This Week in Startups - The “AI wrapper” debate, AI-powered lead gen, auto-clips, resumes, and more with Sunny Madra | E1792

Episode Date: August 15, 2023

This Week in Startups is brought to you by… Embroker. The Embroker Startup Insurance Program helps startups secure the most important types of insurance at a lower cost and with less hassle. Save up to 20% off of traditional insurance today at Embroker.com/twist. While you’re there, get an extra 10% off using offer code TWIST. Lemon.io - Hire pre-vetted remote developers, get 15% off your first 4 weeks of developer time at https://Lemon.io/twist. Eight Sleep. Good sleep is the ultimate game changer. Now you can add the Pod Cover to any mattress! Go to eightsleep.com/twist to check out the Pod Cover and get $150 off at checkout! * Today’s show: First up, Sunny Madra is back to discuss the “AI wrapper” stigma, the first official demo-to-investment, and demo more useful AI products! (2:41) Then, Tonebase CEO Igor Lichtmann joins to demo his product, break down how he achieved product-market fit while maintaining capital efficiency, and more! (42:47) * Time stamps: (0:00) Sunny Madra is back for more AI demos! (2:41) The “AI wrapper" debate; VenturusAI is joining the LAUNCH Accelerator as the first full-circle demo-to-investment (10:09) Embroker - Use code TWIST to get an extra 10% off insurance at https://Embroker.com/twist (11:23) Klap.app turns YouTube links into multiple TikTok-style videos with captions and ranks them on "Virality Score" (19:08) Kickresume uses generative AI tools to help users build a fully-fleshed out resumes and prepare for job interviews (25:15) Lemon.io - Get 15% off your first 4 weeks of developer time at https://Lemon.io/twist (26:34) Browse AI helps users build AI-powered web scraping/data extraction bots with no code (34:16) Oppwiser is an AI-powered lead gen platform for sales teams (41:16) Eight Sleep - Go to https://eightsleep.com/twist to check out the Pod Cover and get $150 off at checkout! (42:47) Tonebase CEO Igor Lichtmann joins to demo his music education startup, break down how he achieved product-market fit while maintaining capital efficiency, and more! * Check out Tonebase: https://tonebase.co Follow Sunny: https://twitter.com/sundeep Check out Klap: https://klap.app Check out Kickresume: https://www.kickresume.com Check out: Browse AI: https://www.browse.ai Check out Oppwiser: https://oppwiser.com * Read LAUNCH Fund 4 Deal Memo: https://www.launch.co/four Apply for Funding: https://www.launch.co/apply Buy ANGEL: https://www.angelthebook.com Great recent interviews: Steve Huffman, Brian Chesky, Aaron Levie, Sophia Amoruso, Reid Hoffman, Frank Slootman, Billy McFarland, PrayingForExits, Jenny Lefcourt Check out Jason’s suite of newsletters: https://substack.com/@calacanis * Follow Jason: Twitter: https://twitter.com/jason Instagram: https://www.instagram.com/jason LinkedIn: https://www.linkedin.com/in/jasoncalacanis * Follow TWiST: Substack: https://twistartups.substack.com Twitter: https://twitter.com/TWiStartups YouTube: https://www.youtube.com/thisweekin * Subscribe to the Founder University Podcast: https://www.founder.university/podcast

All right, everybody, welcome to Monday. I am back in Tahoe, just for a little summer jaunt to see the lake, take the girls out for a little rope course, maybe get on the lake on a boat or something, some hiking. They like to go to the beach, the pool, everything. How's your summer going, Sunny? Summer's been good. You know, I'm in build mode. You know, it's a summer of AI.

It's the summer of guests. It's all around. And so this one's been pretty heavy on the work front this year. But that's good. It's sort of, it's an important set up summer, I think, for a big, big fall. Set up summer. Oh, I was calling it no, I started the summer, stress-free summer. Oh, no, no, no.

And then I got halfway in, and now it's the set-up summer. I tried to have stress-free summer. It didn't work. Yeah. Yeah. But I am doing. We had three years of that, 2020, 2021, 2022. Those were not stress-free. Those were gangbusters.

But yeah, they weren't depressing and like licking your wounds and trying to fix everything in your portfolio. Yeah, I've been meeting with a lot of LPs and that has been very instructive because now venture capital is kind of back to basics, back to square one. Like back to, you know, just find a great company, invest in it, burn a low amount of money, get product market fit, be super focused, be super lean. It's a different world. It's the J-Cal's sweet spot. I kind of like it. It's kind of like exactly where I like to get, you know what Carmelo Anthony would like to get that little 12-footer?

Go ahead, back you up. They try to ride it, do that little floater. I kind of like I got my spot on the floor. And, you know, they're feeding me the ball. Constantly. This weekend startups is brought to you by in Broker's Startup Insurance Program

helps startup secure the most important types of insurance at a lower cost and with less hassle. Save up the 20% off of traditional insurance today at a broker.com slash twist. While you're there, get an extra 10% off using offer code twist. Lemon.io. Need to speed up your product development without draining your budget? Hire vetted engineers from Europe at lemon.io. Go to lemon.io slash twist to get 15% off for the first four weeks.

and 8Sleep. Good Sleep is the ultimate game changer. Now you can add the pod cover to any mattress. Go to 8Sleep.com slash twist to check out the pod cover and get $150 off at checkout. And the thing that's most interesting is I'm getting a lot of great AI companies. And they could be dismissed as AI Rapper companies or chat GPT Rapper companies or, you know, pick your language. model wrapper companies. Yeah.

But what I'm finding is they're finding all kinds of pockets of where to add value in the wrapper. So, you know, there's a website or an app, a wrapper around an iPhone or a wrapper around AWS. I guess you could diminish it to that way. But I'm curious your thought on this debate in Silicon Valley. If you can explain it from your own words, the AI rapper debate. Well, look, I'm going to lean into a company that you are famously known for Angel investing in Uber.

