Why Your Ideal Customer Profile Is Broken and How AI Can Fix It ft. (Co-Founder of Stage 2 Capital)

What is the output metric that we have to evaluate these segments are to decide if it’s in our ideal customer profile?

Josh Schachter [Host]:
Oh, man, I feel like I’m back in business school here.

Mark Roberge:
Dude. It’s LTV. Our job is to go find ltv. Yeah, in very correlated to retention. Right? That’s where we have product market fit. Okay, so like the, the people we want to go find are not the people that are easiest to sell. They’re the people that are going to succeed with our product.

Josh Schachter [Host]:
You’re listening to Unchurned, brought to you by the Gainsight Podcast network. Mark Roberge didn’t just help build HubSpot from zero to $100 million in revenue. He wrote the playbook on how to do it. His book, the Sales Acceleration Formula, turned sales from an art into a science. But today, Mark’s focused on something entirely different. Reimagining what happens when AI agents replace the very sales teams he spent a career perfecting. As co founder of Stage 2 Capital and a senior lecturer at Harvard Business School, Mark now invests in the future of Go to Market AI backed by a network of thousands sales, marketing and customer success leaders from the world’s best tech companies. And he’s not just thinking about productivity gains.

Josh Schachter [Host]:
He’s wrestling with something much bigger. What happens to capitalism itself when agents do the work humans used to do. Today on Unturned, Mark Roberge on the four phases of AI transformation in Go to Market and why we might need to rewrite the entire playbook on work identity and how we live. I’m Josh Schachter. This is Unchurned. Welcome to this episode of Unturned, everybody. I’m your host, Josh Schachter and I’m delighted to be here today with Mark Roberge. Mark is the co founder of Stage 2 Capital, an investment fund that’s doing some really exciting stuff in the AI and go to market space amongst other companies that are killing it in his portfolio.

Josh Schachter [Host]:
He’s a senior lecturer at Harvard Business School and is one of the or was one of the OGs at HubSpot. It took them from 0 to 100 million and ultimately left as CRO. Mark, thank you so much for being on the program.

Mark Roberge:
Hey, thanks, Josh.

Josh Schachter [Host]:
Great to be here. Where do we start? We want to talk. All things Go to market. All things. In this world of AI transformation, I want to learn about the, all the kind of bubbling startups that you’re seeing in your portfolio and that are applying to become part of your portfolio where you think there’s opportunity, where you think it’s just noise what the future is. You know, I want to hear about this book that you’ve, that you’ve contributed to that’s coming out. Yeah, let’s pick one.

Mark Roberge:
Oh my gosh. Like, I mean, I guess let’s start with the AI stuff and we can back into the book stuff. So let me give you like the quick arc if I can Josh, just to give folks the context of where my perspective comes from. All right, so I think we all in the tech world will remember November 2022 when the was that the year? I think that was it the chat. I know it was November, I’m pretty sure it was 2022. 2022, the big chat GPT. That version, you know, launched and went viral and it just kind of like showed to the world what how much of a leap AI’s going to be. And so you know, as I, you know we’re running stage two, I think we’re in the middle of fund two at that point and we had our LP meeting on the east coast and at mit there’s a couple hundred people there and as you know, you probably, you know, you know Josh.

Mark Roberge:
But like for the listeners our uniquenesses were backed by the heads of sales, marketing, customer success, rev ops, all the go to market function and the we what we think are the best tech companies in the world. So there’s a thousand of them that are our investors and that’s what we bring as a unique contribution to the ecosystem.

Josh Schachter [Host]:
I mean what a freaking advantage by the way as a portfolio company like to have all those people like as ammunition for your to build your business.

Mark Roberge:
It’s a blessing every day to get up and play that accordion. It really is unbelievable. Like such a blessing. And nothing that I necessarily set out to is just really the vision of my co founder Jay Po and me listening to my first principles. But anyway I wanted let’s stay focused on AI but so you got the contacts. I’m standing up in front of a couple hana random at MIT in the spring 23 and I’m like hey, how many of you need to do more with less? And at that time it was remember like everything had crashed. Interest rates are going up, a lot of pressure and tech companies, no more hiring freezes, all that stuff, right? Efficiency, all their hands go up and I’m like great. How many of you think AI is going to be a key piece for that.

Mark Roberge:
And then like all their hands go up. I’m like, all right, I’m going to send out a form after this meeting if you want it. We’re going to like start seeing a bunch of these go to market AI startups and we can like sift through them and kind of surface the best ones to you if you’re interested in seeing them. So like 110 of these folks fill out the form and, and it’s like the CRO of like Snowflake, Zoom, like big companies, right, that any native AI company would just love to get a meeting with. So I promote that on LinkedIn. Like, hey, I got 110, you know, go to market leaders that are looking for the, the greatest AI breakthroughs. And within a month we get 250 submissions of like AI startups in the go to market arena. So we see like a pretty quick glimpse of what’s out there.

Mark Roberge:
And at that moment, I mean, I don’t wanna offend the people then it’s just this very typical innovation cycles. But it was like not exciting.

Josh Schachter [Host]:
It was rappers, I remember that.

Mark Roberge:
It was a lot of rappers, co pilots. Oh yeah, like, and we like, fortunately we, I think we were able to see through that a bit. We surfaced a couple decent ones and I think it generated about 80 meetings and I think it was like one pilot. So I’m like, I go back in the fall meeting and now we’re out in like Silicon Valley. I think we did the fall meeting at like, I don’t know, Google’s office or Oracle’s office or something. And I’m like, what gives? You guys all said, yo, you folks all said you want to see? And they’re like, oh yeah, like we, we tested it, it was cool, but then like legal and it shut it down. Like we didn’t have an AI policy. So that was like the meme of that, like second half of the year was like the big companies didn’t have their stuff together from like an IP perspective.

