This Company Spends $32m Per Year On Claude Code!?

Growth Newsletter: https://levelingup.beehiiv.com/subscribe Need marketing help? Visit: https://www.singlegrain.com/ and https://npdigital.com/Want to recruit great marketers? Find them here: https://marketingschool.io/hire Eric opens with Spawn, a company with 10 people that spends about 30 times its payroll on AI, and what an AI-native team looks like in practice. They cover iterating fast while changing less, the Money Prompt, and why experience still matters when AI does the work. Eric runs through Jen Abel's list of what startup sales teams get wrong and the advantage of founder-led sales. Neil explains why a win rate over 50% is a warning sign, often that you are not charging enough. Then GPT-6 Astra being distilled, what that means for IP, and how to make your offer a no-brainer for buyers. Speed and a strong offer beat protecting IP. Key takeaways◾AI-native teams can spend far more on AI than on payroll◾A win rate over 50% is a warning sign, not a win◾When IP is easy to copy, the offer and the speed win Chapters00:00 Spawn: 10 people, 30x payroll on AI02:05 Iterate fast, change less03:49 Money Prompt and experience05:17 Jen Abel's startup sales list07:48 Founder-led sales advantage10:46 Win rate over 50% is a warning14:23 GPT-6 Astra distilled: IP is dying16:16 Make the offer a no-brainer 𝗔𝗕𝗢𝗨𝗧 𝗧𝗛𝗘 𝗖𝗛𝗔𝗡𝗡𝗘𝗟Welcome to Marketing School, one of the top business podcasts with over 61 million downloads. Each episode delivers actionable marketing tips and strategies from two entrepreneurs who truly practice what they preach. The show is hosted by Eric Siu, founder of Leveling Up and Single Grain, and Neil Patel, co-founder of Neil Patel Digital and recognized by Forbes as a Top 10 Marketer. 🎙️ Learn More About the Hosts Eric Siu – Leveling Up: https://www.youtube.com/@LevelingUpOfficial Neil Patel: https://www.youtube.com/@neilpatel 📩 Free Resources Ubersuggest: https://www.ubersuggest.com/ Answer The Public: https://www.answerthepublic.com/ ✅ Subscribe for Daily Marketing Insights!

• 0:00Speaker 00

  Finding customer information shouldn't mean digging through emails, call notes, and spreadsheets. HubSpot Starter Plan brings all of those details together, so it's easier to track what's happening with every deal and what to do next. It pulls information from calls and emails. Then it gets to work, logging and summarizing them. Right after your call ends, HubSpot pulls next steps from your call notes and drafts suggested follow-ups. And it stays up to date for you, so you can build demand, win more deals, and spend less time on repetitive work. Your records, insights, and next steps are ready for you to act on. And you don't have to miss a beat searching for any of it. Try the Starter Plan for $10 a month. Learn more at hubspot.com. Would you look at this company, okay, this guy Jacob over here? He's building this company called Spawn. Ten humans. Apparently, he's saying he's spending $32 million a year in Claude spend, okay?

• 0:44Speaker 00

  So he said, "How I built the most AI-pilled company in the world." So Spawn is 10 people who spend 30 times payroll on AI tokens that ship almost as much as Anthropic. In the last 38 days, we shipped 77,000 substantial changes at about 108 an hour using the fleet of permanent AI agents. Each agent owns an area like the engine or finance with real permissions and memory, and this only works with Anthropic's Fable. We killed pull requests, PRs. So agents coordinate through the living code and product like Stigma G. Shipping hourly while fixing breaks themselves. Five of us now focus on taste and entropy while AI judges two years of recorded decisions, so nothing waits.

• 1:24Speaker 00

  So that's, I mean, there's way more that he's putting in here, and you can kind of see this chart where he's doing more and more. But I think there's a couple of things here. One, when I was talking to my CTO about this earlier this week, it's like, we have to fundamentally reimagine how we work. Because marketers, it used to be for marketers, it's like, oh, well marketers shouldn't touch the code, they shouldn't ship PRs. But now you can today, right? These are things that were unreasonable before. And you look at these guys now, they're spending 30x payroll on these tokens. I'm not telling you to spend 30x on tokens. But what I am asking you to do is, I'm saying, go to your people on your team and say, hey, what's the most unreasonable thing that you've asked your agent to do in the last seven days? And you can even ask your agent harness too.