Oh, yeah, I was in that one, I think. Yeah. And one could argue back then, someone would say the location API on an iPhone through iOS, oh, you're just building a wrapper around that. Yes. Right? But that was a very important API. That API has led us to have Uber, Instacart, DoorDash, Dhab.

a whole bunch of, you know, services, with location kind of being at the core of it, but then they had to build a whole bunch of infrastructure beyond just the location API. But without it, you know, you probably couldn't get off the ground. I view AI as the same thing. AI today is an API. It's a very powerful one. But if you're going to build a business, you need to build a lot of infrastructure, a lot of scaffolding around what's happening.

And I think the companies that are focusing on that are going to get there. Now, obviously, Apple benefited tremendously off of building the platform and off of building a place for developers to build off of. But many companies, and probably in aggregate, the value of those companies exceeds Apple's or at least maybe close to it. And so I think when people go to that argument and say, oh, everything is just a wrapper, well, we can boil that down to almost all technologies that. have scaled off of a good platform that got off the ground. Yeah, I think it's very well said. And, you know, we had in our discussions here, you brought me a company called Venturous AI. Remember on episode 1774?

I interviewed the fan you, Emmanuel. I like the product so much. I offered me a 25K investment live on the air. And now Ventriss just accepted the $100,000 off from the launch accelerator. Oh, right. So it's our first full circle demo. as an investment. So I'll make sure that you get a little tasty poo.

You figure out how to do a little carry share. No, I'll just write a check. Or you can write a check. I'm good with that. That was the perfect example of a rapper company, right? And you could dismiss these, but I can tell you, like, the level of detail they went into in doing a SWAT analysis, the pastel analysis, all this stuff. They're getting traction amongst, and for folks who don't know, it's a very simple

concept. I'll share it with you right now. Or you explain the concept while I pull up my screen. Yeah, the concept is, you know, if you have an idea, what it does is it takes your idea and it breaks down the basic types of analysis that you would do for any idea. You know, everything from SWAT, which is strengths, weaknesses, opportunities, threats to pestle. And, you know, it gives you, like, here's a framework right here, exactly. And it can formulate a business plan for you, an MVP, unique selling points and go to market motions. This is, you know, look, for a long time, this is what you'd pay a consulting firm to do for you. Yes.

Or you'd go get an MBA and you'd understand this through, obviously, some type of business school. And for many people, I think what this does is help them understand. Because a lot of people have ideas. And you know, you and I have both seen that throughout our careers. But what people struggle with is, how do you take the idea and then start to test? its viability and which framework should I use for testing it. Now, some people will just go and build something and see that it works or fails in the market.

But this allows you to do some higher level analysis before you get into the build phase and maybe, you know, before you spend your money, your energy on building something. And this is, as we've discussed with generative AI, you may not want to make a health care decision on it right now or a finance decision. Obviously, they can hallucinate. but when you think about just ideation coming up with ideas fancy word for coming up with ideas and developing them brainstorming another word for that chewing the fat you know jam session this is just giving you so many jumping off points so when you start doing these analyses and the one i'm

showing on the screen here if you're watching is for the venture capital school i'm thinking about starting like i want to start a program to train people to be associates at venture firms and have them work at our venture firm as part of it, it really does do a great job of giving you some ideas. You know, some of them are obvious. Leveraged social media to share success stories of alumni and showcase the expertise of faculty members. That's actually very succinctly said.

If you were going to start a school, those would be two blocking and tackling things that you might forget. And you're, oh, wow, you know, we should move that up the list. Obviously, testimonials and, you know, the dossiers of our faculty members would be critical. So the other thing is when you, look at this, when I make investments in early stage companies, everybody asks me, like,

how do you make your decisions? And one of them is product velocity. You could see here, they have updated their, you know, whole site since we talked to them a month ago or something. So they definitely are adding things. Like path to an MVP and unique selling point, I don't think we're part of the analysis. And here they're upselling you. Hey, you got to pay for this, right? So back to the wrapper companies, I think there's so many ways to add value to, you know, some verticalized app, whether it's travel, whether it's business analysis,

sales, you can add so much. Imagine taking each of these ideas and using it as a jumping off point and just having group discussions inside of this app, right? Like just being able to store the,

you know, people using these as jumping off points, right? Or moving them from a bullet point to a base camp, like a punch list, you know, a project management platform, right?

You can go from ideation project management. Pretty interesting idea, right? And I, you know, I've used it a few times now, and a lot of people, similar to you, Jake Kellogg, reach out and have an idea or something like that. And I literally will just take what they sent me, put it inside the tool and send it back to them. And now you're a genius advisor. Yeah. I mean, I'll be very transparent. I say, look, like, these are the things you need to think about, right? And then, you know, I'd create, I think it's, I really think it's an incredible tool. I'm excited we demoed them. I excited that they join the accelerator and,

you know, writing a little check. That'll be nice to. Yeah. Yeah. All right. So the voice. Demos. Voices you're hearing.

Jason Kalakhanis, agental investor, Sonia Madra, co-founder of definitive intelligence. They provide AI enhanced data analytics of public and private data, including autonomous data science agents. That means people going out on the internet

and working on your behalf. Not people. We like to do demos. Robots. Bots. Yeah. Agents. Yeah.

Pick whatever term you like. Listen, I work with super early. stage companies at launch, like literally year zero. They haven't even incorporated yet. And then we hit the Series A. People have thousands of dollars in MRR. And maybe they've only raised a couple 100,000 before that Series A. And they don't have their insurance set up. And in fact, we recently had a great startup that didn't have DNO and we had to really stop everything because they were having board meetings. They were making massive decisions. There were legal issues. And they didn't

have the basic D&O insurance that protects directors and officers. So we send them right to Embroval. And broker is business insurance built specifically for startups. A single application will help your startup get four quotes for four lines of coverage in 15 minutes. Think about that. Four quotes, four lines, 15 minutes. And they're going to connect you with one of their expert brokers for unmatched service that goes beyond your policy.

We use it at launch. It's easy, peasy, lemon squeasy. It's easy, breezy. What more do I need to tell you? I use it. I love it. A lot of our startups use it. They love it.