Mark Roberge:
And so anyway, we fast forward to 2024 and that was kind of the year that like we got through all. Like we figured out how to sell through it and security and we got some of this stuff into like proof of concepts, which is cool. But then we started to see a lot of resistance from frontline workers who thought they were training their replacement. So we had to kind of like get through that. And you know, we’re, we’re seeing like step functions, like 10% improvements yeah. Even though like it’s clear that this stuff has the potential for 5 or 10x improvements and anyway we’re just not there yet.

Josh Schachter [Host]:
Do you think it’s data? Like what do you think the biggest speed bump is right now? So it was I guess like governance and just the overall. Just putting together program.

Mark Roberge:
Yeah, through that for the most part.

Josh Schachter [Host]:
Yeah. And then there’s. Then there’s fud. Right. Then there’s fear of loss of jobs and there’s. I think that still is in the back.

Mark Roberge:
Hurts a little bit but I think we’re working through it. Like I think I’ve seen some really good implementations of like scientific method control group experiments, you know, that help get through that and there’s a whole other side. I don’t mean to come up as like this capitalist, like I think a lot about too the social impacts here and I don’t even know if we have time to talk about that. But I just. There’s a whole bunch of stuff that I’ve gone to my like spiritual philosophy like this stuff. So I don’t want to just poo poo that outside. Right. That’s real stuff that we’re not talking.

Josh Schachter [Host]:
About like this Andrew Yang type of stuff that we’re.

Mark Roberge:
Yeah, yeah. I mean like, yeah, it’s. We’re not talking about enough that enough either. But like anyway we’re like through. Yeah, we’re getting through like the adoption stuff and yeah, I think there, yes some of it is the data but I honestly think it’s still the products. I just don’t think they’re there yet. Like I think a lot of the startup visions are still iterative, they’re not inventive and the underlying tech is there and it’s just like I think really. Yes, we can blame the customer because they don’t have their stuff together, their data, their workflow, their change management.

Mark Roberge:
But that’s our job as the vendor and in sales. And so you know like that’s where we are now. I feel like we are seeing the 2-3x and I think on our way to 10x and some people have done it on the product and engineering side. I think there’s still some issues on like moving from you know, like wireframes into like true production, you know, but we’re getting there. We haven’t seen it. We haven’t seen 2ers and 3ers in sales. Right. I just haven’t seen a lot.

Mark Roberge:
We’re looking for them and so now what we’re doing is like we have an OKR to try to write an article every other week around anyone. It doesn’t need to be in our portfolio. It doesn’t need to be using one of our vendors. It doesn’t need like anyone that’s seeing a massive step function in go to market performance. I think we’ve put out like, I don’t know, 15 articles on our blog with like down to like the prompt, you know, in. In the GPT on how to do it. And I think 2026 is going to be the year to do it. So that’s like.

Mark Roberge:
I guess the only other thing I’ll add is I take 50 students every January from HBS out to Silicon Valley and me with the CEOs of like, you know, Waymo. Waymo and you know, like Nvidia’s COO and the head of sales at OpenAI and the chief Business Officer of Perplexity.

Josh Schachter [Host]:
And just kind of these close.

Mark Roberge:
Yeah, come out, dude. So like I just put you on.

Josh Schachter [Host]:
The spot right there.

Mark Roberge:
I’m not trying to like, I just want people to know where my perspective’s coming from. And I spend most of my time just so folks know, looking at the B2B application layer, not the infrastructure layer and not really the consumer side. And I can definitely talk about my current vision, but I’ll stop there. Josh. To give like. No, no, no.

Josh Schachter [Host]:
That maps to our audience too. I mean, we’re talking to folks that are looking at the B2B application layer for the most part. That’s. That’s in their workflow. Right. So, so what is it that’s so iterative? That’s that just not the 2 to 3x and I get what you’re saying, like on the, on the dev side and the design side that we are there, we’ve got the lovable.

Mark Roberge:
I’ll just give you an example. Like right in this, like the heart of this audience is like, there’s so much work and like this look at customer support tickets, like, help the support rep put the ticket out. It’s like that’s the story right now. Yeah, that’s boring to me. Like it’s prevent the ticket.

Josh Schachter [Host]:
Yeah.

Mark Roberge:
You know what I mean? It’s like the AI has the capability of like when you look at the first principle and a user is using a product, like, let’s. What leads up to the ticket? Like, you start fumbling around, you’re clicking, you’re like pausing, you look for the chat section or the help section. Like, that behavior is highly predictive of what tickets coming and what solutions come in. So let’s just get rid of the ticket. And that’s like a classic example of like workflow disruption and innovation. And we can go on and on and on around renewals and expansions and sales and setting meetings and like all the stuff that we’re just taking the current workflow and streamlining. Like, we’re just kind of like iterating it. We’re not reinventing, we’re not breaking the mold.

Josh Schachter [Host]:
You know what, Honestly man, I don’t usually use this podcast to plug Gainsight so much, but like, you are validating. I’m feeling good about what you’re saying with respect to what we’re doing on the agentic side.

Mark Roberge:
Hey, you’re paying for it, dude. Plug away.

Josh Schachter [Host]:
Plug away.

Mark Roberge:
Right, right.

