E58: Replacing Your Sales Team with AI Agents (Jason Lemkin, Founder & CEO @ SaaStr)

Jason Lemkin is arguably the godfather of B2B SaaS. He's the founder of SaaStr, the largest community of SaaS executives in the world, and an early investor in unicorns like Salesloft, Pipedrive, Talkdesk, and Owner.com.

I am done paying an SDR $150,000 a year for basically inbound spoon-fed leads.

And Jason is speaking from experience. In today's show, he explains how SaaStr made the radical decision to replace most of their human employees with AI agents.

Our AI agents are better than a mid-pack AE or SDR, and so we just don't need them.

But Jason has a stern warning to CROs and CMOs who think they can simply outsource this transition to a consultant.

Do it yourself, dude. If you don't roll up your sleeves in the age of AI and AI GTM, you will become obsolete.

We also discuss why the old benchmark of triple, triple, double, double might not be enough anymore, and the realities of what it takes to win in 2026.

If you're not willing to step up in today's AI age, which is so much faster, honestly, maybe join the fuckers.

Welcome to the Revenue Leadership Podcast. Here's a familiar scene. A deal's in commit and the buyer says, can I talk to somebody like me? Next thing you know, sales reps are asking CSMs after hours. 5 people are in Slack trying to remember who helped last time. And that one customer who always says yes is getting burned out after 3 requests in a week. That's why this episode's brought to you by User Evidence, the platform that turns customer proof from chaos into a repeatable GTM system. With advocates, references, and real customer evidence all in one place, your team can share credible proof instantly and keep deals moving forward instead of stalling. Companies like Gong, Ramp, and Workato already use User Evidence to capture customer proof, build buyer confidence, and shorten sales cycles. See how at userevidence.com. That's userevidence.com. Today's guest is the one and only Mr. Jason Lemkin. And for those who you don't know, Jason is one of the OGs of B2B SaaS. He runs one of the most important and prolific media companies in SaaS, which is Saastr. That's where I learned all about tech when I first broke into, into this space over 10 years ago. And so Jason was a founder, founded Eloqua, sold to Adobe back in the day, and then has been writing about SaaS since 2012. He's got the very best podcast in the market, which is Thursday's with Harry Stebbings on 20VC and Rory Oldriscoll, which everybody should listen to every week. It is, it is a must listen for me, including all the Saastr content on their own feed. And Jason has been an investor and owner since the very beginning. He led the seed round when Adam was probably like 17 years old or something crazy like that. And Jason has had a pretty prolific hit rate. Salesloft, Pipedrive, Talkdesk, Front, Intercom, gorgeous Algolia Greenhouse, Automile, and littleoldowner.com. And Jason's been incredibly helpful for me personally along, along my journey to owner. And so I'm really excited for this one. And we're going to talk about the future of go-to-market, which you've been talking a lot about. So welcome for part 2 because you were one of my first guests and thanks for doing this.

Thanks, man. Um, hey, as an aside, I wanna go on your question. It's funny that you brought up that Adam was 17 or whatever. Just an aside for folks, it's, it's, it's something I've never really talked about. Um, but I think of all founders as the same. Uh, this is just my nature. You know, whatever your age is, background, country, ethnicity, gender, I just don't care. Um, there's some negatives with that, right? Obviously some founders that have more life experience, have different perspectives than me, but I, I treat everybody the same. And the reason I bring this up is I think if you're looking for a sales role or revenue role, you should do the same. Like, don't be biased either way, but don't go in and say, hey, this is a younger founder, or this is, hey, this, this founder has got 80 years of experience and took 8 companies public. Just treat everyone as a human being you look up to and respect, and you may make better career decisions because of it. Yeah.

Certainly has helped me betting on it.

Yeah. I mean, Adam's the best, right? I mean, I don't really, but it's funny. Some folks in the beginning, it's been a long time, right? I mean, people, it's funny, even with Harry Stebbings, Harry did our SaaS podcast when he was, I think 17 or 18. And people thought of Harry as the young guy. Now he's becoming a grizzled veteran at 30. Yeah. So the world keeps rebooting. Founders are younger, right? So many founders are founding companies out of college, just don't go into that with any preconceptions either way. And just, just, we, innovation's everywhere today, including in sales and GTM. It's just everywhere. So don't go in with any, just go in and fucking crush it.

So we're 3 minutes in and I'm already going to take us off, uh, off script because I don't think I've asked you this publicly, but like, what did you see in Adam at, at that age? Because, you know, there's not a lot of data on a founder when you're investing in them that early. So like, what, what was it?

Well, you know, it's, it's, it's the, I try to, I'm not like actually that outgoing or social. I care deeply, but I'm not, I'm not Mr. Gregarious. So I try to do a lot of interactions with email and Adam was just, and now Adam went and copied a Saastr post I wrote where I wrote about the best cold emails to investors and he copied it for Owner and it was, it was a 10 out of 10 and he knew it was a 10 out of 10 when we talked later. Right. Yeah. And a fun thing was, and this is, this is just general sales because getting an investor is selling stock. It's still sales, right? So he sent me a great email pre-seed. Um, but the timing was off and, uh, I said it just wasn't a good match. And then when the seed came up, he reached out back to me. So I had two points of connection. Now that's small, but I can see that even before I met Adam, even before I got on the call, or certainly this was during, you know, I hate to even talk about it today during the global pandemic. So it was harder to meet people in person, but even back then, those 2 points of like high quality interaction, value added in a sense, right? To track progress, uh, do it in selling anything, including stock, right? 2, 2 great interactions is 20 times better than 1. It's so imagine you just get an email from a rep saying, hey, I want you to use Ownr or any product in the world. And you see a lot of great ideas and Here's another Mexican restaurant just like you that uses it. And the email's great. And then 90 days later, you get a follow-up email that's even better. Hey, did you know that, that, uh, that, uh, Sancho's, where I just used last night for, with Owner to get food to go. Did you know they used Owner? Did you know that just last month they saved, uh, they, they made, they brought in an extra $12,000 to the business? Like those two connections are so powerful. So Adam did that, advertently or inadvertently, that helped. And then, um, And then just, and then, so that's, that's specific to here. Those two points with lots of high quality interaction works in any sales. And then you just gotta know that someone's a force of nature. They have, it's so hard. And even since I invested in Odor and since we met, the world is, AI makes everything so much faster and more competitive. It's so hard, like it's so exciting to see companies grow at owner's pace or heck at cursor's pace or whatever. But The world is brutally competitive, brutally competitive. And so you've just got to find someone that's going to walk through walls and glass to do this. I was literally just DMing with a friend this morning that's an investor. There's like, yeah, CEO in our portfolio just quit. Would they raise $20 million last week and they quit? It's too hard. Don't want to do the work. And probably it's not that simple. If they raised $20 million, I don't think that they were like incompetent. It was probably the competition was probably AI. They woke up and they had 11 clones. Yeah. So you gotta find folks that will walk through that want this irrationally or don't invest. Right. Um, and then the third thing, this was a tough one and this was true of some of the other investors on the cap table too until recently. This is a quirky one. And, um, it was hard for me to get my arms around the space back then. And this will happen for folks joining spaces. I was trying to do diligence in 2021, understand what was happening. I, it was clear to me that some of the competitors in 2021 were kind of dated and mediocre. So that was a good thing to see. Okay. So you've got it. You don't have an owner back then, right? You don't have someone that's breaking the mold. So that, that's helpful to know, but it's, that's, it's necessary, but not sufficient. But like, it's trying to figure out what's changing in the world. Like, why would I make this investment today? What is the step function in the world? And I couldn't figure it out. I couldn't figure out why now for owner. I couldn't figure it out. Did the investment anyway because Adam was great. And then I met Dean, who's the CTO, the co-founder. When you have two, that's, that's as good as you're going to get at the seed. But I couldn't answer the why now question, which kind of stuck in the back of my mind. And it's okay. You got to take risk. But that was the one, one where I, one, one of probably my only really good investment where I couldn't answer the why now, but it was fine. Like you bet on the founders. Um, what helped was some traction, right? But it was interesting one. And, and, um, I think the answer honestly was just Adam wanted to do it. Yeah. Now we're seeing AI change the space, right? Now we're seeing some, and some of it a little bit was like 2020, 2021 changed how we ate at restaurants. It obviously, it made pickup and to-go, and it changed the ratio of table service versus other rest. So that was, that made Owner a bigger play, but it didn't really answer why now from a technology perspective. Not, at least not completely. So I don't mean to ramble, but, but it'd be nice to, I mean, if you meet a startup and you're going to go work somewhere or learn something, it, you know, I hate it when the sales rep does zero work and just asks a lot of questions. I think that's the lamest play of all of interviewing, but why now is a good one for early stage. Just, I'd ask the C, especially if you're going to work for a CEO directly, just say, I love you. This is great. I just have one question, right? Because you might not figure this out on the website or in ChatGPT. Why now? Yeah, it's a good, it's a fun question, right?

Yeah. But the meta takeaway is like, it's always about founders, especially at pre-seed, seed, Series A. I was, that was my bet. You know, like it is like, you know, it's a, it's a messy, noisy space. Restaurants churn at a, they close at a very high rate. And so there's a bunch of reasons not to do it. Sort of like investing in GTM tech. Nobody wants to invest in GTM tech right now. Like Clay is such an anomaly in that space, but non-consensus bet is actually to bet in some of these spaces that seem like they're—

now is a great time to be in GTM tech.

