---
title: "AI Changed GTM Forever. Most Companies Haven't Caught Up."
episode: 65
podcast: "The LeanScale Podcast"
publisher: "LeanScale"
guest: "Alex Shartsis"
guest_title: "Founder & CEO, Skyp"
date_published: 2026-05-15
date_modified: 2026-07-22
duration: 01:14:11
word_count: 14505
topics: ["ai-in-gtm", "gtm-strategy", "outbound-sales", "demand-generation", "brand-positioning"]
canonical_url: https://leanscale-knowledge-hub.netlify.app/podcast/alex-shartsis-ai-changed-gtm/
source: "LeanScale Podcast Knowledge Hub — https://leanscale-knowledge-hub.netlify.app"
license: "Free to quote and cite with attribution to The LeanScale Podcast."
---

# AI Changed GTM Forever. Most Companies Haven't Caught Up. — Full Transcript

> Episode 65 of The LeanScale Podcast, with Alex Shartsis.
> Published May 15, 2026 · 01:14:11 · 14,505 words.
> Machine-transcribed and **not diarized** — speaker attribution is inferred, so verify
> attribution against the audio before quoting a specific person.
> Structured breakdown: https://leanscale-knowledge-hub.netlify.app/podcast/alex-shartsis-ai-changed-gtm/

## 00:00 — Meet Alex Shartsis and Skyp

**[0:00]** (logo whooshing) Today, I'm joined by Alex Chartzis, founder and CEO of Skip, an email platform combining software simplicity with done-for-you service. Alex was employee number five at Drawbridge, helping build it from a cocktail napkin to over 10 million in revenue in just one year before LinkedIn acquired it. He founded an AI company back in 2013 before AI was cool, raising 6.5 million in venture funding, and he's been on both sides of the table as an operator, advisor, and investor, and now he's bootstrapping Skip with a refreshingly contrarian take on why speed beats perfection and go-to-market. Alex, stoked to have you here today.

## 00:41 — How AI changed the speed of building

**[0:41]** I think we have a lot of good stuff to dive into. I wanna start with the perspective that you have on speed, the importance of it and startups, the importance of it in today's environment, and your opinion on why you think some companies are really, really slowed down. - Well, I'm super excited to be here. I think it's just a bunch of fun topics, and I love the podcasts and love the show. So AI has changed speed, right? I think that it's completely different than it was. It's almost like going back in time to the '80s or '90s where a team of one or two developers could build a world-beating product, a world-class product.

**[1:23]** Now one or two developers can build a world-class product in six weeks, eight weeks. And so if you look at Mike Olsen who started Caudera, who originally wrote a database for chips, for embedded database, he built that into a nice business, sold it to Oracle. It was him and two other guys back in the late '90s. You didn't, it took him six months. You don't hear about that. You didn't hear about that for years. It was raise a hundred million from Sequoia, raise a $5 million seed round, hire 15 people. And so I think the biggest thing about speed is AI enables just so much faster product development and so much faster iteration across the entire organization.

## 02:07 — Smaller teams, faster iteration, cheaper failure

**[2:07]** And part of that is AI. Part of that's just having smaller teams. Like with a smaller team, you have less communication overhead. You have less meetings about meetings. You have less meetings, period, right? And you end up, you end up just getting to do stuff. So to me, like the current moment we're living in is just so exciting because you can have an idea and you can have a prototype for that idea later that same day, right? Maybe it's not a good idea, but you can put it in front of customers and figure it out. I think one of the things I've noticed about my own work habits is that when you ask an AI to do something,

**[2:42]** a lot of the times it just runs off and does it, right? And I don't know if you've had that experience, but sometimes it's frustrating 'cause it isn't what you wanted it to do or it didn't do it quite perfectly and you needed to give it more instruction. But think about an employee, right? Back in the day where you're like, "Hey Bob, I need you to go put a list together "of a thousand CTOs of Series B companies "on the Pacific Northwest," right? Bob doesn't take 18 seconds to do that and come up with an 80% okay list, right? Bob goes off for a couple of days and does whatever he does and comes back with a list that's, you know, maybe it's 80% good, right?

**[3:21]** But now, because it's just the iteration is so fast, I catch myself getting angry at the LLM for doing a bad job so quickly. It's like, this sucks. You wasted 30 seconds of everybody's life. Nevermind that I was working on some other window and some other LLM during 28 of those seconds, but like, you know, you end up realizing just how much faster you can iterate. The cost of being wrong is so much lower now, right? You know, Bob costs you hundreds of dollars a day or thousands of dollars a day, right? Two days, like, oh, we've lost real time here. 30 seconds, a minute, it's a lot faster. So, you know, I think it really, the speed thing is so important

**[3:59]** because the companies that are evolving in this AI world are not only shipping products faster, running marketing experiments faster, doing a lot of stuff faster, but they're also getting a lot more comfortable with being wrong and failure and learning from those failures. Whereas, you know, I think if you, just culturally, it really changes your approach. And I think that, you know, that's one of the things we do at Skip where, you know, you can write, we can go generate emails for you. If you don't like them, like, just change them, right? It isn't like you sent a copywriter off for a week to go write a bunch of email templates

## 04:33 — Living in Claude Code: work you'd never attempt before

**[4:33]** and you got them back the next Monday and you didn't like them. Like, 40 seconds maybe have gone by. Like, you know, tell it to do it differently. So I think it's really the sort of change in mindset coupled with the fact that like you can just do, actually do things faster with AI. - Yeah, I think there are a ton of new capabilities and I'm pretty much addicted to cloud code at this point. And I run most of my day through that agent platform versus even living in Chrome or anything else. - Are you using cloud code in cloud or are you using it in terminal somewhere? - VS Code is, it's like one step above terminal. The pro pros use that.

**[5:11]** I'm on maybe like a pro minus, but yeah, mostly operating out of VS Code and just the types of things that I wouldn't even do. So the type of analysis that I wanna do on my company, type of analysis I wanna do in our own go to market or things that we're doing for our customers, I wouldn't even ask the question because I knew, oh, this is gonna take eight hours to do it. I'm the only one who could kind of do it right and I don't have the time to do it. So it really, before having these tools, it's not like it's just doing the things that I was doing before. It unlocks all of these things that I always wish I could do that now I can.

## 05:51 — The Google Ads experiment: compressing failure

**[5:51]** - Yeah, so I mean, I couldn't agree with you more. I think that one of the things that people who haven't played with those tools don't get is, yeah, there's a learning curve, but it's not a big one. Like it's, you can get value out of them incredibly quickly and there's a bunch of tutorials and stuff online. I've written articles. I'm sure the other guests have talked about how to use them, but I think you just have to try it. But the unlock to me is that you lose before if you were building even like an internal tool, right? Like I built, we tried Google ads in December and January. I've been pretty public about it.

**[6:23]** We're finally getting returns from them. But I built a little tool in Cloud Code. I use Cloud Code in cursors, VS Code. So it's basically the same setup as you, but I can always also have cursors do stuff at the same time. And in doing that, like, you know, it's, it's, I just, you know, I'm like, great, generate some keywords. Are those keywords on any good? Do it again, like, okay, generate the upload file so that I can just drop it into Google, right? And so, you know, all this stuff is, you know, would have taken, I would have paid an agency five grand a month and it would have taken them six weeks

**[6:57]** to get it off the ground and it probably would have failed really hard for the first month. Instead, I compressed all that failure into two weeks, right? Like I did it myself, I failed after a week. You know, I failed again the next week. And then by the week three, it was kind of working. And, you know, we're at week six or seven now and it's actually working. But, you know, you just, you get rid of that communication chain of like, tell the account manager about your business. And then they tell three analysts or three junior people who are gonna go set it up about the business. And then one of them wasn't paying attention

**[7:28]** and another one was sick and a week goes by, right? It's just like you sit down, have the AI do the thing, review it maybe, and then it's done and you've checked it off. So yeah, I do the same thing, it's phenomenal. - When you're talking about speed of iteration within the context of go to market, why... Two-prong question. Why in go to market is speed even more important than maybe other areas of the business? And two, what are the big things holding companies back? I know adoption of AI can be one of them, but what else on top of that is really, really slowing companies down and preventing them from having that agility that you recommend?