Josh Schachter [Host]:
We don’t have the midroll ad yet, so I guess I could. So, you know, but on the, I would say like at the surface, it’s like, you know, it feels like it’s a little slow getting out the gates with our agentic. I mean, it’s, it’s, it’s, it’s the focus of the company right now. But, but you know, we’ve avoided kind of the, I don’t know, wrapper is the right word. But like this agentic wrapper, you know, these like discrete workflows that are just iterative like you say, right. That are just like the 1.1x because we’re focused on trying to get the 3, the 5, the 10x. I mean, right now, our kind of marquee agent is a renewal agent that can, you know, automatically autonomously serve the long tail segment. And then where we’re, where we’re headed, our attention to now is an adoption agent, which honestly, we don’t know exactly what it means yet because it means so many different things to so many different people.

Josh Schachter [Host]:
It’s very complex, but it’s speaking exactly what you’re talking about. It’s okay, let’s preempt all of these issues and let’s figure out how to, to automate that adoption experience upfront. You know, that’s the big, you know, the big rock to move.

Mark Roberge:
Well, that is awesome, man. And like, if you want, I would love to write up your, your best customers with that stuff. Like, that’s just like what I’m really focused on in the next nine months or so through stage two, through my podcast is just like helping the world see these step functions. So let’s do that. If you’re up for it.

Josh Schachter [Host]:
Let’s do it.

Mark Roberge:
I want to. What I want to do, Josh, is like just kind of based on that foundation. I want to share my like hypothesis of where this goes from a high level and you can overlay it against what you’re seeing, you know, and, and just like what you guys are building.

Josh Schachter [Host]:
Yeah.

Mark Roberge:
And so like, and get. And again, just in the hindsight of like this is very probably wrong. We’re just so early. It’s hard. Like no one was predicting Snowflake and Twilio in 1999. Like we just don’t know what, where it’s going. You know what I mean? Like how this technology evolves. But here’s my take on like the first piece I’ll say is I think it’s, it’s difficult to predict where it’s going in the timeline on how fast we are abstracted from today’s point of work.

Mark Roberge:
And so I’m going to propose four phases of go to market evolution that each phase represents a further abstraction from today’s human point of work. And this is kind of where the buyer and the seller, the, either the buyer or the seller or the post sale person and the customer. Right. Like it’s kind of the same thing. It’s a very fluid thing. So I think in the first phase it’s really about maximizing customer facing time and prospect facing time. And on the pre sale side we call that selling time. I don’t know what the word is on the post sale.

Mark Roberge:
And just so folks have perspective here, like I ran the services and customer support side of HubSpot for a couple years. I don’t, I’m not like totally clueless about that side and obviously we work a lot there too. But um, so I have that context. But I do think like we have the potential. And do you have a guess, Josh? Like what is the average quote unquote selling time in the industry? Meaning if you look at a Rep’s week at 40 hour week, what percent of the time do they spend with a customer as opposed to doing like admin and meetings and stuff?

Josh Schachter [Host]:
Well, I would want it to, I would want it to be 75% of the time, but I think you’re probably going to tell me it’s like 30% of the time.

Mark Roberge:
Dude, you’re spot on. It’s 25. Yeah, okay. And like that’s, and like there’s a lot of people who are 10. Right. So just right there, if you picture today’s technology, we can go from 25 to 75 in the next year. It’s not that hard.

Josh Schachter [Host]:
Can you Imagine the dent that that would make in, in sales forecasts and what you could buy.

Mark Roberge:
I mean it’s just mathematically speaking, it’s 3x productivity.

Josh Schachter [Host]:
Right, right, right.

Mark Roberge:
If you literally go from your rep is doing 25% selling time with a customer to 75% and that rep did a million dollars last year, they will do $3 million next year. If you have 100 reps that did $1 million last year and you go from 25% to 75%, they will each do $3 million. That’s the step function we’re talking about and that’s very real. And we can, we can unpack that level. I get a decent vision on it and it’s pretty customer success oriented. So I want to run it by you. I think phase two, you know what.

Josh Schachter [Host]:
Now, now I want to run back to gainsay and figure out what that, what that number looks like for us. Maybe we will. Yeah.

Mark Roberge:
Yeah, I bet. I mean you guys are tight. I bet you’ maybe 40. The scene is like in the 40s.

Josh Schachter [Host]:
But what do you think? So I said, you know, I’d like to be at 75% but I’m just, I’ve never been a CRO. I’m just throwing out a number. Like what do you think would be the ideal happy path for us to get to?

Mark Roberge:
I think you can get to 75. I think the limiting factor that is very difficult in phase one and we’ll get to phase two and phase three and then I’ll unpack phase one fuse. Like in phase one, the rep is still selling to the customer. The customer success manager’s still working with the customer. They have to prepare. Like they have to just read the person’s name and back. Like they need to read the notes that the AI generates. Right.

Mark Roberge:
They can’t just like show up to meeting 30 seconds before and be ready. So there’s a little bit of like time, maybe you could push it to 90, I don’t know. But like there’s still some prep time that’s needed. Now phase two is, and I don’t know if phase two or phase three come together, whatever, but like I’m just seeing more progress on the, the buy like the seller side. So phase two is the selling and onboarding and renewals is, is done by AI instead of a human, which you guys are starting to do a little. And I love to learn more offline from you on that and examples. And then phase three is the buy ins done by AI and I, again, I don’t know I know there’s some progress being done there, but I just see more progress on the selling and vendor side. And then phase four is the functional boundaries of an organization that we know of today.