I think so too.

You know, we want to know the truth is, because we use all these tools, which is your conversation, they're all good, but they all could be 10 times better.

Mm-hmm.

They all could be 10 times better. So like build the 10 times better. Don't be intimidated. Listen, we love these tools. We can talk about them. We use them. I'm eternally grateful for everything. But yes, AI-enhanced GTM is, and we, we'll talk about it, is 10 times better than pre-AI. It's a step function. It's a step function for AI SDRs, BDRs, parts of marketing.

Yeah.

But we, but, but it's still in some sense just automation on steroids. We've just begun to tap what we can do for one-to-one personalization, one-to-one sales. We've just, and, and we talk and Amelia does, we show all the data on how now we send 70,000 hyper-personalized emails and we say that, And they are better than humans sent. Okay. This is step 1 for A. The 70,000 emails that we sent going into SaaS or London were better than like the 7,000 emails humans sent before that. So 10 times, but they're only a little better.

Mm-hmm.

Imagine when it's, we talk about, well, you know what hyper-personalize is? Just a little bit of data on a, on a customer. Okay.

One dynamic field.

One. Yeah, we do 3. Um, we do 3, but it's really barely 3. It's almost 1. What if this really is incredible, right? What if we really pull in every single competitor, every single adjacent things, we really remember every single data, every single time you visited owner website, every interaction you did with SaaS for 10 years, we could really pull them all together. We know exactly who Kyle is, why he would buy, and what he cares about today. And we send you the same quality email that Adam sent me to invest in owner. Those products don't exist today.

Yep.

The email Adam sent to me that he probably spent several hours crafting with a guy that's in top 0.1% IQ, he still probably spent several hours writing that. AI should be able to do that. It's just not there in GTM. And when it is, buy that product.

Yeah. We're not far though. And so I think the product might not be there, but you can build to that with a few different things right now and get pretty close. Like you can get, you can start to, maybe not the Atom Bar, that's like a different thing, but.

So maybe, but here's the thing, AI is smarter than us now. Yeah. So the emails that will be sent by an AI SDR, an AI AE, an AI marketing engine, they should be like the best email that could be sent by all of the hive mind. It's like, what's that show on Apple TV, Paradigm? Have you seen this one or whatever? The new Breaking Bad guy? All humans are connected.

Oh, pluribus?

Pluribus, pluribus. Yeah. And so all the answers are perfect because we're all connected. AI SDR should be as good as pluribus.

Yeah.

And they're not there today. It should ingest, it should know everything that has ever happened across your whole org, everything in your database, and be able to synthesize every piece of data in every bit of your database in real time, right? And send the world's best email. Today there's a bunch of reasons we're not there yet. One, there is real cost to that. That's one of the reasons these tools don't do it is because the token costs are high. To do everything. You can do a little bit of it with RAG and vector databases too. It's slow to do this with— if you want to add any reasoning to AI, it's going to take seconds or even longer. So it's really hard to send 100,000 emails if each email takes a minute to process, right? Um, and there's cost. But in a year, man, I just think we're, we're just starting. We're just starting. So I don't mean to take us off track, but I, I, but I think the innovation— we're— I honestly think we're in the bottom of the first for innovation in AI and GTM. I don't even think we're in the second inning.

Totally agree.

Right?

Yeah.

Inning one is approach the best person on human on your team. This is what inning one of AI GTM. Take the best human on your team, try to reproduce it at mass scale. That's all it is. If you train any of these apps, these Artisans, Qualifies, OneMinds, even AgentForce, we can talk about. We're in production AgentForce. Really, you're trying to take the mind of your best person on your team, Maybe 2 or 3 if you can, and just productize it with AI, which you couldn't do with general automation before. And that, that's a step function, right? If every email is as good, every communication, every chat comes up massively. Yeah. Yeah. But it should go, it should be like pluribus. It should go way above number 1. It should be the whole hive.

Yeah. So one of the questions is the, like recursive learning in, in like when you have all of these tools, one of this is one of the questions I'm grappling with today. So we've got OneMind and we've got like an SMS follow-up agent that we're building. We've got X and Y. How do you send, how are you thinking about centralizing the context engineering, the, the learning loop? Because the, one of the questions we're asking is like, do we build one mega model of our own to power all of the customer journey from AI BDR all the way through to AI support? Or do you have a, a mix of all these tools, but how do they share insight from one another? Like, have you thought about that?

It's a— if you're as far as we are or further, or where you are, it's one of the great questions you'll start to ask.

Okay.

And I want to give you my answer before we go, just for folks that are going to watch or listen, bear in mind, if you're not as far as we are, don't ask this question. Get a couple of cool tools into production that work, get it going for 3 or 6 months. Then you'll realize once you get 3 or 4 agents working, not just one, then you'll have data conflict that you'll, they'll be answering questions inconsistently. You'll have trouble sharing data. And after 3, 4, if you have 3 or 4 agents in production for real that are working, not complaining they don't work. If they work and you've done it for 3 or 4 months, you'll start thinking about meta agents. How could I have one agent manage all these conflicts and issues? What we're doing today is not that clever. We are using, for example, for outbound, we use AgentForce, Artisan, and Qualified. Okay. We use 3 and we may add more. We're, we're just basically taking our Salesforce database and giving each of them a portion of it. It's not that clever. And that requires Amelia, who is our, went from our head of CMO and everything to chief AF. So now she's gotta manage 3 different segments of our base and use 3 different tools. It's fine, right? Because it's better than humans. Um, but we could, but they don't re, the only way they communicate with each other is by pushing data back into Salesforce.

Okay.

Right. And that's okay for today. Right. And in fact, there's been a little bit of renaissance, as you know, with Salesforce. This is both a negative for Salesforce and that the agents in some ways are more powerful than the CRM. But it's also renaissance because Salesforce becomes the hub for all these agents. Right. So Salesforce is more important to us. Right.

I have a question on Salesforce that I'm going to get to because you and I have both been leaning into Salesforce more than before.

So, so, but it's a big question that, that I have a lot of answers going into 20 and for 2026, this meta agent versus the human having to be the boss of all the agents. I think, um, I don't think we're there yet. At least we haven't figured it out yet. Maybe you'll figure it out ahead of us. And if you're asking the question like another, maybe we should build our own orchestration layer that is automated with AI. I mean, listen, more power to you and love. And you know, you guys have raised 9 figures of capital and have an S-tier team. I don't think that is the right answer for 99.8% of the world. It might work for owner and I'd love to learn from you. It might work for us, but most folks should not think about, most folks should realize you need a nerdy human managing your agents. Someone that's a GTM nerd in a good way that loves this stuff, that's quantitative, that likes to get in the weeds, that likes to interact with agents, that likes training, that loves data. Someone that just loves data. You need that, that is at the top of your stack and your agents will have conflict.

Okay.

Right. And, and, um, we haven't figured it out. Um, but you, like you said, we all quickly see this issue, but you've gotta be in production with several conflicting agents to really feel this issue. So, you know, don't, don't put the cart before the horse. Okay.

That's really good advice. And so just going back a step for those who haven't been following us, uh, along with the podcast stuff, LinkedIn, Twitter, like just give them a perspective of how drastically your business has performed over the last 12 months and what you've built.

Yeah. So let me, so basically, and I think this is where when you hear from leaders like Ohner, you could almost get bad advice today. And what I mean is, and I'm gonna answer your question. Oh, we were just talking before we went live. Ohner isn't, isn't yet in the, in the echelon of Cursor and whatever, but, but, but you're a unicorn with an exciting product and great leadership and dynamic leadership. Your ability to recruit, as hard as it is, is, is better, if not easier, it's better than 99% of folks out there. Okay. And it's still not easy. You spend probably 30 or 40% of your time recruiting one way or the other. Right. Adam's insane. CEO. Right. I mean, I got a, you know, he was, he was, I got a text from him on Saturday. He's flying by Palo Alto to meet a candidate. Then he's flying up somewhere on Saturday. I mean, he's an insane recruiter too. Right. But for most of us, the single best, the single best I've ever been around in my entire life. Single best. Right. Most of us not only are the single best, including me, but we don't have the brand behind it. Like Adam was always great. He found you, but it helps to have a few great brands and a billion dollar plus valuation. It like, it, and so we don't have that. Okay. So that's why a lot of the advice from owner or better, just be skeptical of those CROs. Like it's great advice from Kyle and above, but it's, and so then I'm going to answer your question. As strong as the Saastr brand is. Okay. This is an enduring brand. Okay. Helped hundreds of thousands of folks. Half the public B2B companies have founders love Saastr.

Okay.

It's still hard to recruit. It's still hard to recruit. We, over the years, I've desperately tried to recruit GTM folks. And at first I tried folks that were refugees, right? They got fired from Owner, but I still knew Kyle and he, so I hired a bunch of like folks that were almost like Kyle, but got topped or fired. And I'm like, well, come hang out at Saastr for a while. Help me sell. And that was a, that was fun, but they either thought it was beneath them or had other issues and they quit. Okay. Then I tried to hire some kids, right? But they didn't want to learn, learn what we do or sell sponsorships. They, they, they, they just wanted to sell it. They just wanted to basically sell clay, something that was easy, right?

Yeah.