## 08:16 — What holds GTM back: fear of failure and the 18-steps problem

**[8:16]** - Yeah, I mean, so I guess I'll do the first question, the second question first, right? What's holding people back? I think that to me the biggest thing that holds people back in go to market is a fear of failure, right? Like they either don't know how to do it and so they don't wanna try it. Like I was like this with Google Ads. I was like, I have no idea what I'm doing. And it was a friend of mine who's a PM at Google. It's like, your product would be perfect for this. Like you should really try it and figure it out. So there's this sort of knowledge gap, which AI can solve a bunch of, but also experts can solve a bunch of as well.

**[8:53]** Let's say there's this knowledge gap and then there are all these, kind of what you said, right? I wouldn't even been able to do this before 'cause it takes eight hours. My kids went to Montessori preschool a long time ago and the Montessorians would talk about how many steps were in a task and like cutting up an apple isn't like one step, like cut apple. It's take out the tray with the knife on it, go get the apple, cut it up and then put the stuff in the garbage, go wash the knife. Like it's like 18 steps and to them, the development of kids is about like, how well do they handle doing all of those steps and do they do it consistently?

**[9:31]** And I feel like a lot of work is like that too. It's like, well, starting Google ads isn't like starting Google ads, right? It's like 70 steps and so previously you'd be like, oh, if I do this, I gotta hire a guy or I gotta, you know, it's like all the steps are what's preventing you from doing it. You don't know where to start. Once you've started, you don't know what the next step is. And I think, you know, and there's so much content out there now, some of it's good, some of it's bad. And so it's sort of like, did I watch a good YouTube on Google ads? Am I doing it right? Or did I watch like the charlatan snake oil YouTube

**[10:08]** and I'm setting myself up for failure 'cause just 'cause it's on the internet and has a hundred thousand views doesn't make it like the right way to do it. And so I think what, you know, what really holds people back more than anything is like not knowing where to start, not knowing how to do it. And I think the unlock for that, depending on what you're doing is either hiring an expert, whether that's an advisor, like you, because of AI, you can do a lot of the work. Like with Google ads, I have an advisor who's been doing Google ads for decades, right? And she's telling me what to do, but I'm doing it. So, you know, I'm learning and understanding it.

**[10:43]** I'm setting up the tooling so that I can do it the way I like doing things with, you know, an AI first approach. But at the same time, you know, like I'm not, I was failing for the first week and when I talked to her about it, I was like, "Oh, this, you know things that I don't know." And I don't even know that I don't know them. Like I'm, you know, and you know, it's true of a lot of go-to-market channels, whether it's LinkedIn posting, whether it's other socials, whether it's influencer marketing, whether it's obviously cold email, there's a ton of stuff people don't know they don't know. And so, so that's kind of that sort of second thing,

## 11:17 — Urgency and the nine-month sales-cycle math

**[11:17]** like what stops people. I think the speed question in GTM is so important because the market is changing so quickly. And I think people don't really think it through. So I was talking to a friend who just raised the Series B, you know, 33 million bucks in the bank all of a sudden, their company's obviously doing it pretty well, right? And he's like, "Look, we, our goal is to hit, is to close five big, you know, retailer logos by the end of the year," right? And I was like, "So you better get moving on that now." He's like, "I know, like we gotta, like, that's a nine-month sales cycle." So he's fully aware that if the goal is like

**[12:00]** five big retail logos and the sales cycle is a nine-month sales cycle, he's got like two months to figure that out and get those leads through the door and turn them into opportunities and start working those, right? And not five, probably like 20, right? To actually close five by the end of the year. And so a lot of people just don't get that urgency or they get it too late. You know, it's like November and you're like, "Oh, I got this goal. I got to close five of these people. I need to send some cold emails." And it's like, it's too late, you know? Like those guys are, they're planning their Christmas holiday

**[12:33]** and they're 2027 right now, like you missed it, right? And so I think that urgency, you know, if you act with that urgency all the time across, especially across everything, go to market, not just like the foresight of like nine-month sales cycle that takes us long, but more like, "Hey, what can we do this week that'll have an impact?" Whatever the sales cycle of your business is. You would just, you would try a bunch more stuff. You'll find the stuff that works faster and then you'll optimize it faster and you'll see sort of exponentially better returns versus not knowing where to start, having a bunch of meetings about it,

## 13:07 — The growth model and two kinds of founders

**[13:07]** not being able to just take a risk and decide, so. - Yeah, and I think knowledge gaps, if you don't even know what you don't know, you have all those knowledge gaps, it's hard to even begin going in the right direction, which can slow you down or you end up going in the wrong direction. And then also having this sense of urgency and understanding the timing of how long things take. We see it all the time. One of the biggest projects we do that lean scale for our customers, and most of our customers are series A to series C companies. So they raised a fresh round of capital. They're usually gonna go put that money into scale and go to market.

**[13:43]** Where we normally start is what we call a growth model and saying, "Okay, great. "You need to double or triple your ARR target right now. "Let's reverse engineer that. "Let's go look at the sales cycle. "Let's go look at how long it takes to ramp a rep. "Let's go look at how long it takes to do those things. "Okay, you need to create this much pipeline here "and then start working through that engine." And having that visibility unlocks that sense of urgency as like, "Wow, we need to get moving quick." Where they had no idea that they needed to get started that early, so. - Yeah, I think there are two kinds of founders.

**[14:14]** I would put them in the rookie bucket that raise their series A and are super excited about it and don't realize the crushing responsibility they've just put on their shoulders. And it takes them six months before they're like, "Oh, I've got to change how I run my company." And they're already behind and they're not gonna catch up, or some do, but few. And then the more seasoned ones who would do what you do, or have good advice and listen to it, and are like, "Oh, we've raised this thing. "We've got to scale up. "We've got to shift gear and do things very differently." And I've seen both and one, if you're on the venture flywheel,

**[14:47]** if you have to hit venture goals, the sooner you tool your company for that. And once you have the resources, right? Don't do it early and waste all your money on perfect rev ops when you don't have any customers. You've got to find product market fit first. But once you do that and you close your A or your B, you've got to make your company more efficient and start managing it more to a growth plan. Or you're not gonna get there. Maybe you get lucky, but that's not a, lucky isn't a winning strategy. - Totally, totally. I think something I'm really curious about, you've had a really, really strong career arc. You've been at some really strong logos.

## 15:26 — What used to work that doesn't anymore

**[15:26]** You've been in a number of leadership positions. You've kind of been outside companies advising, deep in the weeds, operating within companies. As you go through your career arc, what are some things that used to work that simply don't work anymore? And is there anything that you think people are still trying to do, or they have this idea of what sales and go to market is, that has just left in the past? - That's a great question. There's so much that doesn't work anymore. (laughing) - It's a long list. - Not that it's business related, but I took AP computer science in high school and learned a dead language. It was dead by the time I got to college.

**[16:08]** So Pascal, that was popular for a brief moment in the '90s and then it went away. So there's so much that changes, I think go to market's even worse. I think the fundamentals have not changed. Know your customer, focus on your customer journey. In sales, know what a logical next step is. There's a lot of stuff that are just fundamentals that are, I think, never gonna change at this point. But the tactics have changed a lot. Back in 2015, 2016, there was a conference we went to that we found out about in an industry we weren't in two weeks before it. We found out 'cause we sent cold emails, we were testing different markets. And my SDR was like, hey boss,

**[16:49]** a bunch of people are going to this conference. Are you gonna go? And I was like, I don't know, give me a few meetings. If you give me five meetings, I'll fly out there and buy a ticket, right? - Sure. - We got 18 meetings in two weeks in an industry nobody had ever heard of this in. I mean, we had some things going for us, but like you can't do that with cold email from a standing start anymore. Like you need to build the infrastructure, you need to test a bunch of stuff like that. That like have one guy send some emails from his Gmail account and get 18 meetings. Like that's no longer a thing. So, I think there's stuff like that.