Mark Roberge:
Like you have a finance department, sales department, marketing, customer success, you have engineering product, they start to blur. Like those boundaries exist because of limitations of human capability. Like there’s not a lot of people that get a master’s in finance and then write code or a master’s in computer science and then join an HR team. Like we’re just limited. But those limitations go away when we’re very agentic driven. And when you do that, there’s massive benefits. Right, so. So those are the four phases and I want to unpack phase one because I think we can get there.

Mark Roberge:
I think we’ll start to see some early adopter companies get there in the next year. But do you have any reactions? Yeah, so.

Josh Schachter [Host]:
Yeah, so, I mean, so. So tell them what you said. Tell em again. Tell em again. Right, so phase one is more efficiency play. It’s more augmentive.

Mark Roberge:
Maybe it’s still humans are still like working with human sellers and customers. Support people are still working with human buyers and customers.

Josh Schachter [Host]:
Okay. Phase two is, is more autonomous.

Mark Roberge:
It’s selling the vendor side is driven by agents. Phase three, people and the customer success.

Josh Schachter [Host]:
Yep. Yeah. Is by agents have human buyers still have the human. Right. Phase three is your agents talking to your agents, your boss talking to your bot. Right.

Mark Roberge:
Agents buying from sell agent sellers, agent customers talking to agent customer support people.

Josh Schachter [Host]:
And I would imagine honestly we’re going to get there fairly soon. It’ll be like very clumsy at first, but it doesn’t seem like that’s so far away.

Mark Roberge:
I mean everything takes so much longer than we think, man. Yeah, like I, I was like, yes. In like SOMA of San Francisco. Yes. I mean like, but like, I mean that’s fine. There will be a couple customers where in some obscure use cases. But like it’s so funny, I was listening to this like very forward looking. I think it was Moonshots podcast with Kelly Perdue.

Mark Roberge:
I think so. Like my son’s actually classmate, a high school, his dad is one of the guys there and I love it, it’s very forward thinking. But I’m literally like listening to this thing and I’m like freaking out about the future of society and I’m driving through upstate New York to go visit Clarkson University with my younger son. Driving by an Amish horse and buggy. I don’t know what that it was. I don’t know how to explain that moment because I’m like literally driving through like rural New York picturing how AI is going to change this town, listening to all this and then driving by an Amish horse and buggy.

Josh Schachter [Host]:
Yeah.

Mark Roberge:
So it’s like, I don’t know, like, yeah, it’s going to come by. I just think it. When we live in these tech sectors, it always takes longer, I think, than we think.

Josh Schachter [Host]:
But yeah, there’s no greater dichotomy than that. And then phase four is like you said, is the roles. I mean, the roles is probably changing as earlier phase, but it’s just, even the functions themselves, you can just reimagine an entire organization. Yeah.

Mark Roberge:
Like, think about all like the red tape that exists from product figuring out what to build, salespeople figuring out how to message it, customer success, people figuring out how to get it adopted, finance people figuring out how to count the beans. There’s, there’s so much like, stuff that we live with today that’s unnecessary and it’s just because it’s a. The humans are on the front line. Yeah. And then like, you know, anyway. Yeah.

Josh Schachter [Host]:
I mean, listen, I can unpack phase.

Mark Roberge:
One, but I don’t know if you have reactions like, correct me, Josh, push back on me, tell me like, things you’re seeing that agree with it or disagree with it.

Josh Schachter [Host]:
No, no, I, I agree with, I agree with it all, to be honest with you. I think an interesting piece of it is the role transformation. There’s the functional department transformation, but the role transformation. And at what point does our job become, you know, say in cs. Right. Or in sales? It’s, it’s more of the customer face time. It’s being very strategic. There’s also an element of this is I’m going to be managing my agents, you know.

Josh Schachter [Host]:
Yeah. And that, I don’t know, that’s phase 1.5 or wherever you would put it. But that’s another kind of piece I’m thinking about these days.

Mark Roberge:
It’s kind of like, yeah, I think it’s through 2 and 3 because the agents start driving the front line. And I think if you were. Who knows, Josh? And like, we can, we can take a moment for, like to evaluate the societal impact for a moment that like, okay, on one hand you can like, oh my gosh, unemployment’s gonna go to 60%. Like, what is happens to capitalism, all this stuff. And like, I do agree that capitalism is not compatible with the post AI era and we have to talk a lot about that. But at the same Time. If you read the newspapers from the 1880s.

Josh Schachter [Host]:
Wait, let’s. But you, you bury, you buried the lead there. Should we just, should that be the episode? I feel like we can go very deep into that conversation.

Mark Roberge:
I guess you could get in trouble. Scary. But like, I’ll just. On the other side, we can if you want. Okay. But like on the, if you look at the newspapers in the 1880s, in the 1920s or whatever, 1950s, through each of these arcs of like industrial revolution, electricity, personal computing, there was the same narrative. All the jobs are going to go away, you’re going to destroy. And it didn’t.

Mark Roberge:
So that’s kind of where a lot of like the, you know, Ivy League economists are. They’re smart. They’re very, very smart. Right. And like they’re just like, yeah, we’ll, we’ll find new jobs, it’ll create more jobs. And you can see some of that. But I do have concern. And we can, we can go back to that, Josh.

Mark Roberge:
But like. Yeah, so wait, the quick point I want to make on what you just said, Josh, is managing the agents. Yeah. Like this, that could be like a little bit of like a symbol of where we head. And if you are early in your career, you might want to consider this, which is like if your job suddenly becomes managing a bunch of agents, you’re essentially like constantly refining models and iterations and getting data correct and like looking for opportunities and coming up with creative ideas. Your job is a little more like OPS today. You know, like that’s something I kind of wonder is like, are the future go to market leaders more like rev ops than CROs? Today’s CROs are like really great people managers. They’re really great at enabling people, hiring models, people coaching models, people enablement models, people training models, managing motivating people.