Try and broker it today with the code twist. and you'll get 10% off their startup package in broker.com slash twist. That's eM-B-R-O-K-E-R.com slash twist, and use a code twist for 10% off. Okay, let's get back to this amazing episode. So let's get into our demos. We always like to talk about demos, and you and I are on the chat thread, and we're constantly sharing things we find on the internet. People talking about stuff, generative AI.

It's obviously moving quickly here. So what do you got first this week? Okay. So this is a build off of something. that we've been continuing on a theme on, which is from transcription to clit or to subtitles. And you know what? After last week, I found this tool, which is called clap.

And what it does is you give it a link to a YouTube videos. In this case, the free version is limited to under 30 minutes. So I found this clip from your session with Tim Ferriss. Oh, cool. And in this particular session, you were talking about wealth and happy. happiness. Yep. This might, you know, it's Tim Ferriss.

I'm assuming the whole thing might have been a couple of hours because he doesn't do seven minute podcasts. Yeah. And I haven't listened to it. But now I've listened to this clip. And so I gave it the clip. And then what it does is it, it's generated. It's called clap.

dot app. K-L-A-P-A-P. Yeah. Okay. And what it's done is it's generated nine TikTok style videos from that seven-minute clip. Got it. And those videos have all the standoff. subtitling and all that.

But what I really like is it found a way to generate nine videos and it gave them all a virality score. And so the first one here has a virality score of 85 out of 100. And it explains that this video has a high virality potential because it touches on a relatable topic of redefining success beyond financial wealth. What a really interesting thing that it kind of pulled from that discussion, right? And then it sees you. your enthusiasm and passion for life can resonate with viewers.

And it gives a reason of why this thinks this is 85. That's pretty scary. I mean, good. Why is it scary? If I asked a video editor, you know, like an offshore video editor, why this is a good viral clip, they wouldn't be able to say, right? In all likelihood. Yeah.

And this goes back to our theme. AI, remember J-Cal we're talking about last week and said 18 months? It'll be as good as humans. And even better. I mean, this is the crazy bet that you have. made. In 18 months, we'll be sitting here in the new year, basically January 2025, you're claiming, Sandeep, that in January 1st, 2025, like, it's going to be as good as a human to do these kind of tests. It is a bold prediction, but I would have a hard time saying

over under. I might say over by a little bit. Yeah. But it could be under by a little bit. Yeah. Yeah. But we're not in orders of magnitude anymore. We're just within like, you know, you're not saying 10 years away, right? No, no, no. It's not the 20 year. It used always be when you made a prediction, hey, when we have flying cars, you say 20 years because nobody remembers you made the bet.

Yeah. Or so you can do it under 10. You might know, and you can see it's done all that. And then, you know, as you scroll through it, shows different scores. See, this one is a 72 out of 100. This is a 72.

This is a 70. And you can go 68. And so this, I'm just looking at number six out of them and says, well, why is this a 68? Well, this is moderate. potential for virality. And it's the topic about finding fulfillment and doing what you love is relatable and can resonate with a wide audience, etc. And so, and it explains, however, the lack of visual elements and the length of video may limit its reach compared to, and it's

kind of looking at like just the real nuances of what creates engagement. So they've trained it that the clip is too long and it's not visually exciting enough. Yeah, because it would be cool if I could say add collateral and make it more. Well, you know, it's, funny, that's exactly what Nick and I were talking about when we were just doing like our dry run here. We're like, that's the next phase where you're like, okay, well, make it better for me. Take this from a 68 and make this into 100. You know, that's a standard collateral thing.

Like, what we did, we did a video one time on our TikTok where I was talking about the movie The Founder and it got a little bit of traffic. But then I say, can you remake that and put clips from the movie in it? Yeah. And then that one was the one that went viral. Yeah. So, look, you know, this is just an evolution. and, you know, I'll make another prediction.

In less than three months, we'll find one that does what, you know, both of you guys are saying, which is now it's taking a seven-minute thing, turning it into ten clips and scoring them. I think in less than three months we'll have one of these that'll not only take the clips, but it'll give it the boost of what's needed through an understanding of, you know, either doing, like, you know, adding in transitions and things like that in, yeah. Well, I mean, and if you fed it, which I'm guessing, well, what do you think the founders did there. Do you think they said, here are some heuristics we know. These are great rules. So, you know, look at each video under, you know, with, and in the prompt or something, they're saying,

hey, examine the video for these qualities, visually appealing quotes, whatever. Or do you think they said, here are, these are trended videos from TikTok, you know, infer what should come. Yeah. You know, in the next video. Because if it did, it'd be like, you should be a sorority girl dancing during rush week. Do, yeah. those are pretty popular right now. Unbelievable. What I would say... They don't have a block on TikTok.

Why don't they have a block on TikTok? Can I just say no, none of this? Just more food, more dogs. Yeah, exactly. I need to just tell them. Please, no more dancing. Sorority girls in my feed. And it's like the group chats that get you.

Because someone will send you something in a group chat and you open it and then it's like in your algorithm, right? Oh, right. Yes. If somebody sends you something, yeah, now you're tied to it forever. Yeah, exactly. So that's like TikTok bombing. You could TikTok bomb somebody.

If I sent you like, oh my God, these five TikToks are amazing. And I send you five TikToks about like people who like to dress like evil clowns. Yeah. And now I've like turned your TikTok talking to like evil clown talk. Yeah. It's terrible. It's like there's, remember the Rickrolls? There's an equivalent of Rick rolling.

Yeah, yeah. So what do you think they did here? I think they just put it in their heuristics, right? I think so it's very similar to like, you know, Venturis, right? where look, there is some set of frameworks that people understand, right? In their case around business, in this case around what it takes to make a viral TikTok, I think in building a startup, it's built by two folks. I think it's like, look, they've started.

I think they have their names at the bottom. We can give them a quick shout out here. So it's built by, you know, Theo and Victor here. Oh, hey, Theo. Hey, Victor. Yeah. And so look, my guess is that they haven't done the, you know, deep training of like, here's a bunch of videos, figure out what happened, apply it.