So we never cracked the code on our sales team, at least the way I want. I never had the stable. high-performing sales team. I had pockets of excellence, but it was never stable. So we go into May, SaaStr Annual 2025, and, um, we also pay people really well, dude. Really, really well. Um, and we had two folks on our sales team making very high six-figure salaries that just quit going into SaaStr. They just quit. They just quit for no reasons. No notice, no reasons, just ghosted. And listen, maybe it was right of them. It sounds critical. I'm no longer critical. And Amelia and I turned to each other, we're like, we're going all in on agency. Tonight, we're done with this. I am done paying an SDR $150,000 a year, an AE $500,000 a year for basically inbound spoon-fed leads and renewals and quitting on me. Maybe you can be critical of me. Okay, fair enough. Right. But I'm not like the worst boss. Like I'm pretty loyal. You've seen me. I'm pretty grateful. I pay people well. I don't, I don't tolerate laziness or other things, but do your job with me. I'll stick with you for 20 years. And I, at this point in my career, I couldn't take it anymore. So I'm like, okay, Amelia and I said, listen, anything we can do in AI in May, and May is a long time ago now in AI, like we're going to do it. We are going to do it. Right. And the first learning was, so we just went around all the sponsors at SaaStr this year and we just said, who wants to help us? And Artisan was there and we didn't know what the right vendor was. Okay. And you and I talked to SaaStr about what we were doing and we were still on the journey, but Artisan said to their credit, they're like, we'll, we'll help. We'll do it all for you. We'll give you. The right F forward deployed engineer, and we'll do it for you. So, and this is a meta-learning. We picked Artisan as our first agent, not because they were the best. They're very good and we can talk about it, but because there's a, there's a 2x2 of quality of vendor, but quality of FDE, quality of folks that will help you deploy it. Right. Another vendor, you know, is nameless. We talked to their head of sales, argued with us. He was like, well, you've got to pay us $100,000 upfront. We'll see if you're worth it. And then, um, I'm not gonna do anything until you agree to pay me $100K upfront. It— yeah. And then, and then I talked to the CEO later and the CEO asked me, why did we lose the deal? I'm like, you didn't lose the deal, but Artisan came in and did all the work for us, as did Salesforce with Agentforce, as did Qualified and others. So when you go to pick a vendor, pick someone that is the right partner with you to train and deploy your agent. We are not smart enough to do this. This is not SaaS of 2023. Where someone buys a product and hopefully later in the year they get it going.

Mm-hmm.

With agents, agents are too much work to train. Every agent that you're gonna use in GTM, including Clay, including the rest, requires weeks of training before you can go live. There's a, we can talk about why it requires. So if you're not getting the help that you need, don't, don't do it. And ask to speak to your forward deployed engineer or your, or, or your solution architect or whatever it is. And if you can't talk to that person, just find another vendor. I would rather have a worse vendor. And know who's, how they're going to deploy my agent, then the world's fanciest brand where I don't know who's going to do it, don't sign the contract until you talk to your forward deployed engineer or sales agent, right? So, I mean, we're going off course, but then we just went all in and I'm like, we're not going to have SDRs we pay, you know, 6 figures for quit on us anymore. We're just done. And then we said, let's go read their emails. And like, some of their emails were just terrible. And then the worst one was like, we had, we had a senior SDR that we paid a lot of money for, made up to $200,000. And basically what this SDR would do is I would go on LinkedIn and I would see, oh, Kyle left from owner to go to, uh, to, to qualified, right? And qualified's in our ICP. So I would say, go follow up with Kyle. And the SDR would be like, I'll add it to my list and get to it when I can. And half the time the SDR wouldn't even follow up. Now the agent just doesn't, the agent doesn't argue. The agent just follows up with Kyle Norton. So anyhow, we, we, the reason we pushed the envelope and have 20 agents running now was because, and, and listen, criticize me, I was just exasperated with turnover. I was exasperated with overpaying, not helping or not working. And I just couldn't do it one more time in my career. Now, if I was you and I had a large team and I had $200 million I wanted to spend, maybe I would come to a different conclusion. So I'm pushing the envelope on how AI can replace not the best folks, not the best sales teams, not your team, but 99% of the world. They can't hire your team.

Mm-hmm.

And it's better. And this is, this should scare people. This should scare people that don't wanna work harder and smarter than a year ago. Our AI agents are better than a mid-pack AE or SDR or BDR, better than a mid-pack. Not, maybe not mid-pack owner, but I've been doing this for a while. Better than the mid-pack people I've worked over in my career, not better than the best. And so we just don't need them. We're good. And so these jobs, these mid-pack jobs, they're, they're just in terminal decline in GTM and you should be aware of it. They're in terminal decline. We won't need the mid— now there will be pockets. Okay. And the hottest AI companies in the world can't find enough people to hire today. We could talk about that. And so they may absorb the mediocre ironically, right? Because they have to hire so many humans, but step up or, or maybe find a new role.

That's sort of the Jevons paradox of sales, which, which I think people are stumped on. Like, if reps are so much more productive than ever, then does the— are the unit economics of an incremental rep so good that you're just gonna— you're gonna actually hire more of them? Like, are— where we are, we've— we have like 9 high-impact production use cases of AI, ARR booked per dollar out on a per AE basis is 3x any team I've ever managed before. And I've managed some pretty good teams. But the argument is not, hey, let's hire less reps. I'm hiring like 25 people in sales over like a 75-day period. But how do you think that that applies more generally to the market? Like, does, does a rep being worth so much more mean we hire more reps? Or do you think this like mid to lower pack, these jobs just go away?

Well, let's step back for a minute. For how, I know it's different because it's a different company, but if you had to try to normalize it, how much more productive on a dollar basis, on a quota basis, is your AI-infused team today versus prior teams? Like how much more productive literally are they?

Yeah, 3x. 3x.

Okay. I might challenge you on it, but it's whatever it is, it's a lot. Okay.

Yeah. So it's not like for like, 'cause we do month-to-month contracts and it's SMB.

But even that, if it were truly 3x, if it were truly 3x, I'm spitballing. You could probably get away with a 30 or 40% smaller team than before and still hit your growth numbers. Not, not a third, right? But 30 or 40%. So there's a little bit off in your, in your math, right? Um, because if you look at a lot of the AI leaders that are hyperscaling, they all have sales teams. And in fact, their leadership is the same people we know, right? I mean, the CEO of Slack that was at Salesforce for 14 years just took over as CRO of an OpenAI. Okay. The CBO of Vercel was the COO at Stripe. These are not like brand new people. These are the classic best B2B people we know are running AI companies. But for a variety of reasons, they are running them leaner. Some of it is because so much of the revenue is PLG, right? You know, if the first $300 million at Cursor was all inbound, same thing's true with Replit. You just don't need as big a team, right? Because you don't, you, so there's some reasons, some of it's cultural. Like if you have that much self-serve, do you want the headaches of a human-driven sales team? Right?

Yeah.

Um, and then some of it is just assigning massive quotas, $2, $3, $4, $5 million quotas. The quotas, I mean, uh, the quotas at OpenAI for the enterprise team are pretty damn high.

Oh, I bet.

Yeah, they're really high and they don't have enough people. Like they don't have enough people, but they're really high. So I think there's a lot of confusing factors. If you're growing at OpenAI's rates or if you're Salesforce and you're repositioning your whole company for AI, you might hire a massive amount of salespeople. But it doesn't mean that the salespeople don't have to be much more productive. So I don't think it leads to— it will only lead to more net jobs because AI is such a huge amount of our growth in our economy today. Right? If— I don't know how many sales folks are at OpenAI, or all of them, right? All, all, all, and, and Anthropic. But let's say they have 10,000 reps on their sales teams, or could at the end of next year. Let's make up a number. That might be too high, right? That's plus 10,000, but how many are we losing at the public companies growing 8% or 2%? Probably more than 10,000. So I don't think it's net positive for, for if the new folks are, have bigger quotas and are more efficient and the old ones are being relentlessly brutal on efficiency. I don't know that's going to lead to massive job improvements. I do think that what will happen, and I wrote about this, the $250,000 SDR. I do think that the elite folks, the truly elite folks, not the ones that think they're elite on LinkedIn, the ones that really are 5 times, 10 times more productive for real, they should be paid compensation that is 2 to 3 times higher than it used to be. And this is already what's happened in engineers in AI, right? The average, you know, the, the average employee at Anthropic makes $400,000 and the average engineer makes $800K and plenty of engineers at OpenAI are making $20 million. But it's not, it's not charity. Like the productivity that they're expected to create is like nothing we've ever seen.

Yeah. The leverage is wild.

Yeah. Super high leverage. So traditionally in sales, there's been no leverage. Every year you've got to add more. In fact, there's been anti-leverage because it gets harder and harder to get that incremental dollar, right? For folks that crack the code here for real, which is not going to be most folks, but folks that crack the code, you may well find your way into a job that pays 2 to 3 times more before, if you can have someone managing instead of 20 SDRs, right? If you could have one human managing 20 agents, no problem paying that person 2 or 3 times more than a traditional SDR director of outbound. Yeah. Like no issue, right? But we're, but it's not charity. We're expecting 10x the productivity. Yeah. 10x.

But that's probably not a, that's probably not a BDR turned agent manager. That's probably like a, growth PM or a data science turned GTM?

Probably. I think in the short term, like, there are no magic hires out there. We're going to have to look at our own teams. Who— and it might be someone, it might be that nerdy SDR that has a math degree that still joined Owner because it was cool and is like pretty good, right? But loves, like, loves all the data, right? And so Kyle says, I'm going to promote her. And then he promotes her again and she's running the whole thing, right? I think we're going to find just as many of these people inside And if you put a job description on LinkedIn, I want a senior GTM engineer for our sales team, I think you're going to be unhappy who you get.