## 17:24 — Influencer marketing: B2C vs. B2B

**[17:24]** There's a lot of the what kind of gray area, Zynga style, feed polluting viral stuff that no longer works. LinkedIn used to suck in every contact from your phone and send them a message. Hey, Alex is on LinkedIn, you should join too, right? I mean, like thousands of people at a time. That stuff works and it doesn't work anymore. So, the targets evolving, things have changed. I mean, but now there's whole channels like influencers that didn't exist before, right? That work incredibly well. So, it's in a lot more organic advertising. I mean, back then LinkedIn didn't have ads. So, if you wanted to reach people on LinkedIn,

**[18:06]** like you couldn't, it wasn't a thing. So, as channels kind of closed out and new ones open up and that's part of why being quick to experiment is important. - Yeah, one thing I'm hoping we could dive deeper and I'm genuinely curious. Influencer marketing has worked really, really well in B2C. Have you seen any really good successful B2B influencer plays or any examples where like, hey, this company really pulled it off or this is a good influencer in B2B? Have you seen it done well on the B2B side? - Yeah, I mean, there's a ton of examples of influencers. I don't know if they were an influencer before they started the company

**[18:43]** or they started a company and became an influencer, but there's a lot of examples like Adam Robinson who has very loud $20 million business, right? It's a hundred percent driven. Well, it's largely driven by his LinkedIn presence, but they also send 150,000 emails a month. So go figure, right? So, one wouldn't work without the other. Well, the LinkedIn might work, but you're not going to capitalize it without sending the emails. The emails certainly wouldn't work if he wasn't in everybody's LinkedIn feed. But that's one where I would say like the personal brand and that influencer kind of status is what makes that company work.

**[19:21]** I'm seeing more, I think the influencer marketing, so there's a different kind of influencer, right? Like there's the business influencer is also the same as it's ever been, a VC that sits on eight boards, an advisor that advises top companies, right? Those people, the banks and even the VC, like people get brought in through relationships, which is just like non-digital, low-scale influencer marketing, right? I have a relationship with, I'm a banker at Silicon Valley Bank, I have a relationship with a bunch of partners at Kleiner Perkins or at Andreessen. They do a deal, they bring me in to do the debt part of it. Like that's, it's relationship-based,

**[20:04]** but it's a little bit, to me it's sort of in the same vein. I think that when you deal with enterprise level, well, so two things. When you deal with like the mass market, so like the Adam Robinson, 200 bucks a month, right? For everybody, right? Then it works pretty well. And when you're dealing with enterprise level stuff, being somebody that, or having a relationship with somebody that everybody's heard of is also helpful, right? Like it just, it gives you legitimacy much faster than you can get legitimacy by being written up by the Wall Street Journal or Bloomberg or something, right? You probably end up in both places,

**[20:41]** but if you're working with a prominent influencer and you're sponsoring posts or you're getting mentioned or that sort of thing. So I know that Apollo does 'cause I, their event, the sort of lead database, they have, they work with influencers, they add them at their conference posting and stuff. I don't know what the ROI was on that, but I think by the time you put on a conference for 40 or 50,000 people, it's probably a drop in the bucket. So why not? - It's probably worth it. - But the ones that have been the most successful are the people that are kind of building their personal brand and that is their go-to market.

**[21:20]** - Yeah, I just think it's completely taken off in like CPG and like real consumer product type of industries where it's like, that's the main channel. You build a new water bottle or something, like that's where you send it. You go find the influencers, they have the following, they talk about how much they love using your product and then that's how you're getting sales. And I've seen, yes, the thought leadership, founder doing their own influencing for their own company. I just haven't seen a lot of like, oh, this is a famous person who's using this B2B product and then like they're getting leads that way.

**[21:58]** And maybe there's something, now what you're saying is, hey, investors and connections, getting your foot in the door that way still feels a little bit different. They're not gonna sell their time for that, but it still feels like maybe there's something structural where it's not gonna work as well in B2B in the same way it does in consumer because of XYZ. - Yeah, it's a great point. I think that the relationship thing is slightly different, although some of those people have huge followings, right? Like if-- - Oh, sure. - I mean, some of the VCs have, you know, like Tom Ash Tungus has like 100,000 or half a million,

**[22:34]** some huge number of those-- - I just don't know, would they sell a spot on a podcast where they're like, oh, and by the way, I'm using Apollo. - Yeah, I was thinking more organic than like paid influencer. I guess that's a good, no, it's a good point. But I, you know, I do think that like, I'm seeing a lot of the payments, like there's a payments company that's sponsoring Kyle Poyar's "Growth Unhinged." And so, you know, they're in a bunch of places. I've seen them also pop up on a different newsletter, you know, as a sponsor or a guest post. So I do think that there's more people trying that. I think the reason why it works so well in consumer is,

**[23:15]** I mean, maybe it's just 'cause there are more of them, right? Like you don't have to go, creating a million YouTube following is hard, right? But as a brand, like you don't have to do that. You just go pay the guy who has the million followers, right, maybe you hire them full time, but either way, like, you know, you don't have to worry about solving that problem. You just go to the person and pay them. I don't know if there are B2B people with those kinds of followers. I mean, you know, part of it's just that like, 'cause you're selling a $10 water bottle or a $30 water bottle, right? Anybody can buy that. Like you're selling a half million dollar CRM.

**[23:51]** Like, I don't know. Air's a little thinner. - Yeah, that's true, that's true. You're not gonna reach an audience that's big enough through that type of channel where, you know, it's better to just get in probably through relationships or other more tight circles. - I mean, I think that you can create those, like, as a company, if the question is like, if I'm a company, how do I think about influencers as a channel? Like, I think about it all the time. I mean, our basic two motions right now are Google, 'cause that's, you know, Google paid, 'cause that seems to be working. And then, you know, podcasts and influencer type stuff like this, where, you know,

**[24:28]** it's really just having conversations about our not sharing knowledge. And, you know, then if somebody's like, oh, I need to do some emailing, that Alex guy seems to know what he's doing. I should give him a call, or I should look up his company, right? And so it isn't the same as like, oh, I wanna be like, you know, a Kardashian, so I'm gonna buy this water bottle. Maybe there's a little of that, I don't know, but. - No, maybe not, maybe not. I think most B2B solutions are pretty, like, functional or economic, not necessarily, like, status driven and things like that. So I think there's probably just different buying incentives

**[25:04]** that are conducive to either one. But I just always, I'm just kinda looking at that and I'm wondering, like, are we just slow to adopt it or are there some structural reasons? - Well, we're definitely slow. I mean, B2B is always a decade behind consumer, right? So, I mean, there was direct mail-in consumer long before there was direct mail-in B2B. And there was, you know, TV ads for consumer long before TV ads for B2B. I mean, I think it'll eventually catch up. It just takes time. - Right, great.

## 25:35 — The ideal GTM tech stack: it depends

**[25:35]** Shifting a little bit, I know as you've gone through your career, I'm sure GTM Tech Stack has looked a little bit different at each stage. And I'm curious, I think there's a whole, seems like there's a whole lot of consolidation going on in the market right now. And at the same time, a lot of solutions popping up too. So there's a lot of point solutions, but then a lot of companies that have been out there for a while are really gobbling up smaller companies and putting together comprehensive tech stacks. I'm curious your take on what an ideal tech stack looks like today and maybe peek into the future a little bit of where you think it's going.

**[26:14]** - Yeah, well, I think the tech stack really depends on the company, right? So there isn't like, oh, this is the way, like that you have to do it this way and like only use HubSpot for everything, right? Like that's, to me, like there's, it depends on the nature of your company and it comes down to what meets your customer, your prospect, your customer journey needs the most and what fits best with your culture. So as an example, for some companies but fits best with your culture, I really mean like helps you move fast as fast as you possibly can. And so for some companies, like putting everything on HubSpot is great 'cause it solves regulatory problems.