I think it's like they have some notes. own, you know, formats and things that they want it to talk about. Because in some of it's a little bit repetitive. You can see it say the same thing. But look, if you're going to build these things, that's where we started the conversation. A location API just gives you, you know, where a person is, just a lot long, right? You've got to build a lot of infrastructure around to build a company. And you don't build, start up by building all the infrastructure.

You kind of crawl, walk, run. So I would start in the same spot if I was then. Yeah. And you can totally see that the people who are jumping in figuring how to do these verticals, which you could be cynical and say, oh, if I had the right prompts and I had an unlimited amount of time, I could be on chat, GPT, doing this myself, right? Yeah. But we all don't have unlimited time.

And we all don't, you know, sit there and obsess over it. So, yes, you could theoretically build your own Airbnb, like make a message board software. And then, you know, email all your friends and say, does anybody have a place or know somebody? I mean, yes, you could do all that work. Or you could let Airbnb maintain it, right? Yeah, build the network effects. Okay, what's next here? Okay, so next is an interesting one.

This is like one that is kind of near and dear in some ways, but like, and I think it, it makes sort of the use of the technology easier here. What is the name of this? Oh, sorry, this one is called Kick Resume. Kick Resume. Okay. And like, these resume builders have been around for a while. Like, that's not a new concept.

But like, this one uses the power of AI in a couple of, of different ways. And so I kind of did like a, and it could pull your stuff in from LinkedIn. I didn't want to have it do that because I just didn't want it to scrape my LinkedIn, but it has the ability to do it. And what I can do here is I can say, I'm going to add work experience. So I've just got like, you know, for the folks listening, I have a resume UI on the right hand side. And I have, on the left hand side where I can add. So I'm going to add work experience. You have like prompts over there. Like add a work experience. Like a four. Yeah. Yeah, Exactly. So I'm going to say, I say working at Quist and I'm going to say I'm a podcast host, right?

Yeah. And you know what? I'm going to do and say, oh, you know what? I don't know what to do because like resumes are tough. And so I have this AI writer and I can basically say generate content. And it's like, well, for what is I say podcast owes and generate. And you can see here on the left, it starts to write the content. And on the right, it starts to put it into the resume and fill it in for me. And so, right. Look. And it does this. for all the different sections, like your skills and, you know, and other areas. It's an incredible starting point because it says here, like, listen, I collaborated with a team of producers, editors, and sound engineers, switch your seamless production and delivery of high quality podcast episodes, maintaining a strict release schedule and meaning all project deadlines. Who knows where, you know, chat GPT or whatever they're using here, got that information. But somebody apparently had put that on their resume. And so it scraped it or, you know, and then put it into its language model.

guess the next word. What's the most next, likely next word after collaborated with for people who are having podcast producers. And here you have it. And it's a great way to start because you'd say, you know, we didn't have really sound engineers. We just had editors. So collaborating with and oh, it wasn't really a team. It was a producer and an editor. So collaborate with, you know, a senior producer, an editor. And you can just edit it from here. But again, brainstorming with you and suggesting things that you might not have, of and giving you great perfect grammar and structure. And look like, this goes back to, yeah, can you do this in Open AI?

You can, but you're not going to get a full-on framework, like the resume that's coming out of it, right? And you may not know like the template you want to do it in around work experience and the tense and the voice and this helps. Now, beyond that, it does some interesting things. It has like this AI toolbox, which I'm just loading up here. So we'll give it a second. And what it'll do is it has a.

a, it'll score your resume for you. Yeah. And it'll tell you, hey, you know, you've got these repetitive verbs. Maybe you're doing it. It's got like a built. Yeah. Yeah. Exactly.

Results. You're using like, you know, words that are overused. Yeah. And look, and, you know, it's got a checklist. Make sure you got your email address. Last name, work experience. It's telling me I didn't put my phone number and I don't have an education section. It's giving me some benchmark scores in terms of like word count, start date to end

dates, you know, just a bunch of framework stuff, which, you know, I thought was really cool. It then has a section that will help you prepare for a job interview. Okay, so put in here you want to be a podcast producer. Podcast producer, sure. I want a big shot. I want a podcast host to producer. Right. Can you tell us about your experience as CEO co-founder and how you contributed to the

company's growth and success. So that was based on your first title on your resume. My first. Yeah. Yeah, exactly. What strategies did you implement to increase market share and establish the company as a key player in the industry? How did you secure venture capital funding and utilize to it to expand operations and accelerate product development? These are great questions. And yeah, this would be great. If you think about this, like a resume coach to run you through an equivalent here would be hundreds to low thousands of dollars. Like, what does a resume? may coach slash writer charge 500 bucks, a thousand bucks probably. Yeah, I never had to use one other than when I was in college when we used to do co-ops.

And then our department, our co-op department helps us with that stuff. But, you know, look, this is a real thing for folks. Like, you know, people still expect resumes and even if you're applying maybe even to be a fantastic. A job at any level. And look, not a lot of people have access to these kind of tools. And, you know, they may not know what to do. doing, and this going back to just, but maybe we'll stick on that theme today, a lot of people

may not know what to do in chat GPT if you got access to it, even for free and say, how to, how to have it help me with a resume. This puts a framework around it. Yeah. And it makes it a lot, more straightforward. Great. Awesome. Yeah. Really. And then you can download it, proofread it. It's got this career code section as well, all the basic stuff. Yeah. I mean, and when it gets more and more data, it's going to get better and better. So, you know, it's right now they don't have, you know, think when they get a hundred podcasting related people to build their resume in there, now they'll have a training set for

podcasters. I mean, of course, they're probably scraping a ton of resumes they're finding on, you know, career sites or LinkedIn, yeah, yeah. You know, building up data, but, you know, that's going to be one of the interesting big debates. There was a company

in Israel that sued LinkedIn and won, or LinkedIn sued them and the company won. That was just scraping you know, basic data off of LinkedIn's public profiles, and they wound up winning. So LinkedIn tried to stop them from taking that data. So now, of course, building a derivative product, I don't think you're going to win. But it was IQ Labs that did that. You can look at up H.I.Q. Labs. Okay, listen, you got an idea for a tech startup. Great. You think you want to change the