Yeah, we, we were trying to hire this GTM AI lead and it was really difficult to fill that funnel with people who, who had enough business acumen and, and sales savvy. We ended up hiring a guy who was at Atomic in the venture studio and like had started a couple of companies and so had, had the. Had the business savvy, but just loves AI and loves go-to-market.

You need someone, they gotta love it.

Yeah. And he's ripping, like ripping through projects and shipping stuff. And we built our cold outbound email program. He built the infrastructure in like 3 weeks and like everybody else told us that was like a multi-month project. And it was just like, prompt layer and a bunch of models and he was writing some custom code and then we were sending 1,800 cold outbound emails every single day from all of these domains. And, and we had help. We, we had like Eric Nozewalski doing all the domain infrastructure and helping us think through those.

Okay. That's a lot, dude. Like, like it is a good story, but most people can't do that. Why not? I just don't know if an ex-CEO that worked at Atomico that founded 4 companies wants to go to most startups. Yeah.

Yeah. Okay.

Okay. So again, going back to the beginning, I, I do love your stories. I want to hear 'em, but I, but I think I worry that people, like for most folks, I think they're, they're doing the opposite. They're putting up a job notice for an AI executive to manage this stuff. I actually think in 2 years that might work because we'll, we'll have veterans. We don't have veterans today. So when spaces change rapidly, you've got to find the person on your team that's the AI nerd. Some folks can hire the person, but I just don't think most folks, and we talk to like a lot of CMOs, Amelia talks to a lot of folks come to advice. Okay. They do not have the capabilities to do what you just described. In fact, even worse, we just, Amelia just had a discussion with the CMO at a leading $10 billion company that, that is AI first. Right. And they said, listen, we think we want to do an AI SDR tool. I'm going to buy it and just hand it to the SDRs to figure it out. No training, no anything, just, just had budget. So I'm going to spend $400,000 for this tool and I'm just going to give it to each SDR like, like we used to do with Outreach or Salesloft or Mixmax in the day. Here, go run your cake. And we can, you and I will laugh, but this is a very sophisticated team at a $10 billion plus public company. Just because that, that old paradigm doesn't work today.

Yeah.

Buy Salesloft, let the reps use it on their own. Right. Yeah, maybe you can centralize. So, so, so most like, it's just, um, uh, you gotta find that nerd, that GTM nerd that loves it, that wants to run it for you internally. And in fact, it is chomping at the bit to do so.

Yeah.

Chomping at the bit. Right.

That's, that's good advice. And so one of the questions I get, which is related here, is like, okay, my CEO says we want to do AI. And I want to pursue being an AI native company, but like my CFO isn't really giving me the headcount I need. Like we've got, you know, he's giving me like one extra RevOps headcount. They're asking that team to figure it out. Yeah. I don't really have a budget for like a bunch of speculative tools. And I'm really, I'm surprised by the number of friends of mine working at good companies who are in this position who are like, I just like don't have the budget to pursue it. And I'm like, What are you talking about?

It's the wrong way to think about it. Okay. So, so yes, this is, this is a real story. Here's what you have to do. This will solve all these problems. Okay. Pick one tool that you, that, that, that you think you believe in, that looks like, that is, that is the easiest to, to deploy, that solves a medium or higher ranking problem. Okay. And for many folks, it is this AI SDR, but it doesn't matter. Pick, pick one problem. Could be RevOps, right? It could be— pick one, find one tool you believe in based on a demo and some research. Okay. And here's the second, just one that, that is easiest to deploy that you believe is, and that it solves a real problem in your org and then deploy it yourself. The first one is not about headcount. It is not about, but it is only about the budget for this one tool. You only need $50K, $60K, or $100K, which is not nothing, but find the budget. If you don't do it yourself, It's like, it's like hiring a CRO that's never sold, but it's not quite the same, but it's the same fallacy. 99% of folks we talk to, and we, we almost are like a free consulting shop. So many folks come to us at SaaS now for help doing this cuz of all of our data. Do it yourself, dude. Mr. CRO, Mrs., Mr. CMO, Ms. C, then they're flabbergasted. They haven't deployed a tool themselves, uh, in a decade or ever, right? Yeah. And you have to oversee it. Pick the vendor, do the training. Understand, do you, do you even know what training is? Do you even know what a forward deploy engineer is? Do you know what ingestion means? Do you know how these tools are trained? Do you know what it takes? And it's only 30 days of work, guys. So you're going to have to take, it's going to take you 20 or 30 hours. Okay. Of those, of those, of those 30 days, you're going to have to make the time. And then my God, you'll be an ex, you'll be ahead of 98% of the world. You've actually qualified, chosen, demoed. Trained, ingested, and deployed an AI agent at scale, that is better than 99% of the folks talking about it on Twitter. They haven't even done it.

Yeah.

You do one. And then what you do is you walk back to your CFO and you do what we do. Hey, we just sent 70,000 automated emails. They're better than humans. And it generated 15% of the revenue for SaaS for London. 15% of the revenue.

Wow.

Can I have, can I have some more budget?

Okay. Sure.

Do it yourself. Pick. And the other mistake is sometimes people try to do 5 or 8 tools at once. Of course you need a lot of budget and a lot of people. No, do it yourself. Roll up your— guys, no one— we stop rolling up our sleeves. Okay. I'm doing it too. At this point in my career, you can see I'm still rolling up. No, like literally and figuratively. If you don't roll up your sleeves in the age of AI and AI GTM, you will become obsolete. This is not an agency game. Not today. Maybe in 24 months. This is not a game of, oh, I've got my new job at Owner as CMO. I'm going to bring in my 11 agencies from my last job that I used in the aughts. Okay. Here's my 11 agencies. I've got my Salesforce agency and my outbound agency and my left-bound and my right-bound agency that I used in 2015. They're all going to fail. Yeah. It's, you are the agency for now, just for, at least for the first, at least for the first one. And once you've done it, because literally if you haven't trained an agent, and it's not that hard, take almost any agent, it's basically the same thing. You, you, you point them to your, to a database. You could even just be a, your website in the base case. You upload your prospectus, you upload some documentation, then you upload some more. Okay? And then out of that, most of these agents will create a list of questions, right? Or sample emails, 10, 20, or 30, and you iterate them and they A/B test them and you talk about it and you see what they're wrong. And when they say, you know, owner, owner is great for 100-chain high-end restaurants, that's wrong today, isn't it?

Yeah.

So you, you X out that question. You say no, owner's side, owner scales much more than it used to be. But our core audience are smaller restaurants with a handful of a single location or a handful with a significant amount of to-go business. Right? And you just keep going through those questions and iterating them. And then you, then it'll, if it's email-based, it'll start sending the emails and you have to read them for an hour or two every day and see what's wrong. You have to do it yourself. No agency is going to get this right. And this is what you do. No agency is even going to really know what you do. What's that?

This is what you were doing with Artisan or Qualify.

Yeah. And you need to do this as Mr. or Mrs. CRO or CMO or VP of sales or you can't hire someone to do this today. Maybe later, maybe later. And HubSpot hasn't figured this out with this AI training. That's why like, like basically half of HubSpot's, um, agencies are obsolete.

Hmm.

Because they can't work. If you listen to what Yamini said this year at Inbound, she's like, we're radically, we're gonna put, we're gonna get more and more revenue through the channel. but we need different agencies. The agencies at HubSpot back in the day, you'd, you'd want, you'd have your local group and they'd hook up HubSpot and they'd, you know, they charge you basically 1x or 2x what HubSpot did and they'd hook it up and, and get your basic campaigns going for you, right? Yeah. This is now work for training an AI on all of your customer data, all everything, and making sure it's all automated and it works. You gotta do it yourself. And once you do it once, and even I was intimidated. The first one we did at the start of the year was a general agent called Delphi, which makes digital clones. And, um, we were early on this and I, I, I saw that, uh, uh, Brian Halligan and Lenny and a few others did it. So I, I just bought it. Um, I'd never ingested or trained data. All right. And I had to do it myself. I did it every single day for a month, every single day, first thing in the morning.

Yeah.

At first, even me, first thing in the morning, wake up, see what, see what the agent said, see what's wrong, and then just start editing it and changing it. And the agents, like, this is the beauty to it. It's the hive mind. They remember.

Yeah.

And every day it gets better. Like, for the most part, they do, they remember everything you tell them, and the hallucinations become a minor issue, if at all, as you train it more. But if you don't do this, you're gonna, you, you, you will have no idea what you're talking about. Like, literally, you will be utterly ignorant in the age of AI. You got, you did it, and I, I did it. And then I did it, and I proved it to Amelia, and Amelia did all the rest of them, right? And she's better and smarter than me, but I had to do the first one, even in our little org. To make, to, for it to make any sense. 'Cause I didn't know to tell her to go do Artisan.