Mark Roberge:
And revops is really good at like enabling tech, tech stacks, data flows, process automation. Suddenly like the job looks a lot more like rev ops. And that, that’ll be true for DevOps and Finance Ops and HR ops and all that kind of stuff. It’s just something interesting.

Josh Schachter [Host]:
No, it is interesting. I mean, think about incredibly strong sales leader. You’ve evolved now, you know, beyond just that. But probably you wrote the book, you wrote, the.

Mark Roberge:
What was the title of Sales acceleration Formula was the one from 10 years ago.

Josh Schachter [Host]:
Right. And it was a very quant driven book. I mean, I love the way that you approached it. Right. It was like, you know, we’re, we’re going to be mathematicians here. Yeah. You know, it’s a science.

Mark Roberge:
Not an art. It can be predicted.

Josh Schachter [Host]:
Yeah, right. And that was incredibly important. But I mean, you wrote the book, Mark. It’s, it’s in the domain somewhere. It’s, it’s in an LLM somewhere. So maybe it’s not the most critical skill. Right. For a CRO anymore.

Josh Schachter [Host]:
Right. You know, and then, you know, Gong and everybody else wrote all the white papers on, on the words to say and not to say and you know, and there’s enough blogs about that and geez, enough clickbait posts, you know, on LinkedIn about that, about how to, how to, you know, how to conduct bant in medic conversations. So that might not be the skill that rises you to the top of the organization. So it’s probably a little bit of everything. But to your point, it’s probably the person who’s now in their 20s or something, just a little bit younger in their career that is AI native, that is going to know how to operate all these mechanics, all these agentic AI mechanics like you said. Yeah, exactly, totally.

Mark Roberge:
But hey, let me spend a minute. If you want to go over to the crazy stuff, we can.

Josh Schachter [Host]:
I kind of do, I kind of do. If you’re ready for it. But we could. Yeah, let’s, let’s finish the phase one.

Mark Roberge:
Let me just, I want to double click into number one because I think we’re going to get there. I think we’re capable of getting there now and not a lot of people are trying. And so let me just give you a taste of it. And a lot of this is rooted in customer success. I think the audience will kind of like it. So let’s just start with like where everything starts in like a, the, the beginning of a go to market process is usually trying to find a company to target a prospect. And a lot of people call that like identity, like defining your ideal customer profile and then like going after people in there and that the way that’s happened, like old school is like maybe once a year, if you’re good, once a quarter, someone refreshes the ideal customer profile and the way they do it and they actually do it wrong. Like most of the people, they do it, they’re like, they look at a sales funnel and they say, okay, which segments do we have the highest close rates? And these segments, they could be SMB, mid market, it could be North America versus Europe, it could be industrials versus Tech versus Healthcare.

Mark Roberge:
Like there’s a lot of ways we can cut this, but let’s just like keep it simple for example, and just say, okay, SMB or mid market. And the way they’re like, oh, our ideal ideal customer profile is wherever the highest close rate is and the lowest cac. That’s so wrong. Everyone’s gonna frigging freak out and clap so hard right now. Do you know? Do you know I hate to like. Can I cold call you Josh?

Josh Schachter [Host]:
Yeah.

Mark Roberge:
What is. It’s not cac. It’s not close rate. What is the output metric that we have to evaluate these segments are to decide if it’s in our ideal customer profile?

Josh Schachter [Host]:
Oh, man, I feel like I’m back in business school here.

Mark Roberge:
I know, dude.

Josh Schachter [Host]:
Professor Robert, I’m sorry, it’s not cac. It’s not ltv. You know I didn’t say that. No, it said ltv.

Mark Roberge:
I said it’s not cac. Yes. Okay, so I’m sorry, Josh, I don’t mean. Because I’m putting you on the spot, man, and this is like so. But thank you for like making this interesting and like going through the. Dude, it’s ltv.

Josh Schachter [Host]:
Yeah.

Mark Roberge:
Like, our job is to go find ltv. Lifetime value very correlated to retention. Right? That’s where we have product market fit. That. That is a customer that is going to succeed with our business. Okay. So like the, the people we want to go find are not the people that are easiest to sell. They’re the people that are going to succeed with our product.

Mark Roberge:
Now of course it has to be like LTV to cac, because I can’t like spend a million dollars to find a good fit customer that’s going to spend a thousand dollars with me. Like there’s a function of that. But like most people just aren’t just oriented in the cac and the easiest to close irregardless of whether it’s a good fit customer in the long term.

Josh Schachter [Host]:
Why wait? But Mark, we’ve tried to shake that for so many years. I mean, we know this, right? Like rationally we know that LTV and the retention of your customers matters so much, but yet it’s still such a shiny object. What your. What your conversion is on the initial sale.

Mark Roberge:
I think a lot of it comes from this, like quarterly earnings on the street. Yeah, it’s like short term revenue targets. When you sell a bad LTV account, you don’t feel it for two years. You’re probably going to move on to another job.

Josh Schachter [Host]:
Yeah, fair enough.

Mark Roberge:
So anyway, there’s like. So that’s just a perfect. And the other thing that’s really effed up and that could be fixed without AI. But the other thing that’s weird. Is okay. Now let’s say we do that right and like once a quarter we’re doing all these segment analyses, we’re doing SMB versus MCMarket versus Enterprise, we’re doing Industrials versus Healthcare versus Tech, we’re doing North America versus Europe versus APAC to figure out where our pockets of opportunity are. And then we have a really clear and we’re doing it by ltv. So good.