**[26:50]** It solves compliance problems. It just, everything's in one place. All the emails are in one place. Like we don't have to tie systems together and marketing can do stuff and sales can do stuff. And it's just, you're gonna move faster even though HubSpot as a platform is it may be the best point solution for anything it does. The speed you get from not having to coordinate a bunch of different departments and people and IT just setting up a landing page in HubSpot 'cause it's set up, right? To me, that's valuable for some companies. For other companies that where they wanna really optimize people, it's a little bit more vulcanized

**[27:28]** and people wanna optimize their individual activity and they can handle the complexity. Choosing the best point solution is a great idea. Maybe you do clay for enrichment and you use skip or something else for email delivery and you use pipe drive or who knows, right? Adio or some other CRM that isn't like as popular as a HubSpot. So it really, it depends on where you code your own. It really depends on like what you're trying to accomplish. And how your company works, right? If you have a bunch of people that are comfortable with NADN or which is like a sort of a middleware kind of AI company, right?

**[28:08]** If you have comfortable people that are comfortable with that and you as a leader are comfortable with that, great. Like, give people the power to do that, set the guidelines of like, hey, everything needs to end up in our CRM or in some other database, but do whatever you want. To me, I think that they're, it's interesting. I mean, Clay has obviously emerged as a clear winner in that space. And yet just yesterday, I forget the name of the company, but somebody's like, oh, this is like Clay, but it's better and cheaper. You know what I mean? There's still gonna be point solutions for your use case that are actually better than the quote unquote

## 28:42 — Build vs. buy: Klarna and the eroding CRM moat

**[28:42]** category leader for everything, right? Depending on what exactly you're trying to accomplish. So again, it really depends on the company. But in terms of where things are going, I think that like, for companies that reach scale and know what they're doing, I think a lot of people are gonna do it, what Klarna did and build some of their own tools. Like I don't, I've never understood CRMs. CRMs are just like the best. It's like a company that's built on the best salespeople in the world selling a giant relational database that you could get for free. Like it's, you know, with vibe coding, with just AI enabled coding, like there's nothing that Salesforce does

**[29:24]** that is special in any way or HubSpot, right? It's just a relational database with a UI on top of it. And you could go to Bolt or Lovable or Replit and build that in like 10 minutes, right? Now there's some workflows that they do and there's some stuff that they do, but Salesforce in particular, like to do anything, you're gonna go pay a third party tool to do it. Like you might as well just get the third party tool and set up a SQL database or something. Like you don't, you know, you don't need that. And you certainly don't need to spend six figures or seven figures a year on it, but their salespeople are incredibly good.

**[29:58]** I think that, I think that, you know, if you look at, you know, AI native companies, I think that a lot of the services that they used to buy off the shelf, they might buy for a few years, but the tipping point between buy and build and the customization math changes where it's like, oh, this would take one engineer a month and we'd have it. Why are we paying a million dollars a year when we could just, you know, have one guy work on it for a month? And because of AI tools, anybody can maintain it. That guy doesn't have to be the one to maintain it. Anybody can load it in a cloud code or cursor, tell it to like read the repo and figure out what does what

**[30:36]** and then make changes. Like you don't, you no longer have this like barrier to, you know, to maintenance that you used to have when humans needed to remember what they were doing when they wrote that object or, you know, did that thing in the code. So I think for the most advanced companies, they're gonna, they're not gonna roll their solution, their own solution for everything, but they're going to do a lot more either with their own workflows using tools like NADM and Zapier, or they're gonna just build their own software. - I'm not as familiar with Klarna. I mean, I know of them as a company, but I didn't know they took a big stance

**[31:14]** to build their own tools. Did they make a huge investment in building their own internal tools? - Yeah, they did. I mean, they've gotten a lot of press over it and it's unclear how much of it is just press and how much of it's real, but I think they ended their Salesforce subscription completely and rebuilt everything. - Wow, and just built their own CRM. - And then they've gone backwards and, you know, readopted a few tools, but I think that, you know, like it depends on the scale you're at, right? If you have a Calendly subscription and you have five users and you're paying 100 bucks a month, like, you know, just keep doing that, right?

**[31:47]** If you have a sales team of 2,000 people and you could just set up a booking link thing on your own, like there's, you know, for what it would cost to use Calendly, like you can, it's a five-day project. Like, you know, you don't, you can just integrate what you need to integrate. You don't need their marketplace of apps and all this other stuff that they built, which is great for the SMB, but I think as you grow or, I mean, we happen to be, you know, 'cause we're bootstrapped, like we buy stuff off the shelf because I think the thing you're buying is that, really what you're buying now is the product manager's knowledge of the problem, right?

**[32:24]** Like, think about a CRM. The one thing about a CRM is everybody goes into the CRM to do something different. And a Salesforce or a HubSpot or an Addio or, you know, the new AI-based ones like Day and Lightfield, like they've spent a lot of time thinking about what people do with CRMs. And so they've got those workflows dialed. They've got the, you know, first you do this, then you do this, then you do this. This is what you need to integrate. Here's the card rails for it. Like they, that's the hard part, but if you know what you need to get done, it's almost half the time easier just to build your own thing. So yeah, I don't, I mean,

## 33:02 — Vibe-coding Skyp; get a customer first

**[33:02]** I realize I'm a little out there on this. Like I'm probably, you know, more graduate level than undergrad 101, but I just don't, so I mean, a concrete example. So we, when we launched Skip, we actually had a customer before he had anything else. And I was messing around with five coding tools and I built a landing page in Bolt. And it was pretty nice. Like it was like, wow, this is pretty good. And it hallucinated some references. Like there were customer quotes and we didn't have customers, but there were pictures and names and quotes. I mean, there were issues with it. It's very hard to remove those it turned out, but they, you know, you had like instantly

**[33:42]** what you would have paid. I mean, my advising site before cost three grand, which was a pretty good deal. But like I had a designer and a couple of developers do it in Webflow and, you know, all in, it was, you know, 40 bucks a month for Webflow. And maybe I'm deleting myself, maybe it was closer to five or 10 grand, but it was like real money, right? It took a month, right? This took like literally 30 seconds and it was good enough to click print on. And, you know, and like that's pretty amazing. So fast forward to, you know, partnering up with my partner on Skip and, you know, he's CTO. I'm kind of the sales guy, CEO type.

**[34:19]** He, you know, at some point he's like, like this is kind of brittle. Like why don't we put it on Framer or Webflow? And I was like, trust me, like this is better. Like I don't, I haven't used Framer. I've used Webflow. I don't, you know, it's like you're fighting templates all the time to do anything takes like real effort. You really have to sit down and think it through. Whereas with cursor, I moved to cursor and you know, where I use a lot of cloud code as well. Now it's just like, hey, make SEO landing pages go. I'm like, boom, 50 landing pages for SEO done, right? And are they perfect? No. Are they 90% as good as what I would have done on my own? Yes.

**[34:58]** Did it take five minutes? Maybe less. Like it's like, how do you compete with that? And so versus like the Webflow method of like building a template, like fighting the bugs, trying to get the template to do what you wanted it to do exactly and then writing the content and then adding the content to a bunch of templates. It's like that took time, five minutes is better, right? And so now, so he went in and refactored it at one point 'cause it was bothering him. Like the fonts weren't perfect and you know, the things that you would get out of a templatized tool hadn't been properly done. And so he went in and he, you know, I mean, he didn't do it himself.