Yeah. And, and it's funny you mentioned that because I, I didn't do this all that intentionally. It was just like, we were a small company and I like this stuff. So Momentum was one of our first AI implementations and I was in there writing prompts and writing custom fields in Salesforce myself. 'Cause I, I implemented Salesforce on my own way, like way back in the day. And so like I knew how to go into the, admin view and create fields and add them to layouts. And so I was just in there playing around being like, ooh, like what can I know about my business tomorrow? And I would build a field, think through, all right, like how, what's, how do I wanna structure the data? And then write the prompt in Momentum. And then I would, it would run like this was back before they did backfilling. And so then at the end of the, the next day I would have my laptop open after the kids went to sleep and I'd be like, oh, I like, I'm really excited to see what's in these fields now. And I would look at a report and I'd look at what's in there. I'm like, oh, like it missed this. Or like, I see my prompt was like too basic, or I need to point it at this other thing. And I was just doing it because I was interested. Like I took a Replit 100 Days of Code course and I was playing around with that. And then it just like, it was of interest, but now, now I fast forward and we've got this like AI lead and my head of RevOps is like incredibly AI savvy. And so I'm, I'm much I still take a bunch of demos and I'm thinking through the broad strokes, but I'm not anywhere near as close.

But you did the first one, but you got the first agent going yourself. Yeah. That's the thing. I will not all going to be as nerdy as you. They may need a little bit of help. They may not do the replica course, but you did it yourself. You, you trained it. You, you, you, you went from evaluation to success yourself. Every day you were reviewing that until every— here's the key. Every day you got to work with the agent. Until it works reliably.

Yeah.

I'll give you another fun example. Like, so we, I like Momentum and Attention both. These are great apps for RevOps, right? Buy one of them. You recommended it back to the day and we had a tiny team, right? So we decided to use it, but we just sort of put it on the, we didn't pay any attention to it, right? Right. Didn't think it worked. Did we? Didn't train it. We did no work. This is not the fault of the vendor. We just didn't have, like, we're like, okay, we got 4 people on our sales team. You know, now we have less, we have agents, we're like, eh, RevOps is not a big problem, right? And then it was the classic issue that, that, that, that you brought up and Marshall brought about. Finally, finally we're like, why is this thing not working? And we're on our team Zoom and we have this sales guy, AE, that we're paying extremely well. And he's like, we're like, we're not seeing any of your data from Moon. And he's like, well, the app doesn't work. We're like, okay. Okay. Shame on us. We never put in the time. We never put in the 30 days. So we, we kind of complained to them that it doesn't work. And then I'm like, I just put them on live on the Zoom. He's like, why not working? I'm like, Do you see that Google link at the bottom left? You gotta link up your account. He's like, I don't really want to. And he links up the account and it showed he'd done nothing for 30 days and he quit that day.

Wow.

So there's a lot of interesting stories. Marcella had the same story at Mango Mint when she brought it in. Someone quit that day too because it showed the lack of activity. So it's a recurrent story. Um, and it's interesting, but she, but like we thought it didn't work and it didn't work until I went to, we, we took the effort to have this one guy just hook it up. That now we rely on it daily, right? But that's an example of even us, because we went from 0 to 1 and then 20 agents. For us, just RevOps was not as high on the priority as it was for you. How many folks do you have on your sales team today?

100.

Okay, so at 100, RevOps is one of your top 3 priorities, right? In some ways it's number 1, right? That and recruiting in some ways are number 1 once you have it dialed in, right? For us, it's like, whatever. If we got, you know, we don't have our process dialed in, it's not It stopped. Like, I got 99 problems. So, so we didn't train it. We didn't do it. And so it failed. The deployment of that agent failed until we had the aha moment that, uh, the reps hadn't actually hooked up their accounts to it.

That's funny.

Shame on us. But now we rely on it constantly, right? Now, now we love it. Many folks that will watch or listen will go through the same thing. Like, they'll buy a product, put it on the shelf, never really get it deployed, get the agent going, get it trained, and they And they will say it's a failure. There are so many, there are many reasons we had the failures of the AI SDR failures in 2024 were all LLM based in my experience. It didn't work. The products didn't work before, before Claude 4, they didn't work. It was slop.

Yeah.

Yeah. It was slop. There was no point. Now they're, they're, now they are above the line. They're all above the line. So it's the, the, the reason they fail today is that people don't roll up their sleeves and train the agent themselves. And if that sounds intimidating to you, how do I train an agent? Right. Like this is not something you learned at your last job, right? Going back to our point, then pick a vendor, pick a leading vendor. It honestly does not matter which one you pick today and talk, bypass, tell the sales rep. I literally did this the other day. I want to talk to my FDE. I want to talk like, great, I don't even need to do a demo. I want to talk to who's going to help me deploy this agent, right? I'm still waiting to hear back from the rep. So that tool will not be one that becomes our point of first agent. But be kind and just say, literally, I want to get on the phone with Jason or whoever it is. And show them what you're doing and have them walk you through it. And then that if you get that partner, you'll realize that training and ingestion and orchestration and all these nerdy concepts are mostly just making sure your data is uploaded to a website and that you, and that you answer a bunch of questions and that you read outputs every single day for a month. It's that simple. You can do that. We know how to upload, we know how to use websites, and we know how to read and say this and tweak, tweak text. It's not as intimidating as it sounds on the— where everyone's throwing around big fancy words like training, it's— oh yeah, yeah, it's like— no, these are, at the end of the day, these are all SaaS apps, my friends. They're just SaaS apps that have an LLM hookup to them, and they look like SaaS apps, they smell like SaaS apps, their GTM teams are SaaS people. So you can do it, but don't— but just understand what it means. Don't be intimidated by what we're saying.

So I want to double-click into picking vendors because we have similar, I think, opinions on, on this. So you're mentioning like pick one thing, make sure you do it really well, be careful about vendor selection, implement it yourself. But I think one of the things I get asked about the most is like, where do I start? Because people are sort of freaked out by AI BDR now because it was all hot and then now everybody thinks it sucks because the—

Why are you getting boxes?

Because everybody's like, oh, uh, outbound is dead. The email, the inbox is dead. They're wrong. I would agree.

But, but where, where to start? Okay. So listen, I don't believe that. We could talk about why. If you believe it, great. Don't, don't pick that first. I'll tell you the lowest hanging fruit for 98% of folks out there. I know I use different nomenclatures than some is adding AI to inbound. My God. Then this is just, I said at SaaS Europe, but there is no excuse today. For you not to get instant answers to your questions. There is no, there is no excuse on planet Earth for some 21-year-old, uh, SDR to qualify me whether I'm worth their time to talk to an AE. There is no need. We use Qualified. It doesn't matter what you pick. You go in, it answers just questions on the SaaStr Annual website. Anyone can try it. Go to saastrannual.com and talk to the digital Amelia, video, text, audio, however you want. And she'll just ask you what you're interested in. And if you wanna spot, buy a ticket, sponsor, access to it, if you want to sponsor for us, it will automatically qualify you. Oh, Kyle, great, you're interested in sponsoring. Where do you work? Owner, restaurant industry. You know, honestly, Kyle, that's not our ICP. It's not a good fit for you. Okay. Um, I work at an AI B2B company. I work at Replit. That's a perfect fit for our SDR. Can I connect you with David right now? And, and it does the whole Calendly scheduling, which is not— which has already been done, but it instantly connects it with the qualification so that that lead is qualified in the moment. Any time. Yeah. And so if you want, if you think about it, okay. And I advise everyone at any, any senior level, any one of the management team or founders over the holidays, slow it down. Go to your website in incognito, in a fresh mind, and try all your journeys. Try, see how good your support is. See how good your qualification is. If you have a brutal contact me, see, see how much you like contact me. Where does that contact me go into? Ghostland?

Hmm.

See how good your support is. Would you wanna buy your product? Be honest. Is this a slam dunk? Right? And so fix one of those things. And literally, if we can talk forever about outbound SDRs, it does work. But even if it doesn't for you, I guarantee 99% of you, your inbound process is an F or D. It's not okay with AI. It's not okay to not have all my core questions answered day or night with no lag, with no delay, answered really, really well. And that I don't have to talk to some 21-year-old kid that doesn't even know what the product does before I can talk to a rep. It's insulting. It's insulting to my time. And I just won't do it. I don't have time anymore in the age of AI. I don't have time to set up a call next week to see if I'm worth your time. Yeah. Right? Literally when I was in, in London, I tried to buy a $10K product. Okay. I emailed them. I emailed a rep that I knew. It wasn't worth his time. So he passed me to somebody else. The other person, it was worth their time, wouldn't answer a single question until I got on the call. Yeah. If they had an AI bot, it would just answer them. Like, I wasn't looking for a discount. I just wanted a couple questions answered and I would've bought it in real time. Like that process is just not okay because it's no longer necessary in the age of AI. There's just no need to have that high friction sales process that benefits some team's funnel in theory. Right? Because they can score a lead. AI can score a lead better than a human, dude. It can score a lead better than a human today.

And, and so how would you more conceptually think about like picking the first use case? Is it, is it going back to like, what are the business priorities? Is it just like actually zone in on what AI is good at now? AI is good at support, it's good at inbound. So like, those are the easiest places to—

maybe not good. I would listen, you could do either of those things. It could be what you're passionate about. That's always a good First place to start. What do you care about? Like you were into the rev ops, so that's what you did, right? If you're, if you're a little nerdy like us, pick something you're passionate about. If you have no opinion, do the clean look at your website, do the incognito from scratch, new email, new domain, and see what breaks your heart.

Yeah.

