand just how big AI actually is. And Jeff Breunsbach: Yeah. Jay Nathan: Claude, Chad GPT, those are just the tip of the iceberg as to what this is, right? So anyway, somebody's gonna have to pay for all this and they're looking for the business model still, OpenAI is and Claude. Jeff Breunsbach: Yeah. Yeah. Well, it's interesting. Like you said, we're like in a technology software world for the most part, and that's generally who we sell to. And you just think about like, there's so many industries or so many, like one of the things I read the other day that I thought was interesting was like another thing that like AI unlocks, especially like the cloud codes of the world is that like, there's a lot of companies that have had technical needs. that they've just basically said, I'll have either written off because they said it's going to cost us too much to build it or they've just kind of pushed it into like their own solution or some solution that they could buy off the shelf that wasn't really fitting their needs. And so just think about all these non-technical industries that like have actually had. technical needs that no one's wanted to go into because it was too costly or you needed too much industry knowledge or niche expertise or something. And now it's honestly like this has opened up the aperture. And so it's almost like ⁓ I forget the paradox, but is it Jevon? No, this is Jevons paradox. We're essentially like you think AI is going to start killing jobs. It's actually like the all of sudden, these non-technical industries, all these non-tech industries could be opening up product manager roles. They could be opening up engineering roles. Like it actually could like go the opposite direction because they've never thought it was a good enough investment. But now it's like, oh, if I have a cloud code that makes me do it in half the cost, I don't know, a third of the cost, like, you know, that starts to open up the door with like, hey, maybe that's a good investment now. Like maybe they can actually do it. So it's an interesting time. Yeah. Jay Nathan: Yeah, I'm just going to do more. I'm just going to do more. the, heard an interesting, you hear all these analogies, right? And, ⁓ there was a guy who gave an analogy of telephone operators. Like remember it wasn't that long ago where if you wanted to make a long distance call from South Carolina to California, you had to call the operator. It was a little before my time to be fair. I never had to do this. You had to pick up the phone and dial the operator and the operator, which is a human would take a plug out of a wall, out of the local network and they would plug it into the national network, right? And connect you to the national network so that you could call somebody across the country. That was a thing. There were 800,000 of these people in America doing that job in every single town, right? It was a very good job. But then we built switches, right? And we started to deploy those switches and you would think, well, that job did go away, right? But Jeff Breunsbach: ⁓ my gosh. That's crazy. Jay Nathan: how many additional jobs have been created because we now have national telecommunications at our fingertips, right? So yes, your job will be different in a year. It will be certainly different in five years. ⁓ But this is the, this is, there will be other jobs and they will be just as important. I don't think the, there's this whole term called doomerism. You've heard this like, Jeff Breunsbach: Yeah. Jay Nathan: These are the people who are out there saying, ⁓ AI is going to take all the jobs. It's like sort of a marketing ploy in and of itself. Right. I mean, by the way, Claude Anthropic is probably one of the world's best marketing companies right now. Just look at what they're doing with the mythos thing. We should talk about that. Have you seen mythos? ⁓ But anyway, the point is like we have, we have way more jobs. We don't have anybody doing that job anymore. We also don't have people operating elevators either. Jeff Breunsbach: No. Yeah. Yeah. Jay Nathan: Right? So, but there are other jobs for these people created by growing economy. Jeff Breunsbach: Yeah, it's true. ⁓ Well, it's a good transition because Chad Hornfeld, ⁓ who's a friend of the program, wrote a LinkedIn post just about what they're doing at Avoco, which is a company where he leads customer success. And ⁓ he kind of mentioned that they killed the CSM role. learning the product and growing the account isn't really the job description anymore. It's kind of the starting point. And so they replaced it with something they call the technical account manager, person's hands-on, technically fluent, obsessed with customer outcomes. And Tams partner with four deployed engineers to unlock integrations and customizations that would have taken months before. So I'm curious, what's your take on what Chad's doing and anything noteworthy that stands out for you? Jay Nathan: ⁓ yeah, I mean, this is, it's a really interesting development. And by the way, I think a VOCA is far from the only company doing this. ⁓ a company that we work with really closely has recently made a big change to where they sort of collapsed all their pre-sales and post-sales roles into essentially the equivalent of a forward deployed engineer kind of role. And