Mark Roberge:
Kudos, great job. And now it’s like, all right, we have a beautiful view of our ICP and we did an analysis to figure out that this quarter we can call on 75 new accounts because we’ve got like 50 reps and 20 SDRs and whatever. Guess how that happens. Some 22, 22 year old SDR that just graduated college four months ago is asked to go out and find those accounts. That’s who’s making the decision on one of the most expensive investments we’re making in our sales team, which is the 75 accounts that we’re going to spend the next six months with. That’s a low bar for AI to like improve. You know what I mean? So like now you throw AI at that problem and instead of doing it once a quarter, they’re doing it once an hour evaluating segments. Instead of doing it on nine attributes, they’re doing it on 900,000.

Mark Roberge:
And instead of like telling a 22 of SDR to spend a day or two days finding the next 10 accounts to call, they’re bubbling up automatically. That’s doable today. And that’s just one example. Right? Like that saves a ton of time and optimization. And now you funnel into, let’s just call that the ICP agent, whatever, you know, like target account agent. And once we have that list, then we gotta go find the 10 people that matter in that account. And again, a 22 year old SDR is doing that. And an agent can do that, an AI agent.

Mark Roberge:
And then we’re trying to figure out the messaging and that’s just like, that’s just like that. The demand gen side. Let me give you a taste of the sales side. So now it’s like, all right, I book a meeting with the director of marketing at Gen Z. So it’s great.

Josh Schachter [Host]:
I’m ready to go.

Mark Roberge:
Yep. So what do I do if I’m good, I spend like an hour. The night before I read through Genzymes 10k. Like you know, look at the, the woman’s background, where she went to school, what she studied, what she’s posting on social media. I, I dust off our like sales discovery call first meeting playbook and customize a little bit, maybe find a case study that’s like relevant to them, blah, blah, blah. And I execute the meeting. Maybe like 5% of the time my sales manager has some time for me to do like a prep role play together. And in a best case, and this probably happened 5% of the time my manager attends the meeting and is like slacking me tips as she’s talking and then I do the meeting and then afterwards I spend like a couple hours like updating the notes, putting the medic into the CRM, figuring out what the demo is going to look like, set up a forecast and then maybe my manager will listen to one of my calls this month and give me some coaching.

Josh Schachter [Host]:
Yeah, it sounds like a pretty perfect, a pretty perfect system you just laid out there.

Mark Roberge:
And that, that’s good, right? And that is like 1/100th of what it’s, it could be. Because what it could be is I booked the meeting with the director of marketing at Genzyme tomorrow, like whatever, next week. And the AI agent which is already pre trained with my entire methodology, I’m using force management and medic, it’s pre trained with all my product collateral, all my case studies, all my content, all my competitive battle cards. And it reads everything in the history of world about Genzyme in seconds. And it reads everything about the director of marketing I know, everything I have. And it perfectly customizes the first meeting based on the information that we have and gives it to me. And then it creates a buyer agent that I can practice with as much as I want with no dependency on my manager the night before I can just. And then it coaches me.

Mark Roberge:
Practice, practice, coach, coach. And then it shows up to me in the meeting as a slack bot or agent or whatever. And as Patty, the director talks, it learns and says, okay, this is what this means. This is the here, maybe try this question, maybe try this story, maybe try this product feature, maybe try this case study. And then after the meeting it does all the stuff, updates the medics, puts the serum, heads up the forecast, sets up the script for the next demo. And then it says, hey, you know Josh, that was a great meeting. You just elevated from a 5.1 to a 7.2 in champion development. You are crushing it.

Mark Roberge:
But you’re still only at 3.6 on sense of urgency development. So I put together a seven minute customized training module for you next time you’re in an Uber. Just listen to this.

Josh Schachter [Host]:
That’s possible today dude, and these are the questions you’re not asking. You’re not getting the information we need.

Mark Roberge:
And now you can start to see how sales time gets to 75%.

Josh Schachter [Host]:
Who’s doing this? Well, you know what, what vendors here, let’s, let’s let you plug somebody, right? Like, sure.

Mark Roberge:
I mean, we just made an investment this summer in Letter AI. That’s like doing that second part. That’s their vision with their platform. Some of their lead customers are like Lenovo and Red Panda. I’ve published. This stuff’s published on their site. You can check it out. That’s where they’re headed.

Josh Schachter [Host]:
Ladder AI.

Mark Roberge:
Ladder AI.

Josh Schachter [Host]:
Ladder AI. Okay, cool, cool, cool, cool. And, and do you know which organizations are being.

Mark Roberge:
Lenovo and Red Panda are some of their, like, lead customers. That’s what I have to study now. Yeah, like I, I need to just go to. If you want, if you’re curious, go to the Stage two blog. You can read a bunch of these different things. And we’re, we’re just trying, we’re. And if you have any, pin me on LinkedIn, like that is like the next year, maybe more of my life is, um, you know, like my podcast. The last two seasons have been like finding the founding CROs of all the best companies.

Mark Roberge:
OpenAI, Snowflake Databricks and tell em the story. And I’m shifting this year to just show me implementations of go to market tech. Whether it’s marketing, sales, customer success, customer support, rev ops, whatever that is. Creating minimum 3x step functions. That’s what you have to find.

Josh Schachter [Host]:
It’s so funny, man. Cause I’m doing the same thing with CS right now. I’m out there talking to everybody, just trying to figure out those case studies. Like my job is just collecting case studies every day of like, who’s beautiful?

Mark Roberge:
Well, let’s share, man. Let me help you promote it.