And do that one, the one that breaks your heart. Stuff's even on SaaS or website stuff breaks my heart. I literally, you know, I literally, I can't tell you how many websites like You know, we've been a Brex customer for years. I tried to get an answer question in the bot. I couldn't get any question out of my bot. It's terrible. It breaks my heart, right? I had another fintech I tried to use. I got an email back saying I would, I would get a response within 48 hours. Go fix that problem. Maybe it's just support. Obviously AI works well for support. Look at Decagon, Sierra Fin. You don't even have to pick a high-end product. You can pick a cheap product and train it. Just whatever breaks your heart, fix it. Because then you can go to your boss or yourself and have a win. Like, if you think back to the early days of SaaS, why did, why did we take off on eSignatures? Why did other people take off? There were a lot of reasons, but somebody could bring an instant win back to the organization, right? That was so— I remember in the old days of eSignatures, we went out to Comcast. So it's kind of brutal who brought us in, and they brought us to the, to the old fax room where there were 100 people that managed the fax lines for cable contracts, for cable, for, uh, for, for, for cable internet for business. It was all empty, dark. The lights were— this was straight out of a sci-fi show, but they looked great to their bosses, didn't they?

Yeah.

We repurposed 150 people and saved all this money and now customers can sign contracts instantly. And they started to send us these, these things for their internal newsletters of I'm a hero at my company. I'm going to hero at Qualcomm. I'm a hero at these hero things. So pick, pick, pick one that breaks your heart that will, or, and if you're not the CEO, pick one that will be fun. You'll be a hero at your company. A hero, a hero, right? Um, and, and if it's not, if you don't believe inbound's a problem for you, I don't believe you, but at least your support is probably terrible. I can't tell you literally how many top— go to, go to like 10 top tech, hot, hot tech companies, try their support. I bet it's 8 of it's terrible.

Yeah.

Terrible. Terrible.

The only support experience that stands out as good was Oura Ring. Because it's ADA, like actually an old school player in this space. And you can get everything resolved instantaneously.

That's it. So there's no excuse. And the fact that ADA is old, but so is Intercom. Yeah. The AI is new. So Replit was old, Qualified was old, ADA is old. Zendesk's AI is really good. Zendesk is old as AF, but the AI isn't. AI is actually, and for founders that are Debbie Downers at the, they, when they watch this, that are like, man, I'm getting lapped by the AI guys. Look at Zendesk, look at Ada, look at Intercom, look at Replit. The— these apps have been utterly changed by putting incredible AI and AI agents inside of very old software. Very old software.

And how do you recommend people look at that decision, either using the AI within their current stack? So like, I have Intercom, I'm adding Finn. Or I've got Salesforce Agentforce or going to the like insurgent startup that's like ripping 10x year over year.

Yeah.

What's your advice there?

This is a good insight for Amelia. We talk to folks all the time. We do so many consulting calls. We didn't used to do this. And they're like, I'm doing a 10 vendor bake-off.

Hmm.

Okay. Not only did this not work pre-AI, but in the AI age, when you have to train an agent, it's just not possible. So pick 2. Pick, pick, listen, if you're already, if you want to try AgentForce and Finn because you're using Salesforce Intercom, great, terrific. And then pick one hot startup, not just because it's hot, but because honestly, someone like you is using them successfully. Ask for references for real. Kyle will respond. Everyone I talk to that's like you, they're getting, they're getting like 50 reference emails a month. Right?

Just don't ask me for time.

Yeah. Just send me an email and I'll give you an honest answer. Yeah. So just have—

and I will hit function on my keyboard in, in Whisper Flow and I'll like dictate a 4-line thing in 30 seconds and I'll give you your answer.

Just don't have to wait. I think Michelle from Mango Mint says she gives, she, she does like 10 or 20 a week. She gives them a one, a one-word answer, but that's enough, right? Find out. And so just do the two and just do it and just try them both and talk and find out who's gonna— and that, that's enough because here's the honest truth, like there are differences, right? And, and, and, and Intercom's great at showing how Fin is better than Dekacon and Sierra and like, and, and you know, who, which, what, who cares? They're both, they're all so much better than pre-AI that what matters is can you get it into production? It doesn't really matter. They're, they, and these tools are evolving so quickly that during 2026 they're gonna get so much better that that's going to lead to convergence and it's going to lead to, uh, you know, it used to be years before vendors would have feature parity, right? Now, now we're down to weeks. Yeah, we're down to weeks. So it's a journey you're going on. So just, and they're all, they're all, all the leaders, if you can get them trained, if you can get them into production, they're so much better than before that just pick the ones that's the right match for you. It really doesn't matter.

And make sure that they can support deployment.

Yes, support. It really, it does matter. And I actually, I honestly think that as we all get smarter and as we deploy our 21st and 25th agents, we, we can focus on minute feature differences that frankly will pro— they'll all clone each other in weeks, right? But right now, when it's a step function, right? It's like, get a great one into production and just do the one that's the right fit for you and don't be overwhelmed. And we list, if you go to saster.ai/agents, we list all the ones we use. Uh, you can see them and we have lots of data, saster.ai/agents, pick any of them or others that are leaders and just go for it. Right. And, and do, but 2 is enough for the big, I don't know why, but we talk to so many folks that are doing bake-offs of like 10, and I guess it's because people are confused because they haven't done it before.

And they want to see it all.

Yeah. In the old days, your bake-off would be the one I used at my last company and the hot one today. You do it too, right? And you'd pretend to look at the third. I genuinely think people are trying to do 8 to 10 vendor bake-offs, but it's, it's even if you could train them in an hour, you still couldn't. But if it's, if think of, if it takes you 30 days to have to really train it, you can't do 10. Yeah. You, it's impossible.

We, we've intuitively done that. Like I would, I gravitated towards the startups where I'm like, they're moving fast, they're AI native. And, but, but just like out of habit, you know, like, uh, looking at, um, AI knowledge base. So we looked at a bunch of, of exciting startups. One of them was like a very high profile founder and I was like, oh, this is going to be awesome. And we stacked it next to Guru, which we use for it was mostly, mostly support knowledge management. And we went through the pro— we, we stood up two startups and played around with them. And then we stood up Guru and we were playing around with it. Like, wow, Guru was way better in all of these ways. Who would have known the like stodgy incumbent would have actually had the better AI tooling and it worked more consistently and it was more elegantly weaved into the UX. And so now we have Guru across our entire go-to-market org. It's where all the knowledge is and it's really good.

But yeah, I mean, I haven't, uh, Rick Nucci, who founded Guru, he, he's wicked smart CTO, OG of the thing. The fact that he could take a, what, probably 10-year-old app and then infuse it with Claude 4 and the latest LLMs and make it great. It's just like Zendesk.

It's one of the best. It's honestly one of the best incumbent, incumbent AI experiences that, that I've seen. And I, I'm curious, uh, to get your opinion. It feels like one of the reasons that so many of these incumbent apps are bad is they're using crappy models and they're like constraining the token consumption. Cause like I'll go into them. I'm like, it feels like I'm using GPT-3.5 here. Like what is going on? And then the startups are letting you rip on the best model and losing money. Like I, I only intuit this.

It's a real issue. Listen, it's, it's, it's a real issue. I'm on the board of a company, you know, Gorgeous, that does AI support and more now, conversational commerce for e-commerce, for stores. Yeah. And at their scale, token costs are a huge deal, right? They just got one of those $1 trillion little trophies from OpenAI. And it's millions. And we were discussing at the last board meeting, approving an incremental $4 million in token costs for next year. Okay. And they're a startup and they have capital, but they don't have as much, like they're not as well funded as Owner. So it's a real question. right? So it's a set of trade-offs and I don't think everyone, and it's a good point. And so that's an example of someone at the bleeding edge that still has to think about these issues versus someone that just raised a $20 million round. Uh, they literally don't, they don't, they don't care. Right. Um, having said that, there's so much fluidity going to like Guru and Zendesk. I just think it's more important to do the two vendor bake-off and pick one.

Yeah.

Right?

Yeah.

And if you, and if the one you, you, you, you pick has some lame copilot that, that is token rationing and it doesn't work, you don't need 150 IQ to see that those answers suck, do you? Just don't pick it. I'm always surprised. Just don't pick it.

People aren't in the tools enough to have that like taste, you know? Like they like can't tell the difference that, that it's like, oh, this is a crappy model and it's spewing you bot breath. Slop versus, oh, th— this is clearly wired in with Gemini 3 or 4.5 Opus. And, and like, I, I just think you need to be in it so much to develop an intuition because without the intuition, like you're, you're sort of hooped. You're gonna keep making bad, bad calls.

Yeah. Although thinking about it, you know, the real issue, I think, I, I don't know what your experiences are. I think that when you see that using crummy models or not using enough tokens, right, it, the real answer is Probably because whatever you're looking at is not core to the vendor. Like it's, it's a copilot. All copilots are terrible. They're all add-ons. They're all mediocre value, right? If it's core, even if you're having to be thoughtful about token costs or not, you're gonna, you, you, the competition is so high, it's gotta be pretty good. It's just when it's an add-on, that's when I might not buy an a, a random add-on, right? Fin is not an add-on. Fin is core to Intercom today, right? It is also core to Zendesk. If it's a random add-on to my old vendor, right? Uh, I, like, I don't mean to, you know, is it core to Gong today? I think so. Right? Is it core to Outreach or Salesloft that are private equity owned or other things? I don't, I honestly just don't know. I'm just out of the loop today. I don't know. I could imagine they're under more stress to ration the tokens when it's, it's close to their core, but not quite their core. Right? Those might, and those might be bad examples. I don't mean to knock on anybody, but I just don't think anyone is going to, everyone knows the bar, the competitive bar today for AI. You got to meet it. If it's your core, you have to meet it. Whether you're going to meet it, whether you're going to be more clever or whether you're going to bring in like cheaper Chinese models and open source models. Those are clever conversations, right? Um, rather than just having a lame copilot.