**[35:32]** He told Claude to go componentize it and you know, clean up the fonts. I mean, it isn't, it took him a few hours, right? And our website now, I mean, it's got a hundred pages. It's pretty comprehensive at this point. And now he's like, I get it now. Like this is unquestionably better because if you want to do some random thing, you can just do it, right? Like we don't need to get on the phone with support and find out if this supports this kind. It is like a React website that does anything you can do in JavaScript or TypeScript. Like anything at all. And so it just makes us, it helps us move faster. I mean, that's, that's what it comes down to

**[36:13]** and be more flexible and do more. So, you know, previously we'd be spending probably our traffic would probably have us in like the hundred to $200 range on Webflow a month. I don't know if framers may be more, right? And we pay Amazon something $4, $2, like not much. So. - Yeah, no, I think it unlocks and also prevents you from going down a road that you don't really want to go down. I love what you said that you got a customer before he did anything else. I had the same philosophy when starting LeanScale. I started LeanScale in 2021. I already had two kids. I needed to make sure like, hey, if I'm going to jump into something, it needs to make money.

**[36:52]** So let me see if anybody even wants to buy this service that I'm thinking about putting together. So first thing I did was just like built the concept of the company, put a deck together, presented it to a few people. I was like, hey, what do you think about this? And then I had someone say, hey, if you put that together, I will buy that tomorrow. I was like, okay, great. We're in business, let's go. But that experience, let's say it was a bad idea. If nobody, okay, then great. Good thing I didn't go build an entire company around it and quit my job and something that's not actually going to work. But now you can test those things out with MVPs

**[37:29]** and messaging and making sure that it's actually something you should start investing real time into so much earlier. I think it can help you avoid so many mistakes. - Yeah, I think that's also true with actual modifications to your product, right? Like why would you ship a new feature if you hadn't put it in front of users and seen how they reacted to it? Like if you could, right? Before it's like, well, it's going to take us six months to build this thing. So we're going to build it, we're going to ship it. We'll do some interviews first. But now it's like our first version, my thesis was kind of like you, my thesis is like, we wanted customers.

**[38:06]** I wanted, I've started companies that were solutions in search of problems and that wasn't fun at all. And so, and it didn't work out very well for anybody. And so I was like, well, like, let me, my hypothesis was that people wanted, they wanted it done for them, but they wanted to think it was software. Like nobody actually wants to send emails for themselves. Very few people do. They just want the emails to be sent. And so, and they want to just know the emails are good. Like that's the main blocker is that like, I'm only going to, you know, I don't trust the AI to write the emails until I see the emails and can trust the verify, right?

**[38:43]** And so I vibe coded like the first part of it, where it's like, oh, you can give it your company information, it will generate, you give it a couple of people, it will generate emails. And it was all just basically front end only so people could see that it would work. And then if you signed up, we would go do all that in our systems on the backend that were completely detached from what that demo looked like the first demo. And we signed five customers because they were like, oh, these emails are good. Like, let's go, this is great, right? And then they never, and, you know, nobody ever asked for reporting.

**[39:16]** They weren't like, let me see the reporting dashboard. They'd asked for reporting after the Stripe, you know, after like a month or two, they'd be like, oh, do you have reporting? And it's like, no reporting, like we'll email you some reporting, right? But we didn't build that 'cause it wasn't actually necessary if we have it now, but at the beginning, that wasn't what people bought. And I've worked with so many founders and then the founder that's like, oh, we need the dashboards to be great. We need to invest in all this other stuff. And it's like, really, like what does your customer need to know before they buy, right?

**[39:46]** And then you can address all that other stuff later or maybe not address it at all. And so I think kind of like your approach, right? Like, let me make sure that somebody really wants to buy this. Not just like says they would buy it maybe one day, but like actually gets out their credit card and buys it. - Right. - That's the key difference. And with, I mean, it isn't just the AI coding stuff, but with the AI coding stuff, it's much easier to do that and kind of capture your vision and like help the, you know, it isn't like a clickable wire frame which doesn't really feel real, you know, it's a product. - I'm curious your perspective on something.

## 40:19 — From fear to an abundance mindset about AI

**[40:19]** I think I've gone through, I don't know if it's the adoption curve or the dunning effect. Like I'm not sure which curve, but basically scared of AI, hyped about AI, thinking AI is the worst thing ever. And now I feel like I have a nice sober relationship with AI. And I think one of the things that is an initial gut reaction is like, oh, it's gonna kind of take away these tasks, take away these things, take away my identity of what I do and how I add value to my job, company, society, whatever you wanna call it. But recently I've had a pretty strong abundance mindset with AI because when you're actually in there operating with an agent platform,

**[41:10]** building repos of data to make things better, you start to realize this like rabbit hole of, oh my goodness, there is so much work to be done. And there's so much work that I wasn't even able to do before, but now I can do these things. And it makes me think that there's almost unlimited work to do. I'm curious your perspective, just like for any-- - Yeah, I mean, I'm definitely of the abundance mindset. I've gone through a similar journey. It's sort of like when the wifi goes out on the plane and you're like, come on, like what do you mean? There's no internet I've worked to do. And it's like, you're on a steel tube flying 600 miles

**[41:43]** an hour through the air. Like, you know, people used to die on the same journey and you're complaining that you can't, you know, you can't check off things on your work to do list or watch YouTube or whatever. So, I mean, I do think when I get frustrated at the eye, it's like, oh, wait a minute. Like, but before I would have actually had to learn TypeScript and now I don't even know how to write TypeScript, but I can still make, you know, my website do whatever I want. So there's a bit of that. But I think I completely echo your thoughts. I think that it's, to me, it's like a calculator, right? Like I, you know, I grew up in my eighth grade class,

**[42:17]** they wouldn't let us use calculators, like just as a general rule. And I thought it was stupid. And I was right, because like you go to high school, you know, like calculus would be a lot harder if you had to get out of your advocates to do every calculation, right? Like, it's like, how is that better? Like maybe you learn the numbers a little better 'cause you know how to use an advocate. But, you know, it gives you just such a, you know, now, I mean, from a coding standpoint, like I learned a dead language, right? And eventually I was like, you know, I'm not gonna be a developer. I like people too much. And, you know, there are people that are,

**[42:50]** that don't like people, that don't like showering, that are better at this than I am. Not that that's every developer, but there are some developers that are just way smarter than me and like didn't like people. And so they can just go hang on the computer lab and, you know, and just code and be happier than I ever would be doing that. But I think the bigger issue was that it's like, look, I've already learned a dead language. Like I don't want to have to learn a new language every three years to keep up, right? Which is kind of how it worked. Now you can tell the AI, like write this in Go, or you can ask the AI,

**[43:17]** hey, which language is the most effective for this task? And it could be like, oh, Python or Go or whatever. And C++, like whatever it is and just write it because it's fluent in all those languages. And it lets you step back and abstract the problem and say, okay, well, this is what I want to do. Spend your time thinking about the actual job to be done and less about like which, you know, low level language or, you know, how exactly to accomplish it. I think, so my excitement comes from this ability to stay closer to your customers and their needs and meet those needs faster. So we were interviewing some great people for a developer job that,

**[43:57]** 'cause we have a guy that helps on the sales side. And I was like, well, it's sort of not fair. Like as a small company that, you know, my co-founder is like the only sort of technical, I mean, I'm technical enough so I can cover for him if he's out, but like he has, if he wants to do some mundane task, like he's got to do it. There's nobody to delegate to where, when I want to do a mundane task, I just tell somebody else to do it. Which just didn't seem fair. So we interviewed these UC Berkeley grads who are great, liked them, but he's like, when has a customer waited more than a couple of days for a new feature? And it's like, they haven't.

## 44:33 — The last mile: taste, focus, and finishing

**[44:33]** Like they, you know, like they just, we ship so quickly, customer asks for feature, feature gets launched, right? There's no translation, there's no, so, so to me, that's like this, it goes back to the speed thing and keeping close to customers. Like AI just enables that. I think the risk to it, the thing I get frustrated about is when I have like six cursor windows open and I'm trying to do six things at once and I look up, it's five o'clock and like the kids need to be picked up and I've done zero of them. I've done like 80% of six things, right? And because it's like so easy to start something and so easy to get that hit of like, oh, it's almost done.