world. You think you got this. This is the one. Well, you've got that same problem that we all do. You don't have an engineer or you don't have enough engineers to make this happen. And you need product velocity. You need to go fast and how are you going to go fast and how are you going to control your burn rate if you got no engineers? Well, what if you had a partner who could provide you with more than a thousand on-demand developers? And those developers were all vetted, experience, result-oriented, and passionate about startups and building great products. Well, what if they also charge competitive rates? Does this all sound too good to be true? I know it does. Well, then you need to head to lemon.io because your dreams have come true right now, startups,

choose Lemon.io because they only offer handpicked developers with three or more years of experience and ones that have strong portfolios. Only 1% of candidates who apply to work at Lemon.io actually get accepted. And if anything goes wrong, Lemon will get you a replacement ASAP. A couple of great launch founders have worked with Lemon.i.o, people in our portfolio, and they have had great experiences. So I want you to learn more at Lemon.io slash twist. And when you go there, you're going to find your perfect developer or an entire tech team. And you're going to do that in 48 hours. or less. And Twist listeners get 15% off the first for a week. I want you to stop burning money. I want you to hire developers smarter and faster. Visit lemon.io slash twisting. Give it a shot.

Joe, Jacob, you have this like uncanny way. And this was like the demo coming up after the next one, but I'm going to jump forward. Okay. Because, you know, the topic you're talking about is like, and I know you've been talking about this a lot for the, you know, the team at launch and inside. So this is a new service called Browse AI. And what they do is they make it super easy for you to build a structured data extraction robot or a site monitoring robot. And this one, I'll just play their video because it's a little bit. What do those words mean in English? Sorry.

Extract structure data monitors. Yeah, if you want to scrape, if you want to scrape something from a site. So if there was a website like ESP. And ESPN had a bunch of player profiles. And you say, you know what? I just want to get their last five games points scored. Yeah. There's no API to do that.

Or I might not be a programmer. I could use this tool to structure, take all the Nix players, and give me how many points they scored and then tweet it out. Yeah. Correct. Where every 15 minutes, X out, tweet out, put a post to X. Doc. Okay, you can tweet. I'll just don't say tweet so we know what we're talking about.

tweet to X.com that, you know, how many points, you know, Brunson score, or Steph Curry, right? You can have Steph Curry point score. Every time he scores a point, it tweets. Yeah. You would be able to get the data through this and then put it in there. Got it.

So one is like, so they have two types of robots. One is like sort of like a one time. Like you just want to pull a bunch of data. Okay. And the second one is site monitoring changes, right? And so one is like imagine you're trying to monitor a competitor's pricing. You're like, oh, I want to make sure if they set their pricing lower, I set my pricing lower. And this first one is more like, hey, I've got a research job and I've got to scrape a thousand profiles and I need to put it into some kind of tool.

And so what they do in this example, and I'll just, I'm going to just quickly scroll through it. I won't play it because it's a seven minute video. But the example that they do here is, and I'm assuming they must be Y Combinator funded. I didn't check, but they scrape the Y Combinator logged in site. So they even allow you to log into something. and it scrapes the Y Combinator site. And the way it does it is, and this is like a fantastic use of AI, you basically install a plugin.

And with that plugin, you kind of highlight the sections you want. You tell it what those sections mean on the page. So you can say like this is the company name, this is the location, this is the description. And then basically it will, from that, create the scraper for you. And then ultimately that turns into like a list that exists on their site here, right? that has the location, the logo, the link, and everything else. And so it does what you know, you would typically ask a programmer to do if they're writing a crawler for you.

And does it sort of by using a browser plug-in and your own, like human knowledge around it and it automates it for you. And then it can dump it into, I think you'll see in here, they'll dump it into like a Google sheet, right? They'll export it to a Google sheet here. And then here you go. So this is that all that Y Combinator information from the logged insight of Y Combinator, of my combinator companies, and they've got all the different parameters here.

So going, you know, I know you've been talking about this a lot, needing this for research or needing this for a lot of different purposes. Now you have the ability to do this, you know, pretty easily without any programming knowledge. And how is it doing the AI? The AI is to figure out what the different columns are in your spreadsheet eventually? Exactly. Like what they're doing and look like I, you know, we should maybe have them on or something to talk about it more detail.

But like, my guess is that when you install the browser plug in and you tell it the sections you want, that underneath the hood, they're taking that and turning it into the, you know, grabbing those sections in code and saying, hey, the name lives here, this lives here. And then they're using that to then, like code interpreter, write a script that says, hey, I know that in these CSS elements, in these HM elements, this is where the name is. So go and grab it from me by going to this URL. Got it. and then how does it know where to go for the next one, the next one,

the next one, and like do the crawling part. That's the AI bit, right? That's like, when you show it two or three, it knows, okay, it figures out the pattern like,

you know, you've written some HTML back in the day, right? You'd see like H1, and then it'd be sitting in a table and you'd just see it over again. That AI can kind of figure that out from that repetitive nature of it.

So there'll be in a way, it's an intelligent crawler. So it's an AI crawler in addition to an AI data analysis tool. It's probably doing, It's probably doing it's probably doing it's an extraction tool. It's not doing any analysis. You would do your analysis and code interpreter. Yeah, exactly.

Like you have it. Yeah. So just extracting. Yeah, I think this is why you may have seen Twitter like didn't allow people or Facebook doesn't allow people who aren't logged in to see data. And this is the reason. These tools are getting so sophisticated. Yes.

And then, you know, if you're not, so they're going to require you to be logged in. So they can, if you load too many pages in too short a period of time, they can turn you off. And then the concept of an open web with data on it is kind of going away, which in some ways is sad. Because if you leave all the pages opening, you don't stop people, they can just fire up crawler after crawler with different IP addresses and then just, you know, over time, have everything you... And it's a tough balance because in this case, like, they'll allow you to log in. So they're doing it as a browser plug in. So they'll even allow you to log in and then let the browser run automated. So they'll work around like sort of some of the automated

crawl issues. Now I think by running it through a browser, it'll stop like, like, it's not feasible to crawl all the tweets that have ever happened on 10,000 pages. Exactly. But if you load 1,000, it probably won't care. And if you do it every five seconds, it won't care. But if you do it too fast, the website might have a trigger that says anybody loading over 10 pages. Yeah. You know, I don't know, anybody loading over 100 pages a minute is a, is a spider. Yeah. So, you know, straddle them. Yeah. And I think like the flip side to that because I know you like the flip side is yeah like sometimes you're doing some work and you want a list of folks right and it is something like look I pay for LinkedIn and LinkedIn's pretty expensive like you know hundreds of dollars a year thousands of dollars a year. Yeah.