I think the expectation is with these products, like Salesforce last week announced that they're going to that they're 100 % headless, which means you can access everything via API, right? And just like we talked about a minute ago with HubSpot. And I think in that world, like the question is, how do you deploy and integrate that solution with the rest of the organization? And that's an engineering task. Now it's not necessarily one that the customer has the skills themselves for today yet they're working on it, but maybe not all have that. And then Jeff Breunsbach: Yeah. Jay Nathan: Number two is knowing how like what are the patterns to go integrate that solution into the organization. So in that world, that makes a lot of sense to me, right? It's a you guys junction is an API first company, right? So how do you think about it? Jeff Breunsbach: Yeah. Yeah. I'm actually hiring for CSMs right now. Yeah. I'm hiring for CSMs right now. And ⁓ I would say a core component of what I've tried to assess for is like technical aptitude. ⁓ And it's less like I'm not looking for somebody that's telling me they like, you know, built their own applications and like are running, you know, some side business or something, but like, you know, show me and like, show me not even like, tell me like, show me something that you've built some, some, something that you've been using AI with, like, but like, how are you kind of getting on the forefront of this? That's like one aspect I'm, certainly looking for. And then the second is like, um, the aptitude of like, I mean, all of our documentation is public. have API docs and everything. And so like, the research, you coming in and telling me that Junction's ⁓ the healthcare data infrastructure company isn't the answer I'm looking for now. It's like, what have you gone to actually read about our API docs and could you teach me something from the API docs ⁓ or tell me how something works or figure out how something might be diagnosed. so those are the two things I've tried to assess for in our CSM role. ⁓ think it's much leaning to me much more. ⁓ I'd say the two things that I'm trying to assess for is technical abilities and commercial abilities. Like you need to be able to talk money and you need to be able to help the customer understand what they're building. Jay Nathan: So you still want them to do both learn the product and grow the account is what Chad sort of saying here. Jeff Breunsbach: I do, yeah. But we're also hiring for a solutions engineer. And this is going to be question I was going to ask you, which is, like, the solutions engineer, how do you see that different from a forward deployed engineer? Like, why has there been a shift in maybe, like, nomenclature? Jay Nathan: Well, mean, the solutions engineer, the sales engineer, as it's more commonly known is has traditionally been someone who can help tie the software together with the with the business needs in a very demonstrable way. And literally they demo the software and that could be anything from, the solution engineer, the sales engineer shows up on a sales call and does a standard demo all the way to like I've been part of, know, we had a We worked on a $20 million deal ⁓ years ago that I got to be a part of. we had literally two weeks worth of scripted demos on site with a client, with every single department of that organization walking through their needs and our specific responses to that. And the sales engineer led the entire charge in terms of preparing that. ⁓ Jeff Breunsbach: Yeah. Jay Nathan: I would say, you know, is that still going to be a need in enterprise? Probably to have someone who can still do that level of engagement around solutioning ahead of the sale. Versus the demo jockey, right? We have to have sales. The best salespeople right now can demo the software themselves. They're, they're a little bit more technical. They understand what's going on. This is not a, you know, slap you on the back, you know, relationship driven world in software sales anymore. So I think that kind of thing is sort of going away and cause there's no room for it. Like all the margin is gone in these businesses. People are looking for the right solution. Jeff Breunsbach: Yep. Yeah. Yeah. Jay Nathan: not a lot of waste, they're not adopting, well, outside of AI solutions, they're not adopting technology just to try things out, right? I mean, it's really gotta be ROI driven these days. So the old relationship sale doesn't necessarily work the way it used to in a lot of industries. Jeff Breunsbach: Yeah. ⁓ Yeah, so I actually probably need to go talk to our head of people about renaming the role because like that's we're essentially taking a solutions engineer and putting them in the they're going to report into engineering right now. ⁓ But they are essentially they are essentially a resource for customer success to work with ⁓ and they're solely dedicated towards customer outcomes. And so ⁓ to me, though, it's like the definition of like for deployed engineer is like what we've gone for, you know. Jay Nathan: Interesting. Jeff Breunsbach: We've just named it solutions engineer for post sales, but I think we should just like probably look at naming that. But to your point, like what we've identified is that ⁓ we need somebody who can help us basically