**[45:10]** But getting things to like 100% done is still hard. It doesn't matter. It doesn't matter what AI you use. Like the myth of like, oh, I'll just go write a paragraph and a vibe coding tool and like, it'll be done is not a thing. And so I think that something that would have taken six months might take like four days, but you still have to invest the four days in it versus, you know, versus like 20 minutes. So I think the challenge is like, it's like great. I want to build, one thing we're working on now, I was working on before this call was like when we, we've shipped so many new features, our knowledge base is way out of date.

**[45:48]** And so, you know, we use intercom, intercom as an API. So there's no reason why we can't just have an agent look at our code base, see the updates and update the knowledge base, right? And it's, you know, it's easy to like, you can basically just take that sentence, hand it to cursor or quad and it will build something, right? To get to something usable, it's going to take probably five hours, right? Maybe eight. And so I've got to sit down and focus and actually do that. And that's what's so amazing about AI is like, I can do that in a few hours, but I got to do the whole thing. Like I can't just like, - You have to take it that last mile, yeah.

**[46:25]** - Right, you know, you got to actually get it across the finish line. And so I think with AI, it's sort of like, I find myself questioning my life choices. Like, am I working on the right thing? Like I could do anything, you know, is this the best use of my time right now? Or should I be doing something else? And even within the scope of a small company, like there's a lot of things you could be doing. - Oh, sure. - So, you know, I think it goes back to just first principles of prioritizing and following through, but yeah, it can be frustrating at times. - Yeah, I think we're going to have to reconfigure how we work and, but I do agree with you.

**[47:03]** I think taking it that last mile, one, because it takes taste, like that's one thing. - Right. - AI can get, like you said, 99% there, but there's certain things it's like, hey, you need to have good taste about what you're building to make those last mile changes. But I do think it will at least get you starting points, get you in the right direction. And if you have the power of focus, then you can get going. So one of the things I was really excited, I know we talked a little bit about Skip. I'd love if you could walk us through what you've built and how customers are really enjoying the platform. And as you're going in, what motivated you

## 47:45 — What Skyp is and why AI SDRs miss the job

**[47:45]** to focus your work and energy here? As you mentioned, you could work on anything and you have vast experience across a bunch of different industries, company types. Oh, it was like, hey, this is it. This needs to be built and I'm going all in on it. - Yeah, I mean, I'll start there. I think it was something I was gonna add to what you said before, which is like, it's more fun to build stuff where you know what you're trying to build. Like you know the job can be done really cold. So like building this stuff for Google ads, like I hated it. I mean, it was better than doing it by hand. I can't believe people used to click through

**[48:17]** like 50 screens in Google and cut and paste, like it's ridiculous. But building all that stuff, it was like, I don't know if this is the right way to do it. I don't know if like I'm basically doing it wrong and just being more efficient at being wrong. Like that, you know, failing faster. Like that, I guess there's some advantages to failing faster but it was frustrating. In this space, like I've advised founders on Outbound for a decade. Like I've done it for more than a decade. I've built businesses with it and it's changed a lot. But to me, it's like, look, I know this job to be done cold and I'm the person to approach this.

**[48:54]** And the fundamental, what we wanted to capture with Skip was really this kind of the same conversation around vibe coding, but it's more like, you know, if I can throw up a website in five minutes, like why can't I send some emails, right? Like why is it so hard to run a campaign to people that were on a webinar or whatever, where I have all their emails, I have all their names, I have all their comments. Like what makes it so hard that I can't just like follow up with them in three clicks, right? I can build a subscription service in three clicks. Like why can't I send a simple sales follow-up campaign?

**[49:29]** And so the insight we had in some of the stories boring was I had a couple of clients that I advise asked like, hey, I think we should use an AISDR. And I was like, I don't, right? Like if you wanna do something in two months, do the thing that works, don't experiment with the new thing. Like that's just risk. It's like why add risk to the whole equation? But I was like, I'll go look at them for you while we do this the old fashioned way 'cause that way I'm sure it works. And the AISDR thing was really interesting. I noticed like two common things. One, none of them were ever like salespeople before they started AISDRs.

**[50:06]** Like only one of the prominent companies, but they were all like 20 something straight out of college. I have nothing against that like Zuckerberg was too. Like that's, but they never like manage sales teams or run end-to-end sales processes or, you know, one of them who's since closed the company, like he was an options trader at I think Morgan Stanley or Lehman or some sort of big trading firm, right? And it's like, oh, this is a great arbitrage. I'll like buy the leads here. I'll write the emails with chat GPT and we'll make money. And it's like, that isn't the job to be done though. Like that's not how salespeople work.

**[50:38]** Like yes, in theory it is, but in practice that's not it. And so what I realized was there was this gap around like enabling salespeople and enabling sales teams to automate stuff 90% of the way. Like not just like push the button and have it just go find every CTO of a series B company and send them a garbage email mentioning their college mascot and seeing if they want to meet. Like that's, 'cause that's kind of what they do. It's like, oh, hey, it looks like you went to Dartmouth. Go big green. Like, I mean, I got one email that was like, hey, I see you live in Oakland. Go A's. It's like, are you not paying any attention at all?

**[51:17]** Like they left, I'm kind of upset about it. So, you know, it's like, are you using chat GPTs 2023 data? 'Cause like that they were still here then. And so, you know, I mean, it's sort of like, it's just, you know, it's that like simplicity, but power that we were looking to build. So I'll show you where it came out. And you can, I'd love to get your feedback. And if people want to drop comments on this video, any feedback is always welcome. So this is the onboarding. Once you, I'm not gonna show you how to put your email in instead of password, but once you do that, you come to the screen. There's a video here I won't play,

## 51:36 — Skyp product walkthrough

**[51:57]** just to walk you through a campaign in case you're new here and haven't seen this podcast. I thought we could do this with rippling, just as an example, but you put in your company website, you can also upload marketing materials. So this is, if you're pivoting, if your company's website is out of date, or if, you know, if you're maybe launching a company that doesn't have a website yet. And what we do is we use an AI to go out and look at your website, not just the front page, but it goes pretty deep, understands your customer pain points, understands your customer context, like who you're reaching out to.

**[52:31]** And the thing with Skip is it's all about flexibility. So this is a starting point. You can edit all of this, and all of this contributes to better emails. So I'll just put in my own info here.

**[52:49]** But, so, you know, you've set up a profile and that's it. Like you have completed setting up your account. Is there more that you can configure? Absolutely. Do you need to do it? Absolutely not. So from there, you obviously, by your team if you want, but you get to start building your first campaign. So what would be, I mean, let's use, if you were advising, we're playing on, 'cause I know you do a lot of this, like what would be a campaign you'd wanna run? I think a post-event campaign is a pretty good like, tight audience, real specific copy and messaging that you need, so. Yeah, let's see, like you went to Sastre and you're following up with,

**[53:32]** I think Rippling sponsors Sastre, which is why it came to mind. But so, let's say like follow up with people who came to our booth at Sastre. Let's say this is like really long time ago. Like this was the 2025 Sastre. And see if they need support or want to book on you.

**[53:56]** So you can do a lot here. The goals, the way I describe it is like, you're telling an employee essentially what you wanna get done. So if you had a junior salesperson, you wanna tell them who you're reaching out to. You might even say like follow up with CEOs, let's just say, who came to my booth. And then because it's AI, there are no templates. And so basically you come up with the calls to action you want people to do and it merges them into the emails in a somewhat random but logical fashion, which improves deliverability. It makes it seem more natural. There's a bunch of benefits. It also saves you a lot of work.