• 2:09Speaker 00

  What are some unreasonable things we can do based on the things that we've accomplished in the next seven days? What should I be asking you that's unreasonable? Because the way that we're used to working right now is we're used to working with humans. Oh, can you give me that report? Oh, can you give me that data poll? Oh, can you give me these numbers? Oh, can you text it over to me? Oh, can you just send me the charts? It doesn't work like that anymore, right? You can ask your agents to do anything you want. Can you spin me 20 iterations of this? Can you test five of these landing pages at the same time? Can you spin up 20,000 ads right now? Can you iterate 10 trial reels at the same time? What are the most unreasonable things you can do? And that's why I say that the demand for this stuff is limitless.

• 2:46Speaker 01

  Just be careful, though, when you are making changes on the fly. So let's go back with Eric's Legion mousetrap, right? You get a free AI visibility report. You can end up making changes on the fly with it, but keep in mind, it takes you time to actually get usage to figure out if your new iteration of your mousetrap is better than the old one. So yes, you have no excuse but to iterate really fast. But just make sure you're not making 50 changes at once or too many changes at once, where you're not sure which changes are actually helping you or hurting you drive more conversions, or sales, or more leads, or more revenue. So that's just one thing I would keep in mind.

• 3:25Speaker 01

  Because in this new world with marketers using AI and not having to rely on developers or designers and just getting shipped, the big thing that I'm seeing in our agency is people are making changes really fast, which is great, but they're making too many changes and they're not sure which ones are actually helping versus hurting. Think of this as like A/B testing. With A/B testing, you need statistical significance. You need enough data to actually show that something's helping you or hurting you. Now, with whatever you're using AI for, if you're getting, you know, if 100 people see it and 40-something of them are complaining about something that's broken, or something's wrong or not usable, of course, in those kind of scenarios, don't wait, just go and fix.

• 4:10Speaker 01

  But just generally speaking, you're like, oh, if we try reducing our form fields from seven to two, we should collect more leads. Well, if that's your test, and you don't really need to test that, but let's hypothetically say you're testing that, you'd want to have enough data going to both variations to figure out if it actually is helping.

• 4:27Speaker 00

  Yep. One more thing I'll say on that too. So earlier this week, so I was talking to some of the engineers on the team, and I said, hey,

• 4:36Speaker 00

  you know, I have this revenue. It's called Money Prompt, okay? So what it does is it analyzes your code and analyzes the entire funnel, it looks at drop-off rates, and it comes up with why, where your funnel is failing. So for example, for one of our products, it's like, hey, you're actually not sending activation emails if people are failing to move to the next step. Oh, you're actually not reminding people to, you know, top off their credits if they're running low on credits, and you're not giving them a deal either. Oh, you know, you have like a 20% of people abandoned at this page over here. So when I walked through those numbers with the engineers, they're like, well, we don't know this stuff, right? I'm like, exactly, right?

• 5:20Speaker 00

  So that whole taste and judgment and experience things matter, it matters a lot. Because I can't expect the engineers to know that. And also, at the same time, the engineers can't expect me to know everything about, you know, having a very, you know, rock-solid build with whatever it is that I'm shipping, right? So experience still matters a lot in this day and age, despite the fact that you can do way more. And so just a heads up to everyone that, you know, you can expect the engineers to kind of just logically figure these things out, but it takes years and years of experience still to, you know, detect these patterns.

• 5:52Speaker 01

  Yeah, and speaking of engineers, one thing Eric and I should have clarified earlier. I know Eric showcased that tweet or expos of someone talking about, I'm paraphrasing here, but the more technical you're going, the more down-channel or the people you're losing. So you want to be careful when you're becoming too technical with your pitch. There's always a caveat where it depends what you're selling. If you're creating marketing material for a company selling GPUs like Nvidia, you better be super technical, because the person on the other end making the purchasing decision is probably super technical. They make it approval to send you a check from an executive who may not be as technical.