I want to get a list of 100 podcast producers. Why can't I do that? And they're like, well, you can do it. We just want to slow you down because you're a paying customer so that you load one page at a time and it would be better if I could say put if I could if I could do searches of up to 100 people at a time and I could do this. a hundred times a year, so up to 10,000 people as a paid user, and I could export up to 10,000 records at a time. That would be fantastic. But instead, people have to use tools like this.

And the people who get the value are not the paying customers. The people get the value are the people who are the most gray hat hackers who are able to do this kind of stuff. Whereas the people who actually have a valid business use, they don't. But, yeah, that would be like a cool tool. If your premium account let you, you know, gave you the results instead of in the kind of, you know, web page after web page results. Yeah.

If it gave you actually a more of a table, you know, which crunch based does. So, yeah, I mean, that's my idea, basically. Premium is you get some structured data as opposed to unstructured data. Yeah. If you're not paying, you get the web page with the data on it. Everyone wants walled gardens, right? And that's the problem versus, like, you know, and if you're using it for business, it's challenging, but another great segue,

J-Cal. You're on fire today. This is going to be a good week for you. This is a new style of like Zoom Info tool. So it's like Lead Research. I don't know if you guys use Zoom info. We have an investment in a company called Lead IQ. There's Zoom Info. Yeah. Yeah. So this is like sort of an AI powered one. And what it does is like you give it like a name of a company and you know,

like I'll give it something one of our customers. and then like it'll just find like similar companies related to it right and so I find it's like really great like it's it's sort of and if you don't know what you're looking for you can ask the AI to help you and say oh like I you know find me companies in this particular industry and you're giving you kind of ideas and names and so the whole the whole it's like an AI lookalike finder which I thought was like really neat because I think Zoom info is great and we use it but like what we're seeing now is this enhancement in and around the search. interface that exists, right? Yeah. And so this allows you to kind of leverage the power of AI for search. And, you know, related but unrelated, like we even saw this, and I think you had a back and forth with Elon on Twitter last week where someone is complaining about like search on X and he's like, yeah, we're going to make it better with AI.

So we're just starting to see that happen in all industries where someone's like, let me make the AI, like let me leverage the AI to make the search so much better on my product. And then that lets them get a chance to even just build a competitive product, right? And so if I go here and I write HR software and it'll basically come up with a list of companies that it believes are good for HR software and, you know, I'm off to the races for my lead gen work. Fantastic. Yeah, that is definitely what people are looking for and there's going to be, you know, and then there's also write the email for me to contact this person or research this person tell me more about, you know, the targets at this company. So you can find people by sector. You can find people by, yeah, you can find people by a specific company, all of

those different things. And this is what people use things like G2 or, you know, other software, captera, different software rating websites. And they'll go to the software rating websites and say, hey, I've got a partnership with Zapier. Who's similar to Zapier? Oh, if this, then that, this other automation tool. Oh, I use Notion and Koda. What are other knowledge management, you know, business.

process tools. Boom, here are the other ones. And so, yeah, it's very, very powerful, I think, to help find leads and then to get you in touch with the right person. And this is going to just lead to many more people contacting us to sell us SaaS software, which, to be honest, I'm there for it. I love SaaS software. I am being able to, I would say conservatively, I've been able to get 25% lift every year using SaaS software. And now with AI, I think it's going to get us to like 40, 50% lift every year. We just did an analysis inside our company, and I haven't really reviewed it yet, but I got it yesterday, where people talked about all the major tasks they do, and then how many they have times they have to do that a year, then how much time it takes to do each time, and then normalize that into minutes and hours. And, you know, it's like, okay, sorting through the inbound companies takes us, you know, each week, you know, whatever it is, 400 companies, five minutes per company to sort them. Now that sounds like, okay, so that doesn't sound like a lot, but, you know, if, you know, Five times 400 minutes is 2,000 minutes, 2,000 minutes,

two thousand minutes, you divide it by 60. It starts to be like, okay, that's a full-time job just doing that. Yeah, at least. And could you use AI to do that, right? And the AI could sort it. Hey, these are the, or you can, we have our own internal ways of sorting it, like, is their product in market or not? You know, we already have some sort of sorting.

But over time, we can look at the companies we've invested in that have been successful and then try to put it against this database. I think that's two years out, but I do think that that's why I'm building up this database of all these startups is to start getting that knowledge. Well, you know, the other thing that's happening is in what you just talked about there, when you force your team to take processes that are loosely documented and are maybe even heavily documented,

but put them into like an AI framework, what you're doing is you're capturing and memorializing that and you're codifying it. And for many companies, a lot of times that knowledge get lost. So I don't know like for you, but if you have like a really great analyst and that analyst, most of it's documented, but there's a few steps that aren't, I think there's a real power for companies to move stuff into the AI world by documenting these processes. Yes, yes. Yeah. It's 100% correct. If you're forced to document everything you're doing or teach somebody else how to do it through the documentation.

And this is the right first culture that did particularly well at Amazon where people had to write like little six-pagers, they call them. And there's a great book working backwards that explains the philosophy. And there's tons of short articles about it on the internet. But basically, if you have to write something down, you explain it to people. They'll even write the press release for a product. They haven't launched or they'll write the FAQ. And they'll have an internal FAQ frequently asked questions and an external one. for customers or partners.

And if you have to write those, you have to brainstorm. Hey, what are the questions people are going to ask me about this new product or service I'm making? Or if you're taking an internal process and doing an FAQ for it, you know, you're just whatever the most common questions people have, which could come from a meeting or it could come from brainstorming or AI. You'll have documented it, which then means you could make an AI process out of it eventually. And I think, you know, that's starting to happen inside the enterprise, right? And that's, you know, one of the things that we've seen on the enterprise side of things is the big consulting firms are making huge investments because they, you know, they've had these eras, right?