circumvent future problems, ⁓ like help the customer think about the setup from like a very technical way. And so like, want some of the CSMs to be able to handle some of those conversations. And I can give you an example here in a minute, but, ⁓ but we also need, you know, an engineer to actually come on and either be working with the customer on those solutions directly or like, you know, working on something on our backend that's going to help them or like, you know, helping to coach and articulate like what some of the best practices are that they need to be following. So, ⁓ yeah, so that's, but like one of the other examples that I have right now that, I'm pretty excited about is just like trying to reorient our onboarding. and really like the first spot is actually laying up a patient journey. Because we think about patients at end of the day, and that's how our customers are working on their patient outcomes. ⁓ And so to me, is truly the customer ⁓ journey is a patient journey. And instead of rather thinking about how do we do onboarding, what I actually want to happen is here's the patient journey. Here's common moments that they're engaging with their patients. And during those common moments, here's when you should be pinging our API or here's when you should be using some of these endpoints. Here's how you should basically be thinking about overlaying junction inside of your environment. And like, that's a place that we should go start with and coach them on. Cause I think 80 % of our customers would probably just cool. Take that, use it. And then it's deployed the best practice. what we've actually, what's happened a little bit now is that, you know, we've, we've kind of said, Hey, go use it how you want. And it's just now we're noticing some downstream problems. And we actually found this out because our support leader. has a meeting every other Friday right now where we're bubbling up ⁓ support issues and we're reading through certain things and he's giving us trends and he's pulling all that stuff out of the support tool that we use. And so we identified some issues and said, hey, why are all these things happening? We went back and looked and we said, ⁓ it's because they've configured a few things. I won't say wrong because they can't do them, but they've configured a few things that's actually creating support tickets for us because of how they deployed it. But it would not be a best practice of ours. It's just that. We've allowed them to do that. so anyways, that's like one example right now of like how I want the CSMs to basically reorient onboarding. And that's where some of the technical aptitude I want them to have, right? It's not about like, again, I think it's if we're calling endpoints, if we're telling you where to put the product in, like that to me is I'll say like slightly technical, but that's not like talking at an engineering level or something, but you need to be able to understand our product well enough to go help them say, yes, this is our best practice and this is why we did it. And this is like how you can call the endpoint and use like some of the attributes of the endpoint. Jay Nathan: So it's industry driven best practice engagement, right? And so I was just reading back over Chad's article. They actually, it says they replaced the role, CSM role with the TAM. Okay, so that's the core role that's doing the facilitation. I would think like, it doesn't matter what you call it, right? The question is, what does it do? ⁓ So what you just described is the same kind of work in my mind. Jeff Breunsbach: Yeah. Yeah. Yeah. Jay Nathan: That person, at least in Chad's example, is ⁓ working with a forward deployed engineer who is doing the actual integration, customization. They're helping the customer at a much more detailed level. And then he doesn't really mention where the sales engineer falls into this. OK, yeah, yeah, yeah. Jeff Breunsbach: Yeah. He says the commercial motion lives elsewhere. Yeah. ⁓ he's he says kind of the commercial motion lives elsewhere. and so he says, you know, like, go ahead. Jay Nathan: Yeah. But I mean, to me, go ahead. No, go ahead. Jeff Breunsbach: I was gonna say he mentions that like going technically he says like asking one person to go both deep technically and carry a number is often how you get mediocrity at both. Yeah. Jay Nathan: Totally agree. Totally agree with that. but I, I think, don't know, this is where there's a whole nother part of this conversation, which is, you know, how to think about your, your sales team and how they're structured and whether you do that, you know, do you, do you keep sales focused on net new logos or do you keep them focused on current accounts? Cause what you, what I see a lot there is over indexing on existing accounts. where when you have a responsibility for both, you go to where the easiest work is and that's expansion. It's not hunting new logos, buying a stretch. ⁓ So that's a whole different conversation we don't have to get into, but it's interesting that he's separated that out, but it's all in support of the commercial relationship anyway. Right, I mean, we're doing this, this is a business, right? So. Jeff Breunsbach: Yeah. Yeah. Yeah. Yeah, and we're I mean, we've got, you know, we're essentially usage based ⁓ on top of like some