Yeah, I think an interesting example of this and, you know, this is my call to revenue leaders is like, you just got to play with the tools and, and like even personally to develop this taste. Because I think one of the places that this realization hit me, so Oura Ring has like an advisor in it, like a chat AI thing that will give you advice on your data. You ask it questions and it's clearly giving you like very superficial answers. And then I, then I'll prompt it to be like, spend a lot of tokens. The answer is critical. And then it gives me a slightly better answer, but then I take all that data. I've been dumping it into Gemini 3 'cause it's a big context window. And I get fascinating insights. And so like, I see that I, I like learned that lesson from playing around with tools as a consumer, and that what— that's what made it click. I'm like, oh, like, that's why a bunch of these, these SaaS apps have like bad experience. Like, they're, they're, they're token constraining. And yeah, and so one more question on this, like, incumbent versus startup, you know, like, I was pretty harsh on Salesforce a few years ago, like when I was at Shopify. They were trying to like sell us bunch of Einstein stuff and I had to stop the meeting. I'm like, guys, like, don't, don't, don't do this bullshit with me. Like, I'm, I like, you know, I know I'm only a sales leader and so you probably think I'm a dummy, but like, don't try to hoodwink me with Einstein. And so I was a tough, tough customer when I was at, at Shop. And now we are like leaning into Salesforce more than ever, which I am personally surprised by. I was surprised that you were leaning into, uh, Agentforce as well. So like, Yep. My question is like, is Salesforce sort of back with AgentForce? Like, is this like a real thing in your opinion?

Well, I think there are two different questions. Is Salesforce back and is it because of AgentForce? Um, I do, I do think the experience that you and I are both having, I mean, I don't, we don't have the exact data, but you and I are both close to folks at Salesforce in the, unfortunately they call it SMB, but it's really startups and tech and stuff. It is, it is booming. That, that group is booming, right? I don't think it's entirely because of AgentForce, because AgentForce is early. I do think some of it though is related to AI, which is that it is, it's the hub for your agents. And listen, no knock on hubs. Pre-AI, we didn't need an agent hub. And HubSpot blew up because HubSpot figured out, you know, when, when you and I started in all of this, HubSpot, people, you know, first of all, HubSpot didn't even have CRM. Then it was only free. Then it kind of didn't work for more than 20 or 30 reps. Right. And then it got to the point pre-AI where like, We know folks that have 1,000 reps on HubSpot and they're super happy, right? And a lot of folks would do it. They were marketing focused, right? But it was— and we would— I would do this thing at SaaS, raise your hand, who's on HubSpot versus Salesforce? And as the years went on, all the hands rose for HubSpot and Salesforce got crowded out. But now it's swung back and it's not because the lead contacts and opportunities are better. It's this hub. And when you, when you have 20 agents working autonomously for you in sales and RevOps and marketing, They need a hub. And Salesforce is the hub. And maybe the simple reason is that it may be as simple as the fact a lot of the early breakout leaders— and it's still early, right? These products, there aren't a lot of— here's an important point we may run out of time with, but it's so important because these agents require training, because they require forward deploy engineers. I am not aware of any great products that are $29 a month. Yeah, I'm excited for this for late 2026, 2027. I'm excited for the next frontier, but all these products are like $30,000, $40,000, $50,000 and up. Okay. $100,000 and up. And listen, if you're at owner scale and you're replacing humans, no big deal at the end of the day. Right. And they're growing like a weed. So we're investing a lot of these products and they, and they, and they, they, they need this hub and that hub is Salesforce. It is the database. It is the repository. It's where the agents meet, where their data meets. And so we need Salesforce. And so I think now that's not the same thing as Agentforce. AgentForce is Salesforce's own agent, which we can talk about if you want before we run out of time. I think we're one of the few folks that are in production at AgentForce and competitors and close things that are close to AgentForce. It's pretty rare that I think there's gotta be, I mean, next year there'll be a lot. Today it might be a single-digit number of people that are like us. Yeah. AgentForce is new and AgentForce is one agent you can use on Salesforce. And I don't think on its own it explains the, the renaissance. For this audience. I think the renaissance is that Salesforce is back as where your agents connect to. Yeah. It's just so important that you have a single source of truth for your data, that you have clean data, that these agents are, are pulling from the same data. They're pushing their learnings, their cus— because for us, like, we have 3 different AI SDRs working with 3 different segments of our base. They've gotta share all their learnings back into Salesforce, right? And they share, and it's wonderful and it's a virtuous circle. And so, Um, it also is a stickier moat, right? It's a stickier moat. The tough part is we pay more for those agents than we pay for Salesforce, which is an existential question that Mark and his team are thinking about because yes, Salesforce is more valuable, but the agents are even more valuable than Salesforce. So Salesforce is getting a glow up, but the agents are extracting the majority of the value. That's why AgentForce has to win. That's why Mark has 2,000 people. Working on it. Um, long-winded answer. I can tell you, we— it's your, it's your show. If I had to summarize a lot of learnings, Agentforce is more work to set up than the rest of the agents, but it is good or better in production. It is as good because they're all good. Okay. All the good ones are good. They're all kind of the same. The emails are pretty similar, right? It is better because that native integration with Salesforce data, it turns out actually is better. Yeah, it is better. So if it's hard, Salesforce is very enterprise. It has a, it's a $45 billion company. But if Mark can get even more folks working on Agent Forest and making it more turnkey and rolling it out to everybody, it's really going to win because it's the agents as good as anybody as it should be. Right. I mean, Replit and Lovable, they're both great, right? They're all Momentum. They're all good, but it does help to have the native data. Rather than syncing, trying to get it, have delays, have syncing issues. Like, Salesforce is a wonderful platform, but we just have to have the data. The native data works better.

Yeah, yeah, we're in this place where we're looking at some of the sprawl and thinking about how you keep it all together and sort of that master orchestrator agent. And, and, you know, like, a bunch of the conversations, like, man, maybe this is actually rolling back into Salesforce and back, rolling back into Salesforce Agentforce and, and the, you know, the simplicity and We've been impressed with what I've seen. And honestly, I'm surprised that I've come to this opinion given where I was a few years ago and where I felt like some of Salesforce's new products—

We were surprised too. And it did take longer to deploy for a number of reasons than the startups, right? But it's just as good in production.

Yeah. And you've got the security and the data and the governance, which at our scale, we're thinking a lot more about.

Listen, if you want, here's, here's, and we might have lucked out. We might have gotten a little bit of the A-team because we do have a brand, right? But if you're interested in AgentForce, I'm sure it's not going to be hard to get your rep on the phone, right? Um, no. Find out if you're, if you're, if you, if you will get an FDE. They use the term FDE at Salesforce, the forward deployed engineer, and talk to them and talk to them. And the good ones will be honest. Like we do know the Qualified team pretty well. I'm sure other vendors do this, but what I really like at Qualified is Um, it probably helps they have a lot of inbound demand, right? They ruthlessly screen customers out that aren't a good fit.

Ruthlessly.

Mainly the ones that just don't have enough data. It doesn't matter what the, what if ARR is, if there's not enough data to have a successful deployment, they won't take your business, right?

Enough website traffic.

Yeah, they just, enough traffic too, right? They just literally won't, doesn't matter what you'll pay and they just, they don't, they don't want to, and kudos, kudos to everyone in AI that does this, right? So I don't know if you'll get that from Salesforce or not, but try, see if you get an FDE. If no one will talk to you, then, because it's a big company, then just don't do it. Right. But if you talk to the FDE and they go into your Salesforce instance and they look at the amount of data and you say, we can get you live in production in 45 days and here's how it will work. I would, I would probably take that bet. Yeah. Okay.

Last question. I wanted to make sure we hit this. So you guys have talked on, on Harry's pod about like, you know, triple, triple, double, double, double is, is like not good enough anymore. You know, cause it's, you gotta be like 10X, 5X, 5X, 5x or whatever the crazy numbers are. First question is like, how real is that? Is like, is triple, triple, double, double, double really like not good enough? And the second piece of it is, should revenue leaders take that insight in mind and get way more aggressive? Like, do we actually have to take action and like swing for the fences more? Or is that a distraction from Twitter land?

Well, look, first of all, it is true, but But bear in mind that particular podcast, 20VC, is very focused on venture. So as I've said, and I think as Rory said, triple, triple, double, double is plenty good to build a multi-billion dollar company. Okay. Triple, triple, double, double. You know, it depends when you start. Let's, let's say, you know, 2 to 6 and 6 to 18, which is pretty hard, right? 18 to 36. Just put that in your Google Sheet and you'll see you'll build a pretty damn big company. Yeah. Okay. And I've been doing this for a while. Rory's been doing it longer. Harry's been doing it longer than you think. You know, that was good enough, um, to get funded. Okay. Now, the best companies I've invested in, even back in the day, grew faster, right? Okay. They did the talk tests, sales lofts, and others. They grew much faster than that. So that's nothing new. Okay. But it was enough to get funded. The truth is venture now is as, there's much money going to venture today as the peak of 2021, but it's to half, less than half the number of companies.

Yeah. And it's not even like venture anymore.