• 6:32Speaker 01

  But the person who decided to choose your Nvidia GPUs instead of AMDs, or instead of Intel's, or whoever, that person's probably super technical. So you want to make sure your marketing copy appeals to your ideal customer profile.

• 6:46Speaker 00

  Yeah, actually, you know what, Neil? Let's go through her list, because I want to go through the ones that stand out to me. So you can, I'm going to pick the one that I like first, you can maybe pick one that you like. But this one over here, so she has a list of 15, so she says, "A round of startup sales thoughts," okay? So number 12 on her list, this is Jen Able again. So number 12, outbound sales often has the highest average contract value. So that can be intuitive or counterintuitive to a lot of people. But this is her experience, you know, selling, you know, enterprise SaaS, for example, right? So that might not be something that's common for everyone. People might think, oh, you know, maybe inbound is, but you know, often it's not, right? Oftentimes, the biggest deals are going to be these people that are, you know, putting RFP out, or you're outbounding for the RFPs. But what else stands out to you here?

• 7:29Speaker 01

  So I'm reading them all, and it's kind of small for me, but I'm going to go with number three, and I'm going to do a variation of her number three. Her number three is, "Demo should, so demo of a product should be never focused on the product. Focus on how it increases or grows their influence in the organization." I'm not saying that's a bad way to pitch a demo. The way I like pitching demos that I've seen that converts really well, we've experimented with this a lot, is you're focusing not on the product, but how it solves problems and makes their lives easier. So if you can show an organization how they can solve X, Y, and Z problem,

• 8:08Speaker 01

  and specifically, you can tie it into money made or money saved, that usually increases sales. Typically, if you can tie it into money made over money saved, that usually performs better. If you can't show money made, money saved is the next best.

• 8:23Speaker 00

  Frictionless Growth Marketing, hosted by Sonya Thompson, is brought to you by the HubSpot Podcast Network, the audio destination for business professionals. Frictionless Growth Marketing, hosted by Sonya Thompson, shows how modern brands grow by removing the friction in their marketing and customer experience that pushes customers away every single day, often without leaders even realizing it. Each episode delivers practical insights, real brand examples, and conversations with marketing leaders, showing brands how to win with the new rules of growth, identity-driven marketing, cultural relevance, and frictionless customer experiences. Listen to Frictionless Growth Marketing wherever you get your podcasts. All right, so I wanted to take a moment to tell you about my podcast co-host, Neil's agency called NP Digital.

• 9:05Speaker 00

  And they work with a whole host of global companies or a global organization. Also, Neil has SEO tools such as UberSuggest and Answer2Public. All you have to do is go to npdigital.com to learn more, and we'll see you on the other side. I'm curious, Neil, if you look at her number one over here, "Having the founder involved in sales processes is a natural competitive advantage." I think, I'm pretty sure you and I both agree with that, but at a certain point, the founder can no longer be as involved. So, yeah, go for it.

• 9:33Speaker 01

  Yeah, the reason I like the founder involved at the beginning is the founder typically, and hopefully, knows the product or service that they're selling. So they're going to be able to answer objections, know the issues that people are having with their potential solution, and they can go back and modify the product, or fix things, or identify issues that need to be changed in the future roadmap. I haven't seen the founder being involved in the long run really impact close rates from a success ratio from them saying like, this is how you should pitch, because a lot of founders aren't great in sales.

• 10:15Speaker 01

  Where I've seen the close rate increase from the founders has nothing to do with the pitch. It's more so by the founder being on the phone call. They can figure out issues that people have with the product or service, and they can go fix the product or service, so that way there's less objections in the future. And, you know, a founder can just make things happen in an organization way faster. It's like, do you remember when Google was struggling to get out anything related to AI? And then Larry Page just came in, and people were causing him issues, and he's like, Sundar, these people are your problem. And he just cranked and got out whatever he wanted, and Google just started jamming when it came to AI. And people were like, oh, wow, Google's actually catching up and building some really cool products.

• 10:58Speaker 01

  But it really happened after the founder got involved and just didn't care to deal with any crap that certain people are throwing out of him, and he just moved him out of the way.