**[54:34]** Like you don't have to go through and like think like, oh, the first thing we're gonna do is ask for a demo. And then when they don't do that, we're gonna invite them to a webinar. You can just say like, you know, book a demo. I'm just gonna use our demo link for this purposes. And then, you know, find a time to meet, reply to book a meeting. So if you don't put in a link, it'll just know basically to treat that, you know, just to write something that makes sense. So it won't always say this exactly. And then maybe, you know, we'll suggest you watch

**[55:11]** recorded webinar.

**[55:16]** And you can even have spelling mistakes here and it will correct them.

**[55:24]** We check every URL to make sure that you're not sending people a URL that's broken. So you can go on like this. I don't think for any given campaign more than three is a good idea. We'll repeat them if we need to. So contacts, I'll just add myself as an example. I'm actually in this ICP.

**[55:46]** And you can upload lists, you can connect it to CRM. There's a bunch of other ways to get contacts in here. And then you can set up your campaign. So there are a couple of different ways campaigns work. You know, this one is one we're doing like explicitly as a follow-up. So we're gonna put in a list, we're gonna tell it to go. You can also have it like target your website visitors and just everybody who visits a website, if we can figure out their email, add them to this campaign. You can have it drop people in from a webinar automatically. So there's a lot of automation. I do think that one big trend right now that wasn't as easy before,

**[56:21]** that's much easier today is automation where things just, once you come up with your plan, like it just happens. Like you don't have to have a human go click a bunch. Like you don't have to worry, did we forget to do that? You just set it up and skip is part of that. It makes it really easy just to set these things up once and let them run. And that's it. Like that's all it took. It's generating a strategy and then it generates some emails. And you can review the emails, you don't have to, you can edit them. So I mean, these emails, what do you think of the emails? - Well, first I'll comment on the experience while I read through some of these.

**[57:01]** Super simple, easy, gets you the right thing. Like how it's hyperlinking everything in there. I think, yeah, I picked up the subject lines well, they're short and tight.

**[57:16]** Yeah, I think it's great. - Yeah, I mean, that's so, I mean, if you don't like them, you can regenerate them and it'll just basically roll the dice again. You can also control everything, right? You can go back and change your original goal if this came out not what you wanted, but you can also explicitly say, hey, use lowercase only in the subject line or use their first name in the subject line or whatever that tactic is that you wanna try, like we enable you to try it. So it's this combination of like, we've been talking about this for five minutes and you're able to see emails and get a sense for, hey, this is pretty good,

**[57:51]** but also that ability to really make it exactly what your strategy is. So that lesson cost me six and a half million dollars in investor money the first time I built an AI company. And that's pretty much it. I mean, at that point, after this, we'll ask you for a credit card because the internet being what it is, you can't let anybody send emails for free or they will only be spammers. - That's right. - Yeah, that's right. We need at least a little bit friction. - But at some point, we have to make sure that you're a real human with a credit card that isn't stolen and 'cause we've had that happen before. And yeah, it's amazing how they find you, right?

**[58:29]** Like the bad actors are looking for you. So yeah, if you wanna try and spoof people's payment info, go to some other service, don't use my service. (laughing) But no, I mean, I think like the idea is that like, if you were a salesperson or even an executive, right? You can be like, oh, I came back from this conference. Five minutes later, I've done all my follow-ups. And we've made it so that you can edit the ones. So if you have five people, you're like, oh, Anthony, like he and I had a really good conversation at Starbucks. So I'm gonna put in a note about Starbucks, right? You can edit the emails yourself,

**[59:05]** but the reality is you're not gonna follow up four times. Like as a human, that's never, nobody's ever done that. - Right. (laughing) - Like no salesperson is that good in the history of the world. But with Skip, like you can in five minutes, follow up four times with a hundred people and it only took you five minutes and it looks authentic. It looks real, gets delivered. We handle all that. So we also do all the secondary domain set up. Like I don't know how deep into this your audience is, but like to do cold email well today, like you have to get all the infrastructure right or you have no chance and it fails silently.

**[59:38]** So it's not like you get a 404 and it's like, oh, hey, your website, it isn't working. Like it just doesn't work. And the only way you know that is 'cause you never get any meetings. And so we handle all that stuff for you so that it's set up properly and your emails get delivered. So you can just focus on like, who am I emailing and do these messages look the way I want them to look. - It's amazing. I think especially setting up those like micro campaigns and being able to iterate really quickly, it's super important. Where normally you need to go into a full engagement software. You're manually putting in the steps, manually coming up with the copy.

**[1:00:13]** It's not quite right. And it takes a long time. And then if you want to change anything, then it's painful. Gotta go back to each one and basically start from scratch. - Yeah, so we let you, we actually just launched this feature 'cause somebody asked for it and a day later we launched it. So basically like originally the model was like, look, you build a campaign and that's it. Like the train has left the station. You can't touch it until it gets to the next station and like deal with it. But now we built it because it's AI native. So that if you're running the campaign for a week and you're like, oh, like we, whatever.

**[1:00:46]** We got some responses and we want to change it. Like this message isn't hitting the way we thought or something changed. You can pause it, have the AI rewrite everything that isn't sent in like one second, you know, change a prompt, click regenerate and then send it again. And it will regenerate everything that hasn't been sent with your new instructions. So that you can, you know, you can learn from those initial responses, which is how we run campaigns. Like you used to just run a campaign with 50 people, learn something and then put in the thousand, right? Now it's like, you can, I mean, if you want to, you can still do it that way.

## 01:01:19 — Point solutions vs. the Swiss Army knife

**[1:01:19]** You can put in a thousand. Once the first 50 have done something, you'd be like, oh, this isn't working. Let me pause it, change it, relaunch it. And again, like our thing is it should just take five minutes, right? Or you're back to the, like the overwhelming nature of some of these tasks. You're like, I'm not going to go into Gong to set this thing up. Cause it's going to take me half a day and I don't have half a day. And that gets back to something you talked about before where I think that one of the challenges in the go to market space is that everybody's trying to eat the whole space. You've got Gong building email tools.

**[1:01:50]** You've got, you know, you've got Apollo building email sending tools. You've got other people integrating contact databases into their applications, right? It's everybody's trying to do all the things. HubSpot now has one of those, you know, de-anonymization cookies or whatever you want to call it. Like an RB2B or a Clearbit, right? And they bought Clearbit. So I mean, there's just, everybody's trying to do everybody else's task. And I think to me, the thing that matters is like, are there people who designed this thing? Like, is it the most important thing to them? Like, do they really know this purpose or are they just like going for some extra NRR?

**[1:02:28]** Like that's, a lot of these are like, it's like Apollo's email delivery is not very good because they didn't, either there were risks they weren't willing to take or they weren't really that committed to it. They're just like, oh, we'll sell more contacts if it's easier for people to email the contacts. So why don't we add some emailing tools, right? Like, I mean, it makes sense on some level but the quality of the tools varies dramatically based on how core they are to the business that launched them. - I think most of those companies, I mean, the only major IPO that I can think of recently is, HubSpot, Zoom Info, outside of that.

**[1:03:08]** I mean, there's a lot of companies that you think would be marching towards that big exit point but just don't have enough juice to get there. So I think they're doing exactly what you're saying. They're scrambling to get as much error as possible to get to that point where it's IPO-able. And but to your point, are they building the best solution? Is their mission to make the best email sequencing solution in the world or are they just trying to create a conglomerate of GTM capabilities in one thing? - Yeah, I think to your earlier question, I think the big thing that separates winners and losers right now is what you don't do.

**[1:03:46]** And I think if people look at HubSpot are like, oh, I want all the features 'cause we wanna do all of this one day. And my partner is amazing at this where he's like, yeah, the concrete example was we were in a newsletter this morning and we wanted to redesign. I wanted to redesign our onboarding flow, the flow you just saw to optimize something towards the end. And I wanted to like blow it up and really change it a lot. And he's like, look, the thing that we're optimizing for is this one button that isn't clear that's currently buried in a menu. Like let's just take it out of the menu and make it really clear and leave everything else the same, right?