They had the client server era. They had the, you know, migration to Linux era, the cloud era. And now, you know, they're having this AI one. And in this particular one, the unlock is incredible and it's huge for them. So, yeah. All right. This has been amazing. Well done. Tons of great, great demos that got us all thinking about how AI will impact our startups and the enterprise. Sandeep Madra, definitive intelligence. Thank you so much. You can follow him on Twitter.

He's at Sunday. Are you sunny or Sandeep? Sundeeb. Sandeep. At Sundeev on Twitter. And in between these Monday episodes, he's talking about AI. And you'll get a little preview of probably what he's going to talk about here if you follow him on Twitter. Okay, next up on the pod, Igor from Tone Base. Tone Base is a subscription-based service that we invested in that has grown phenomenally over the years. And people pay to learn how to play classical guitar, violin, etc. It's an incredible interview about getting true product market fit. Stick with us. If you want to get ahead in your career, then you've got to be ready to fight every day. You've got to grind it out day after day. This is obvious. You're listening to This Week in startups. You're grinding right now. But if you're a workhorse, you need to get a good night's sleep, period, full stop. We all know that getting great sleep equals better performance the next day. Sleep is a secret weapon, and your secret weapon to getting great sleep is eight sleep. You know, I love my eight sleep mattress.

I use it every night. I got it in both houses because I want to be able to control my temperature, and I want to make sure I optimize my sleep. And you can do this with their new product, the pod cover by eight sleep. It fits on any bed, just like a fitted sheet. It's going to keep you cool all night. all the way down to 55 degrees Fahrenheit, if you like, and it will improve your sleep by automatically adjusting the temperature on each side of the bed, and you get personalized sleep

reports. If you're one of those quantified self-people, you're going to love this feature. And when you get great sleep, you get great performance. And listen, everybody's been talking about it publicly on Twitter.com. I'm sorry, X.com. Benocosla, Paul, Graham, Zuck, myself, everybody's been talking about how great eight sleep is. Invest in the rest that you deserve by going to 8Sleep.com slash twist. They're going to give you $150 off the pod cover because you're friends with this podcast. That's E-I-G-H-T-S-L-E-E-P.com

slash twist for $150 off the pod cover and they're shipping, not just in the U.S. and Canada, but they got the U.K. and select countries in the EU and Australia. All right, welcome back to the program. Next up is Igor Lickman. He is the CEO and co-founder of Tone Base. They provide online music education

from expert instructors. How do I know this company? Well, they came to our accelerator's 10th class back in 2018. The launch accelerator is just seven companies. We introduce them to a couple hundred investors and help them with their pitch and how to think about scaling their business. And now they're doing millions in revenue. I won't say exactly how much, but they've got thousands and thousands of subscribers.

And they teach you four instruments currently. Classical guitar, piano, violin, and now cello. it's not cheap when compared to your Netflix, but it's not expensive when compared to hiring a music teacher. So think 50 bucks a month, 300 a year, something to that effect. So Igor, welcome to the program. Thanks, Jason. So when we met, you were trying to figure out product, market fit. You had some videos up on a website. And online education was starting to become a thing. I am most interested in how you've been able to, I wouldn't say bootstrap the company, but you have been incredibly frugal or capital efficient, as we like to say internally.

You've been incredibly capital efficient while getting product market fit. So the subscribers to Tome Base are very loyal. So maybe you could explain a little bit about what you've learned over the last five years in growing from tens of thousands to millions in revenue, from dozens of, of subscribers to thousands. What have you learned about product market fit and how to scale a company so capital efficiently? Yeah, I mean, obviously, there's so many, many things, you know, we will learn.

But, I mean, I think what it comes down to is to really narrowing down and focusing on that exact thing that you can prove that you solve for customers. There's obviously a discrepancy between the ideas that we have and the missions and the visions and the big kind of, you know, idea for the future. But at the end of the day, we'll learn that, like, ultimately, I mean, you know, it's nothing new here. You probably heard this, you know, and this is one of the things that you also recommend to your founders all the time.

Focus on the customer, provide solution to the problems, and do more of that over time. Yeah. And how do you, or how have you, been able to understand the customer needs? What are the tools and techniques that you've used and added to your toolkit over the years to really understand the users so then you can refine the products. And, you know, because sometimes users tell you they want something, you build it, and then nobody uses it. So it's almost like, you know, you ask them like, hey, if you ask somebody, what we should add to the product, they might tell you a bunch of things that they have no interest in using. So now your user is just

guessing and throwing ideas at you, like in a brainstorming session, as opposed to, hey, we've studied their behavior and we talk to them and we know they want X, Y, and Z, and then they actually use it. Yeah, absolutely. I mean, so in terms of like tools, I mean, obviously everyone is using the same kind of, like, if we're talking about data analytics, we're using all the, the heaps of the world and kind of like in all that and just kind of like understand. I mean, we're like a video, like the way we solve problems for our customers is through

video content. It's live and it's pre-recorded. So this is kind of like the canvas, the vessel, how we kind of like, you know, pack its value and then give it to the customer. And I think it was through, yeah, just like over time, just like recording a lot of stuff and releasing it and just tracking, you know, and then you see that some things get a lot of engagement, some things then don't get a lot of engagement. And then like over months and quarters, so you kind of like realize it, huh? Somehow it's always the same things that kind of like tend to get a lot of engagement and a lot of use from our customers. And as we kind of like, you know, try to kind of narrow in and kind of like really zero in on that type of video,

and the type of content and really build things, build things from there. How much of the business has turned out to be live versus library? And this is something content creators always think about. We did a little experiment here with live shows. It was good. People liked it. You know, us talking about the news, but it seems like what they really wanted was founder guests, investor guests, really going deep. on information as opposed to live.