• 11:11Speaker 00

  So I think the, the, so, so there's that. I mean, you know, he didn't want to deal with the, the, the corporate politics. But um, I remember when I spoke to Jason Cohen from WP Engine, you know, he, the way he put it was pretty elegant where it just said that the founder, it doesn't matter if you, like, the founder has this power. It's not the hired gun CEO usually that has this power, but the founder has this invisible hammer where they can just make things happen. And, you know, I visited a mutual friend's office um, a few weeks ago. And uh, you know, people were chasing him around because they're just trying to get him to unblock things, right? So there, those people were blocked on things, so they're trying to get the founder to unblock, right? So to your point, yes, the founder definitely unblocks. Um, and I think in the early days, now I think I know and you know as well, that in the early days, especially with a startup, like you have to get sales, right?

• 11:54Speaker 00

  Like that is the blocker in your way. So you should be involved with the vast majority of deals before you, you pass it off to other people. But it definitely is an advantage.

• 12:03Speaker 01

  Yeah. One other thing I love from her list is number nine, a high win rate, fifty plus percent is often not a positive, but negative signal. I totally agree with this. If your close rate is too high, you're usually selling wrong. Expectations aren't right. You're going to get pissed off people or you're underpricing or there's something that's usually wrong. You should not have a fifty plus percent win rate. Um, there's something wrong. Either with the product or you're lying or you're being deceptive or people don't understand what they're getting. Or, you know, another big problem is you're not charging what you should be charging for the product. Like, I know some people, uh, I remember this in the early days of N P Digital.

• 12:45Speaker 01

  We closed a lady that sold um, those hats from uh, Latin America, I think Mexico, you know, sombreros? I think there's sombreros. I could be pronouncing it wrong. Uh, and she wanted to work with us and was not happy with the results. So she emailed me. I got on the phone with her. And before I got on the phone with her, we were going back and forth and she was really pissed off that she wasn't getting the results. Can you guess how much my team sold her, someone in my organization, they no longer work with us, uh, sold their marketing services for to do all their S E O and rank among Google and all this kind of stuff?

• 13:25Speaker 00

  I don't know. Uh, three thousand dollars?

• 13:28Speaker 01

  Two hundred and fifty dollars a month.

• 13:30Speaker 00

  Haha.

• 13:32Speaker 01

  For two hundred and fifty dollars, Eric, do you think a lot of people would sign up for us to rank you higher and get you?

• 13:38Speaker 00

  Oh, everybody was, your conversion rate should be like ninety-five percent.

• 13:40Speaker 01

  Uh huh. Yeah. Really high close rate this salesperson had. Uh, do you think it's possible for us to generate a R O Y or even break even?

• 13:49Speaker 00

  Not today. Not today.

• 13:51Speaker 01

  Even with A I, it's not possible at two hundred and fifty bucks.

• 13:54Speaker 00

  Not today. Yeah.

• 13:55Speaker 01

  Yeah. We, and back then with A I wasn't around like it is today, your losses were even more. So I helped her. I said, look, I want to give you back all uh, I, I, I want to give you back all the money you ever paid us.

• 14:10Speaker 00

  All two hundred and fifty of it.

• 14:12Speaker 01

  Well, she paid us for a few months and she's, uh, she's just like, I don't want the money back. She's like, I want you to promise what was given to me in a contract. Now, being the owner of it, if my team did something wrong, the buck comes up to me. So I was like, all right. I'm like, look, someone should have never sold you for two fifty. That's not your fault. That's my fault. I can't even make this profitable, but here's what I can do. I'm going to give you back your money. I know you don't want it. I'm going to get on a handful of calls with you and we're going to go and fix your business and grow your sales. It went from making very little to making thousands of dollars a month really fast in less than three months. I think I did less than three or four phone calls.

• 14:57Speaker 01

  Uh, and she was really happy. And then after she started growing revenue, she was making very little to nothing before we started. She's just like, hey, so how do I sign up now? I'm ready. And I'm like, we can't take you on. You're not ideal customer, but I gave her quite a bit of feedback. I introduced her to people who could potentially help part-time basis. Um, but she was happy. I don't know what it cost me in time, probably more than the money we refunded. But at the end of the day, it was my fault because my team sold something they shouldn't. The buck stops with me. I had to figure out how to make it up.