of our like some of how the product works, right? So like to me, there's actually a naturally commercial component for CSMs. That's like if you're helping the customer solve technical problems, you're actually helping us actually access more ⁓ usage effectively. And so like that's that's where I think like their ⁓ component comes in. ⁓ Jay Nathan: Yeah, right. Jeff Breunsbach: So I mean, think the other thing that this comes to mind to you though is, or something that's coming up for me right now is just trying to figure out like, how do you, using a Claude code or Claude code work, can I simplify ⁓ some of the components of our CSMs from like a commercial standpoint? So if we have a renewal coming up, if we have a upsell opportunity, can actually take all those details via Claude code or Claude code work, take them out of plan hat. use our template for our order forms and can basically like hopefully remove any sort of like manual steps that they need to do where like they can basically push a button that says generate, you know, generate order form and it spins up a Google link that does it. And then it's like, cool, now I have an editable order form, but it's already got all the details. and it's in our template, you look that over, you let me look that over, we send it to the customer. But those are the moments where I think, ⁓ can I take those moments out so that they can still stay committed to the commercials and the customer ⁓ as much as possible? Jay Nathan: I tell my team all the time, the most important thing that you can be doing, the most important thing that we can all be doing is what you and I are doing right now. Whether it's in person or virtually is spending time talking to the customer. Like I said on the podcast last week, if folks listen to it, most important data source that my company has is the thousands of recorded phone calls that we have with customers where we're sharing best practices. talking about their problems. Like that's where the magic happens in most businesses. Now that's where the context is. Jeff Breunsbach: Yeah. Yeah. Jay Nathan: And so what you're talking about is just good old fashioned automation at this point, right? Of course it's AI driven, but I mean, it's building agents to do those things that we never could have really automated before and felt good about, and now we can. So that's just operations at this point. Jeff Breunsbach: Yeah, yeah, yeah. Yeah. Alright, I have to run but one thing I did want to tell you that I thought was cool As a business we're doing something ⁓ once a month right now, it go more but we're taking a Friday So it's actually this Friday and we are ⁓ canceling all meetings No meetings on the Friday ⁓ You do not like you essentially are working with your team to automate something within your part of the business or automate or use AI or resolution. But it's effectively like you're coming to the table with your group of stakeholders and so like my team and I are going to basically meet a bunch on Friday and we're going to effectively come up with a list of problems of like what are some of the challenges that we have today? Where are some gaps? What are some things? Do we have solutions that we could be building, you know, through the use of AI or automation with those and then like actually we're going to be building those on Friday and then you basically demo what you've built at the end of the day. So I'm pretty excited. I think it'll be pretty fun. Jay Nathan: Sweet. Cool. It's like a hackathon. That's great. So you guys are doing that, you're doing that this week. Jeff Breunsbach: Yeah, like I mean it's just involving each of the departments. Yes, on Friday, first one of the month, or first time we're doing it. So we'll see how it goes. But I thought it was pretty cool idea from our CEO. He was just like, let's just take it one day a month to start doing this. And if it's working, we could basically crank up that pressure if it's yielding results that we want. Jay Nathan: All right, cool. ⁓ totally. That's cool, man. I, we do, we do every two weeks. We don't go that far. We do our AI show and tell where people basically bring things that they've been working on and something really cool happened yesterday. We had one on Friday, an AI show and tell, and one of our guys built something that was really cool. Everybody liked sort of this template that we could easily spin up for a type of engagement that we do with our clients. And then, ⁓ but it was on a very specific platform. We, actually built the template for SharePoint. in this case, it built the whole SharePoint site. what ⁓ yesterday somebody posted on Slack, I was inspired by Brian's project and I created the same thing for Google sites. So now we have that template for Google. So it naturally inspires people to take the next step when they see what everybody else is doing with it. There's just so much creativity that's being unleashed right now, which is so fun and exciting to watch and to be a part of. So I'm looking forward to hearing about your hackathon. That's cool. Jeff Breunsbach: Cool Yeah. Yeah, for sure. Yeah, we can talk about it next week. right. Good to see you. Good episode. All right. All right. We'll talk to you soon. Jay Nathan: All right, we'll wrap it up. Yeah, you too. Yep. See you next week.