**[1:04:19]** And I think when you look at a HubSpot, it's like, oh, we wanna redo our whole marketing stack. It's like, or do you just wanna send some emails, right? And if you can be really specific about like, we're gonna just test like these two channels, like new landing pages, new emails or emails and Google or LLM search and whatever that is, you'll find that like you can, the point solution vendors for that are so much better than the Swiss Army Knife vendors. And you may end up walking away with like a huge return on that project if you go with the point solution vendor versus like a failed experiment that took way too long

## 01:04:58 — Email delivery: HubSpot vs. Beehiiv

**[1:04:58]** if you go with a Swiss Army Knife vendor. And so, there's so much momentum and there's so much like, oh, everybody uses HubSpot, you should use HubSpot. Picking on HubSpot a lot and I actually like HubSpot, but like, I think there's this sort of like magic bullet, like, oh, we should just get HubSpot, they do that. And it's like, yeah, but they're crap at email delivery. They're crap at marketing email delivery. Their marketing editor is one of the hardest to use. And if you A/B test them against like a Beehive, Beehive is two X better at delivering your emails, which is the only job that matters. Like, do you want 18% open rates or 36% open rates?

**[1:05:33]** Like I've done that experiment and that was exactly the results and it's like, I don't, what are you trying to do here? Do you wanna like, are you, is the job to be done sending the emails or is the job to be done having your customers read the emails? Like, 'cause those are two very different things to optimize for. And so I think if you're, you know, if you're thinking about say email, like look at the core email providers, don't look at the bolt-on solution because your results will reflect how much knowledge and how much effort they put into it. And whether it's, as you say, just like them chasing a bigger TAM

**[1:06:10]** or people that are actually passionate about the project and the product building a great product. - Well, and I think going back to something you said earlier, I think just defining the problem is usually, you know, you're 90% there if you can define the problem well, especially now that you can build solutions so quickly. But if you're building a bunch of solutions to things that aren't problems, then you're probably gonna be spending your meals. - I was just gonna say, the other thing is like firms like yours where you can knit this stuff altogether, right? I think like, you know, this knitting GTM stuff together

## 01:06:43 — Where firms like LeanScale fit; stage-appropriate tooling

**[1:06:43]** is not something a full-time person should do until you're at least seriously. Like, I don't think. 'Cause it's, you want somebody who's an expert, you can duck in, you can have five people and do it quickly versus like have your intern try to figure it out over the course of six months, right? Like, that isn't the way. And so I do think that like, once you decide like, oh, this is the best email solution for our problem, which it might not be Skip. It might be one of the other ones. It depends on what you're trying to do, right? Then you can easily, everybody has APIs now, you can easily pass everything back and forth

**[1:07:15]** exactly where you need it and get a lot of, you know, get better results. - Yeah, and I think that for lean scale, that's where we know the market's changing so much. There's new tools all the time. That's where, hey, this is all we do. We focus only on this. And it's our business to do this at a world-class level. So we're trained on these things. We have pre-built workflows. We have pre-built agents that are ready to deploy some of the solutions that you're gonna need at each stage. So I think it's also as these develop, I think, like you said, you can, now you can do anything. So I do think there's such an opportunity

**[1:07:56]** for companies likely in scale to say, hey, this is what you should do. You can do anything, but here's some things that work and we can get them going really quickly for you. - Yeah, I love the playbooks you have. I think that the, like, a friend of mine's now back as CEO and founder of LaunchDarkly, which is doing a couple of hundred million in ARR now. And she always talks about this stage appropriate thing, whether it's a hire or a tooling or software or what have you. And I think it's one of the biggest mistakes companies make, especially at Series A and a little bit at Series B, is doing stuff for like really big companies,

**[1:08:33]** whether that's like hiring a CFO out of HP or something, or hiring a VP of sales out of Salesforce. Like, you know, there's some things that like you still, you know, or like trying to implement Salesforce or HubSpot when they're really not necessarily there yet. To me, that's, you know, having advisors, whether that's somebody like me that's your strategic, whether that's more a tactical like implementation advisor, like you and your firm to guide them and be like, look, you don't need that version of it, or you don't need that functionality yet. Like companies at your stage like can get by with this. Opendoor, where I was for a while,

**[1:09:12]** like they did their first 10 million in revenue on a sheet, on a Google sheet. Like they talked about tech enabled house selling, but it was a Google sheet, right? And so, you know, like, you know, I mean, I mean, my price income, like perfect price to company, I started raised over $6 million for like, you know, my co-founder insisted on doing it with machine learning, and I could have built a linear regression in Excel. I learned that stuff in business school, would have taken me 20 minutes. But instead, we spent six months trying to figure out a very hard machine learning problem back in 2014. And, you know, I think if we had gone the other route

**[1:09:49]** and just kind of faked it, we would have been much more successful because we would have had revenue coming in and we could have built the machine learning stuff at our leisure. So anyway, I think the stage specific thing's super important and people need advice to sort of get comfortable with like, I don't, everybody talking about this on LinkedIn, but I'm actually, I'm not gonna do it 'cause it's not right for me. - Right, right. And I think that's where the real pros can separate the noise from, you know, hey, what they should actually implement. As we wrap up here, I got a couple of rapid to fire questions. Hope you don't mind, and we can go through.

## 01:10:26 — Rapid fire: focus, favorite tools, who to follow

**[1:10:26]** So first, I think one I like to ask is, best GTM advice you've received throughout your career? - Yeah, this is a tough one.

**[1:10:39]** I think it actually, the best advice I've gotten is focus, right, focus is the most important thing. Whether that's on the task to be done at the moment, like we were talking about earlier, right? Just like focus and get that one thing done, whatever it is, or on your market. So I think the biggest risk for early stage companies is that they try to be all things to all people. That's what I did at my last company, or as I go, anybody who prices stuff is maybe a customer, right? And the more specific you are, the more focused your product is, the better it will go for you. And I think people, you know, why and I is like companies like Rippling and Deal,

**[1:11:23]** and even Salesforce and HubSpot and all these, their original ICP was incredibly narrow, right? I mean, Deal got its start 'cause it was like tech companies that were dealing with offshore teams, right? Like that's what they did. They didn't do onshore, they didn't do PEO, they didn't do any of that stuff. It was just like a better way to manage your offshore team. You know, I mean, like each one of them had its own thing, but I think that the more focused you are, the better, and the sooner you are comfortable with that focus and saying no to stuff, the better things will go. - I love it. We may have a tie on this one,

**[1:12:02]** or think in the same way based on the conversation, but favorite tool that you're using today? - Yeah, I mean, I think Cloud Code would win it. There are a lot of great tools. I don't, I'm not saying it's the only great tool out there, but it'd be unfair to Cloud Code if I didn't give it a shout out. - It's pretty tough to beat, pretty tough to beat. And then last one, who's worth following? Where are you getting some interesting insights from? Where are you learning? Who else should people be watching? - We talked about Adam Robinson, he's great. Kyle Poyar, I follow him, we're getting pretty close. I just wrote an article for his thing

**[1:12:40]** on building a personal CRM using Cloud, of all things. And I think his content is great, especially when it comes to go to market for early stage companies. He does a great job of posting what the stats are from other companies. So if you think your reply rates are good, go look at his blog and his newsletter and see how they compare to everybody else's. Couple of other ones for more like the founder CEO type, Tomasz Tungusz posts great stuff. And then Sean Burns is a good friend. He posts stuff on leadership, which is really good, like making tough decisions as a CEO. And then if people aren't following Peter Walker at Carta,

**[1:13:21]** they're doing themselves a huge disservice 'cause he posts a ton of great data on startup finance, what people are raising at. The latest one was on options pools and like how you should actually set aside a smaller options pool now. So, and it's really useful 'cause you can go to your VCs and be like, hey, this is what Carta's data says. So we're gonna do it my way. And at least have a principal negotiating position. So those are the ones that I think are great. - Amazing, amazing. Well, Alex, I appreciate everything. Appreciate everything you shared, excited about what you've done, what you've built, excited to follow the journey at Skip

## 01:14:02 — Wrap-up

**[1:14:02]** and just thank you for being here. - Thanks for having me, this has been a lot of fun. - Thank you.