• 15:31Speaker 00

  Yep. And so that's good. And, and by the way, Neil, uh, I just want to show you one thing on, on the sidebar over here. So you can see, it said, what does it say over here? It says, Astra gets distilled. So just so everyone knows, Astra is Chat G P T six's model, right?

• 15:45Speaker 01

  So where are you showing?

• 15:47Speaker 00

  It's on the, it's on the right side of the screen. Where you see where my mouse is?

• 15:49Speaker 01

  Okay, I see it now. Yeah.

• 15:51Speaker 00

  Yeah. So there's a live on X right now. It's, it's basically like a live audio session where um, M T F's monitoring the situation. I think which is owned by Andreessen Horowitz. Um, so G P D six Astra, which is uh, Open A I's, it gets distilled. Meaning distilled means that they've, they've basically, you know, copied the model, right? Like it could be like a Chinese uh, company or something like that. Um, and a lot of these, you know, Anthropic complained about distillation attacks. Um, and, and which is funny because in both Anthropic and, and Open A I, they have distilled the internet, basically. They have distilled search, right? But my point of showing, even calling this out right now, you have thirty-six hundred people in this live is because uh, I'm just saying that all this stuff is, you know, uh, software parody is happening so much faster.

• 16:32Speaker 00

  And then uh, Elon retweeted something yesterday where it's like, you look at a, there's like a A I created drug that's in phase three trials right now. There's even a hair loss drug, Neil. There's a hair loss drug now that's in phase three trials that was enhanced by A I as well. And then there's um, there's something that uh, basically doubled your lifespan from like a pancreatic cancer or something like that. So he's just like, we're in a hard takeoff with drugs, but even with this right now, that's why I'm just like, guys, just give all your stuff away for free right now. Because a couple of weeks ago, you had Chama from the All In Podcast, like, you know, I P uh, you know, intellectual property as we know it is, is dead, right? Like it's, it's going to have to evolve quite a bit. So why not? Just give it away uh, as much as you can right now.

• 17:12Speaker 00

  And uh, you know, try to, try to maximize more value for your customers because they're good, they're just going to want more, just like you want more with A I right now. But from a software standpoint, I think all this stuff is going to become distilled and you just have to ship a lot faster. So anything else we want to call out from uh, this Gen Able here?

• 17:29Speaker 01

  Hmm. No, I, I think that's pretty good from the list. That was a good list though.

• 17:34Speaker 00

  This one's good too. Like oftentimes when salespeople pitch, they just want to just keep talking and they just want to talk about their features. They want to talk about their company. But what they don't think about is how good is the offer at the end of the day, right? How is this such a no-brainer where the other party would be stupid not to take it? Uh, and that's not something people think about because oftentimes, even if you're doing agency deals, like the types of deals that we do right now, it's like, okay, they come in and they're like, oh, you know, we're, we're, we want S E O, but it's like, oh no, you, you actually need to do, you should do outbound more. Like you're a consumer company and consumer companies haven't thought about outbound, but like, we have one consumer company we're working with right now. You know, they're doing nine figures in revenue. They're like, we're like, what, what if you guys did uh, cold outbound to med spas and then, and you know, started B two B?

• 18:19Speaker 00

  They're like, yeah, you know, I, I think that's a really good idea because they're trying to go to, grow to like, you know, five hundred million in revenue, right? Um, so if you're able to somehow um, tie it in, price is really good, value is very clear. Um, and they feel like you're smart and they feel like you've done it and you've been there before, right? Then you're going to have a much better chance of winning a deal. So these are good.

• 18:42Speaker 01

  Yeah. Uh, one tool that you can end up using, there's a lot of replicas. I don't know some of the replicas names. I know we've used some of the replicas. Uh, there's Gong. Gong will track how often your salesperson's speaking on a call versus the potential customer. And if you notice that the salesperson is speaking way too much and not listening, you cannot adapt your pitch to solve their problems. You need to listen more than you're actually talking in sales and marketing.

• 19:13Speaker 00

  That is it for today. Please don't forget to rate, view, subscribe, and we'll see you tomorrow.