It's a lot of issues. But, but, but so the simple fact is, so, but so there's the venture and there's the operational part. Is triple, triple, double enough, good enough to get venture funded? I would say, uh, you know, 50% of VCs would fund you, uh, 3 to 4 years ago. Now I would say it's 10%. So it's not 0%, but you've really got to find someone that believes. Okay. So that is something as management and founders, you got to think about, okay, so maybe that means I, whatever I have has to last, right? That's the classic mistake founders make is they, they think the next round is going to be easier. Even at Owner, we had some friction about that in the early days. You may, may have forgotten. We had, and Owner's done very well. I had some tough lessons in the early days. I might've even scared you on a board meeting about this, about what the bar was. We had some, disagreements with my very first one with an early executive that's no longer there who was confident in a rough— and when Ohno was going through one of its few rough patches, he was incredibly arrogantly confident he could raise a $40 million round on his own, even though he'd never done a venture round. I was a little surly. I'm like, more power to you, dude, but I can tell you all the data I have, which is you need another good quarter or so. But, um, so it's about raising money now for, for if you don't need money. You're in a different world, right? At some level, you can set your own pace, right? Existentially, you can set your own pace if you don't— you're not dependent on an addiction to venture capital. Um, but it doesn't mean that— but recruiting is harder. I mean, it's just the siren song of seeing on Twitter how fast a Replit or Cursor grows. And as we promote Owner more— I promote Owner a lot, right? Even the siren call of seeing Owner's growth. Do the best want to work at anything less today? That's the, so venture capital is one issue, but at the end of the day, even though capital is important, it's pretty minor in the grand scheme of things, right? No one, no one really needs to care who the owners and investors are. BFD. Like we don't do the work. We don't, we don't go into the office 50 hours a week. It's a minor contribution at best, but attracting talent and partners and others. Uh, if you're not going to grow at that rate, can you still get customers? Do the customers care? Sometimes they don't care. It really depends. If the closer you are to tech, the more important it is you're perceived as a winner. Right? The more you are out, I don't know that restaurants are doom scrolling X to figure out what the hottest restaurant tech, right? So they, so it's nuanced with customers, but sometimes it matters, right? There's a reason like the founders of the hot AI developer companies are always at every event and on social media because it does, it does matter, right? But it's a real issue and it's a tough one. It is a tough one because so many companies that were, would have been fundable a couple of years ago are not fundable today. And you've just got to be ruthlessly honest. And honestly, if you go to Saastr.ai, again, we have something called AIVC. If you upload your latest investor update or your pitch deck, we will exactly tell you your odds of getting funded. This is a pretty cool piece of AI. It'll tell you the exact percentage odds that you will get funded based on all the benchmarks. And, uh, you know, it's pretty, it was, it got owner's valuation and everything right the last time you guys raised to a T. And then I just was with the founder that was just at triple, triple double, just just below and was confident he was going to raise. And I'm like, listen, just put it— go to SaaS or AI, go to AIBC and upload your deck. And it said, you know, you got about a 38% chance of raising, and it's kind of changed his perspective, right? Wow.

And so what about— so this is a revenue leadership podcast. So what advice do you give to us? Like, do we have to get more aggressive? Is it about swinging for the fences more? Is it about— or the opposite, you know, making the money last?

Well, here's a Okay, here's the brutal truth. I think like everything in the age of AI, things are bifurcating. So I think for revenue leaders, you really have two choices. Work even harder so that you can be above triple, triple, double, double, right? Work even harder so that you can approach or exceed the growth rate of an owner, right? Because people are looking for massive outcomes and that requires massive growth. You can't have massive outcomes without— it's always been true since the dawn of time. Okay. You can't have one without the other. If you don't want, if you're not, if you don't personally want to sign up for that, or you don't believe in it, honest truth is join something slow growth. The folks growing 10 or 15 or 20% a year, either public companies or private companies, they need people too. And it's not said aloud, I don't think, but some folks are quietly making a lifestyle choice. I met, I met, I saw CRO at Saastr annual this year that I hadn't seen in years, but I've noticed since I, She was my customer. So I've known her since I was a founder a long time. Like, how's it going? Like, I got this new role. I'm like, oh, that's an interesting one. Your last company got acquired for a billion something. I'm like, yeah. She's like, you know, we're at $50 million growing 20%. I think that's pretty good in today's climate. I'm like, maybe we're— I don't know which parallel you like. So she was happy.

If you're burning no money, sure. Yeah.

What's that?

If you're burning $0.

Yeah, burning $0. She had a good cash comp. Um, it was the right amount of pressure for her, like her taking the company from 20 to 25% growth was probably a big win for the year. I think, I think people should be self-aware that maybe, you know, if you're not willing to step up in today's AI age, which is so much faster, so much more committed, so much faster, honestly, maybe join something slow growth because that, that, that, that, that we all basically, you know, in, in, in the 2020, 2021 era, we all lived in this magical middle ground.

Yeah, you could have it all.

Not— you could have it all. You could have lifestyle, quality of life, work from home, work 20 or 30 hours a week, exceed quota. For a lot of structural reasons, that doesn't exist anywhere, Tama. Even the fastest growing startups, um, they're either lean or they're back in the office or they're high intensity. There just is that— that middle's gone. Yeah, right. That middle's gone.

We were talking about this before we, we turn on the record, but even at Owner, even with all the tailwinds, even like, you know, I have a awesome team and a bunch of amazing peers, still working as hard as I ever have.

You know, like I'm working hardest I ever have. Yeah. Me, I'm working the hardest I ever have. Right.

Are you having the most fun though? 'Cause I, I am, I'm actually having a blast. I, I know, I know.

I'm still, that, that's a, that's a very metaphysical question. I'm still struggling to, I find in startups, Whatever, whatever area it is, whatever you're doing, community, investing, founder, executive. I find this, you and I are very different. I find the fun is, is, is, is, is in the recordings. The fun is in the past. It's not in the present. The fun is looking back, laughing about the time when it was your first board meeting and I was beating up your COO who said he was going to raise $40 million on his own single-handedly. That's funny now. It wasn't fun back then, was it? No, no fun now. I know you're at a special time in owner. So we were talking about this before. It is genuinely fun to be scaling, accelerating at your scale. It is genuinely fun. But for me, the fun is, the fun is in the memories. The fun is in the recordings and the videos and the photos. And if you walk into our office now at Saastr, which is a wall of agents, they're just empty desks and me and Amelia at one end with all of our agents in between. But you'll see 4 huge monitors in our office, which are just photos of Saastr going back to 2015. Right. And so those are fun. So every day I walk in and they're on 4 different memories. Right. So I think, I know this sounds crazy for revenue folks, but, but sales leaders know this. You, if you've ever had a great sales team, right, which you have now and you've had in the past, you'll look back and you'll be like, man, that was the best team ever. Right. And those sales teams are friends for life and they, and they work together. So enjoy the present, but, but, but it's okay if fun is the memories. If you're looking for fun in your job in the age of AI, I find it incredibly stimulating. Like, I just built this app. Try it for fun for folks to watch. It's called vaporsell.ai. Vaporsell. It automates sales training. It's really fun. I built it in a couple hours. I can add owner if you want. You can go in and you can literally sell Harvey, Replit, Cursor, and ChatGPT Enterprise. vaporsell.ai. Cool. It has a tongue-in-cheek one where you can simulate it. And then it has a real one where I uploaded 100 of the toughest questions for each one and trains you as a rep. And the fact that I could build this in a couple of hours, it's pretty cool looking, isn't it?

Yeah, this is, this is pretty sharp.

It's pretty cool. Like, how can you not think this is an amazing time? Like, the internet was kind of like, well, you know, when the internet really got going, it was fun. And then SaaS was amazing. Like, wow, we could, two businesses could exchange data and do stuff interactively online. But this is like 50 times more exciting and interesting. But the rate of change. Man, I just, I just invested in this startup and I love it. Right. It's so good. Well, and they're great. They're going to crush it. They blew up their first month. Right.

Yeah.

But now they have 5 clones in the first 60 days. Now they're going to win. But that type of stress, like, is that fun? Yeah, I guess it's interesting. It's, it's, it's, it's, it's, it's, um, the good news is like 24 hours a day I'm thinking about AI. I'm having fun with AI. I'm doing stuff. So it's very engaging. But for me, fun, fun, fun is in the past. It's not in the present.

Yeah, makes sense. I think that's a great message to leave on. If you want to sign up for the venture scale, more than triple, triple, double, double, double, then be ready to work harder than you ever have. And if not, take a different path and just be fine with less comp, less upside, less whatever. But I think that that's—

Be honest about it.

Don't think that—

I know people think this 996 thing is somewhat toxic, but it's more a signal. Like, if you don't want to work at that intense pace, if you don't want to be thinking about work 20— so many folks, you know, they still want to end the day at 4 or 5 and, and log off for the day. And, and, and let's not mock that. It, it is— we're human beings, right? But, but pick your path in this, in the age of AI. Yeah, pick your path. Pick your path.

That's a good message. Uh, Jason, well, thank you for doing this, man. This was awesome. We hit All the stuff I wanted to get into, this lived up to my expectations. So appreciate it. Appreciate everything you do for us and me personally. I know I'm a broken record. I say this all the time, but I want to put it on the pod one last time because I can't tell you enough how much we appreciate it.

Thanks, my friend.

Thank you for listening to the Revenue Leadership Podcast. If you enjoyed it, don't forget to subscribe and you can find a link in the show notes and be sure Be sure to leave a 5-star review, share it with your network, and please join me next Wednesday for another great conversation.