Josh Schachter [Host]:
Yeah, okay, we’ll talk, we’ll talk. I’m working on right now the most impressive CCO that I know. He, Colin Slade of Cloud Beds. And he has 150 AI workflows and agents that he’s built homegrown in his organization. Wow. He’s just transformed the entire organization. Yeah, he came in there.

Mark Roberge:
I’m going to follow up with that. I need to learn more.

Josh Schachter [Host]:
We’ll talk, we’ll talk. But he came in there. I mean, step one was he came in there, he turned the ops team into an AIOps team and said, great, whatever you were doing, stop it. We’re devoted now to building out all these Agentic workflows. We’re going to play on our own. You know, we’re going to, you know, we’re going to, we’re going to get this going. So. And then he, you know, and then he, and then he put applied pressure to the team, said, you know, if, if you’re not with us, you know, then as we evolve, there’s not going to be a place for you.

Josh Schachter [Host]:
It’s as simple as that. You know, we need to, we need to all embrace this together. So I’m happy to share more details on that. Yeah. So. All right, let’s, let’s go over to. I don’t have the dark side, but let’s go over to your thoughts on society where, where we’. Mark, listen, let me share.

Josh Schachter [Host]:
You, let me share some, some openness on my end, right? So, you know, I tend to steer a little bit more progressive, right. On that side of the spectrum. We won’t go too deep into politics, but it’s, I know we’re entering a kind of a fuzzy area here with that. Right. And I, I don’t know all his policies or what all his policies were, but about, geez, how many years ago was it that Andrew Yang was running for president? Sure.

Mark Roberge:
Yeah, that’s great.

Josh Schachter [Host]:
You know, and I, I read the War on Normal People. The War on Normal People, and it was talking, I think it was written probably about a decade ago at this point. So I was super impressed with his, his foresight into all this. And Andrew Yang was talking about, AI is coming. The first thing that’s going to hit the streets is going to be, you know, perhaps like autonomous truck driving for all these long hauls, you know, cross Route 66 type of things.

Mark Roberge:
Yeah. Three million drivers.

Josh Schachter [Host]:
You, you’ve read it? Yep. Three million drivers. Right. Like all like in their 40s and 50s. And like, what are they going to do when they get I profession, Right. A lot of them are probably armed, which for good reason. They’re on the road, you know, driving in the middle of the night and stuff like that and like, and, and what’s going to happen, right. And so we need to, to have some kind of security net for the future of jobs.

Josh Schachter [Host]:
Now at the same time, I also, you know, obviously like, very exposed to, to, to the, the analogies that we make to the other, you know, disruptive times in our history, right. When, when all these other technologies came out and like, at a macro level, there were still jobs, the economy went on, the world kept on spinning. I believe that. And I also believe by the way that two things that are dichotomies can both be true. So I believe that is the case, that we will find jobs for people. Jobs will converge, they will evolve. I also believe that it’s a rationalization as well, especially for somebody in my seat that’s helping to build this. So that’s my thoughts.

Mark Roberge:
I like it, man, I like it. I’ll build on it. So the thing that perplexes me and the reason this is rooted for me is like, to your point, like, you don’t want to be destroying the world selfishly. I have no need to do that. And for personal gain, we want to make sure. I think that’s one thing I’ve appreciated most of the tech community is wants to make the world a better place. And so this is an important dialogue now in the real term, like right now, it’s like, okay, should we just stop AI or slow it down or regulate it? There’s arguments for that. But I, I’m really have a lot of conviction that we shouldn’t because there are a lot of people racing to dominate the AI era.

Mark Roberge:
And I’m rooting for people that believe in human rights and freedom. And we’re at risk that some people might win who do not believe in that. And that could be very bad for our species right now. Like, it’s not perfect, but I think the United States is a pretty good believer in the way we’ve set up around human rights and freedom. And so I’m a United States citizen and I feel like I need to like, help us win that race and preserve that.

Josh Schachter [Host]:
Yeah.

Mark Roberge:
Now of course, like you could criticize, is the US good at that? All kind of stuff. But like, I think they’re pretty good at it. So that’s a backdrop now. At the same time, it’s almost like you’re building the plane while you’re flying it because we have to go as fast as possible to defend our freedom, but at the same time, it’s not compatible with our current society. Democracy and cap mostly good capitalism. Because capitalism right now is, you know, you graduate College when you’re 22, work through your 65, take two weeks vacation a year, work from nine to five, five days a week, eight hours a day, and retire with Social Security. That’s just the way. And we were strive for 2% GDP growth and 8% unemployment, 5% unemployment, whatever.

Mark Roberge:
Like, by the way, all that stuff is bullshit. Like, this is like in the same way that we just started the conversation today of like how we’re so iterative thinking. We have to rethink all of that because, like, I guess, like, I’ll root it in this. The way we assess all of our programs is around GDP growth. And at the same time, the United States is falling rapidly in the happiness index. I think we’re 22 right now. That’s ridiculous. You have the most powerful country with the strongest military and pretty good stuff in terms of like, not perfect, but, like, pretty good healthcare, pretty good gdp, pretty good freedom, and dropping in the happiness index.

Josh Schachter [Host]:
We all want to live in Scandinavia.

Mark Roberge:
Yeah, that’s where you. I mean, that’s where they have a good. But they don’t have perfect either. So, like, I think we just need to think about the design of these programs within the backdrop of happiness of people, not gdp. Because, like, if you look at the foundations of Adam Smith’s wealth of nations, which is the inspiration behind capitalism, his theory was perfect. It was like, if you use the invisible hand and open up free markets, you will elevate the percentage of people that have access to healthcare, shelter, and food. And it worked. But even he admits that it has a diminishing return.

Mark Roberge:
And it does. There’s a lot of research on it, and it can swing so far that people lose sight and they end up trying to become billionaires at the expense of being good parents, being good children, being good neighbors, being good, healthy to themselves, spiritual, whatever you want to do. And that’s what’s happened.

Josh Schachter [Host]:
Stop talking about me like that, Mark.

Mark Roberge:
Well, I mean, it’s like I’ve never met someone, like, on one hand, on one side of their mouth, they’re like, what would I do if I didn’t have a job? My identity. And like, like. Cause we’re wired by 100 years of capitalism on the other side. I’ve never heard someone say, f, it’s Friday, I can’t wait till Monday.

Josh Schachter [Host]:
Right?

Mark Roberge:
So it’s like there’s all this incompatibility. And I’m not saying, like, we should all, like, go. I’m not. And then the other thing you have is the word socialism is like a swear word in the United States. And I’m not really advocating for that. What I’m advocating for is a new Adam Smith philosopher that can appreciate where we’re going and help devise and blow up the whole, like, you know, five days a week, eight hours a day, two weeks vacation, work from 22 to 65, not be as good as a parent as you could not. Being as good as a child as you could not being as good as A neighbor, not taking care of yourself, your health, your mental health, your physical health, doing the hobbies. Like we have evolved as a species to be able to put a roof over our head and food on our plate and free healthcare without any exerting any energy.

Mark Roberge:
And I’m not saying we should sit around and play tennis all day, but it’s just like we have to reframe how we think about this. All right? So like, I don’t know, like, I guess my best guess right now is.

Josh Schachter [Host]:
Like, you’re saying we should, we should enjoy a little bit of the, the, the spoils of our.

Mark Roberge:
Just rewrite the whole thing. Yes, rewrite the whole thing. Because like, it’s, it doesn’t make any sense. Everyone’s like, I wish I could retire earlier. And then when you give them the opportunity, they’re like, you’ve taken my identity away. Right? And so, like, I don’t know where it ends up. But I like, you know, the, the big, you know, the big shot execs are talking about shorter work week.

Josh Schachter [Host]:
We should do that.

Mark Roberge:
We should start talking about a shorter work week, shorter work day, because that will help with the supply and demand balance of labor. I think there will be jobs there. My personal opinion is I don’t think AI is good at really true innovation and creation. And so I think it’ll be a long, long time where like, I think humans will still write, author better books, write better, do better paintings, create better music, do better at people want to watch sports, like, you know, create better niche restaurants, apparel. Like, I think like we become a really creative society using AI to do that rapidly. And those that succeed will live a better life. But what we see as middle class or upper middle class today will be the equivalent of welfare today. And we have to rewrite it as welfare because that’s a bad term.

Mark Roberge:
But it’s like if you want to just chill out, then you will have the life of an upper middle class person because we can use AI to like deliver free health care, shelter, food and all that stuff for 1/100th of the cost as it costs today. So we can have a smaller government, lower taxes, a shorter workweek and a better life.

Josh Schachter [Host]:
Mark Robert for president, everybody.

Mark Roberge:
And I’m definitely not running for president.

Josh Schachter [Host]:
That’s what they all say. That’s what they all say before.

Mark Roberge:
I’m just saying, like I. Please, if you’re out there, Adam Smith 2.0, start working on this. And we need a leader. I’m not going to say any political stuff other than the United States is stronger than a divided states. And we haven’t seen a president that can unite us in a long time and we need to find one. This country was killing themselves 150 years ago in a civil war. We’ve been in tougher times. We’re going to get through this.

Mark Roberge:
But we need someone to unite us.

Josh Schachter [Host]:
That’s if there’s one thing we can all agree on, it’s that the United States is a better is better than the divided.

Mark Roberge:
We need a leader to unite us. And we need Adam Smith 2.0 to philosophize and figure out how to redo this thing to be happier.

Josh Schachter [Host]:
Mark Roberge.

Mark Roberge:
Anyway, Josh, good to talk to you, man.

Josh Schachter [Host]:
Yeah, but you know what? It wasn’t so dark where you ended. I actually thought it was a very shining light that you ended on of where we’re going to go. Honestly, man, I look up to all of the stuff that you put out there. I look up to you in that regard. One of the smartest guys that I know in SaaS and respect so much what you’ve built with Stage 2 Capital, what you built years ago with HubSpot. Looking forward to the book coming out on Go to Market AI that you did in collaboration with Momentum. And yeah, thanks for spending some time with us.

Mark Roberge:
Yeah, you bet, man. The one more we didn’t get to you probably. If you’ve read anything about me, I read a lot of write a lot about the science of scaling, which is my new work. And that is like the biggest driver of unnecessary failure that I see is irresponsible growth and the lack of calculating when you’re ready to scale and how fast. It’s been my work over the last decade and I I’m going to do new work in the next decade and I think it’s going to be important blueprint for organizations, whether agentic or human driven. And so that book hits the shelves in February. I don’t really promote it much yet. I like to be available.

Mark Roberge:
But just so you know, it’s coming.

Josh Schachter [Host]:
Okay, well, everybody stay.

Mark Roberge:
And all the proceeds go to McLean Hospital and Mental Health 100%.

Josh Schachter [Host]:
Oh, dude, that’s awesome. I love that. Thank you for doing that. That’s great. All right, so we’re going to have you back in February.

Mark Roberge:
Great.

Josh Schachter [Host]:
Thanks, Mark.

Mark Roberge:
All right, Josh, Sam.