How do you look at the world? What have you discovered about live versus library? Yeah, live is certainly a certain important aspect of the overall product experience. But I think it's just, let's say, 30 to 40 percent, really. I mean, I think at the day, it's content. Because with content, with pre-recorded content, you just have so much more control over the final outcome. Like, you can't design the solution for that problem trying to solve with that piece of content. Whereas with life, it's more like a point in time kind of situation.

It also really heavy depends on the artist. It's heavy depends on the format. We do use some of kind of group workshops and kind of like webinars where people can ask questions and sometimes, you know, play something and present and we bring instructors in. But at the end of the day, you know, I think it really comes down to like content and really targeted content specific to people's problems. Let's demo the product.

What we're looking at here is the dashboard, you know, I can like log in. And the majority of our customers are sort of like intermediate through advanced. So these are not people that just kind of like both the guitar yesterday. They're a little bit more inexperienced and they know what they want. It's basically that this is the content library. You know, people can either learn a piece or develop their skills and knowledge, right? So let's assume, you know, I want to learn the Moonlight Sonata. That's a piece by Beethoven.

That's something that I want to play. Then I find this new chorus by Seymour Bernstein, who is a fabulous, wonderful musician. It's a 96-year-old man, great, great artist. Actually, Ethan Hawk made a documentary of him. And so here, I'll just drop down the volume a little bit. So basically this is the format. This is this wonderful musician who spent his whole life, you know, teaching and kind of like playing this music, walking the viewer through the pieces.

We see here we build a number of kind of, you know, custom tools to enhance the learning experience. We have the scores. So I see you can download the workbook, download the addition. And so this is all a custom video player. So unlike YouTube, you can't have all this stuff on YouTube. and here you can have all this backup material, etc. Goal right into the video. Absolutely.

Yeah. So this is built right into the video. And then obviously we have thousands of these videos. And also, you know, at the beginning of when the customer signs up, they're assigned a level based. They go through like an onboarding question. We ask them a number of questions, a number of to identify kind of like their playing proficiency.

And then based on their playing proficiency, this is kind of like where we go from creating content. and it's been a relative to the recent development, something very excited about, this question, how do we personalize the learning experience? How do we really make sure that each customer is paired with the right type of video?

And so here we have from level one through level 12. And based on that level, people are recommended a number of courses, repertoire, skills, and knowledge. And, yeah, so this is kind of like how the content consumption functions. And obviously, then we have the life events that are in various formats, you know, sometimes it's artists that come in and we interview them. Sometimes it's webinars, sometimes it's workshop, sometimes it's group settings. And then obviously there is a big aspect of the platform, the community, which is very important to a number of customers.

It's really this ability to interact, you know, with each other. But also, you know, we have things like, you know, community challenges, you know, where we ask people, to like this, for instance, you know, it was just one unfinished business challenge. Yeah. Yeah. So we basically here, we ask the people, so we give them a prompt and be like, we all have these pieces that we kind of started, but they're never really finished, right?

So like go in and here's your unfinished business challenge. And yeah, people are participating. Wow, that's great. Yeah. Yeah, yeah. uploading. Yeah. And so, you know, typically when we invest in a startup,

We look for great founders who understand their customers really well and are just have product velocity. But, you know, we really watch what investors say when they look at companies. And I think anytime you pick a niche, hear classical music or, you know, music instruction for the higher end, people might say, hey, the Tam is too small. Talk to me about, you know, what you actually think the Tam is for this product ultimately. Can this be, you know, you've proven you can make millions of dollars a year. Can it be tens of millions, hundreds of millions?

And how do you get there? And how do you explain that to investors who maybe, you know, are skeptical? I mean, so the way we think about this is really bottoms up, really. I mean, I think, you know, we approach things from, you know, this understanding that, you know, if we are serving everyone who wants to finesse and prove their playing ability in an instrument, then I think there's no skepticism that there's a lot of people that play instruments. The way it works for us is that we see the overall Tam not so much expressed in the number of individuals who participates, who are basically part of that market, let's say obviously,

you know, the market for beginner guitarists, you know, or the market for, I don't know, people who want to ride a bicycle. I don't know, obviously larger of people. But the way we kind of like qualify a market is less about the total size of that market. But by the market, the kind of like the value, how much people are willing to pay, right? I mean, I don't know, to make very remote example, Palantir, they have like 100, 150 customers, you know, and they make a lot of revenue. Yeah.

And they typically have three-letter acronyms. So they have large budgets in here. The equivalent is you know, you're not selling a $5 a month app. You know, we're talking about a $50 a month app. These people are 10 times more committed and probably even more than that. I would say 100 times more committed than a casual person looking for music instruction. The teachers, I understand it's tough to make money as a musician. obviously and a music teacher, tough to make money as a music teacher, really hard to find students.

If I'm a music teacher and I live in Reno, I've got a population of one hour drive and I might have to drive to them. So is remote music and these live seminars and stuff like that, is that just mind-blowing for the teachers in terms of the ability to make money? Or how does the model work in terms of their compensation? Are you helping them make a business out of this? The way we work with instructors is both, I mean, on the pre-recorded side, it's a one-off production, they're fairly compensated. On the live side of things, no, they're also like, they're vet that they have to be invited. So for us, really, it's about controlling what we present to the customer. So we don't really focus on one-on-one, kind of like people can't really book one-on-one lessons

with instructors as other players in that market. personally, I don't necessarily believe this is a very viable business solution, kind of like taking a cut of a transaction that's small already, and the solution is a lesser version of the real thing, you know? So this is where the control kind of, you know, comes in. And this is where the live aspect, the live, kind of a portion of the product is an important add-on, but it remains an add-on and kind of like, I think. And it does seem to build community.

saw when you were going through the community site, that's like a great way for people to meet each other. And it's hard. I mean, if you play cello or violin, you know, those are more niche instruments. Those students probably don't get to interact with each other all that much. How many could there be if you lived in Reno? How many serious players are there? So the community aspect, does it drive revenue or does it just keep it from help with retention? It certainly helps with engagement and retention. I mean, I think we haven't, we haven't found a very direct correlation between revenue and community engagement, but we certainly know that some of the most of the power users, you know, they are very, very active. And this is also where word of mouth happens,

you know, where people talk to, you know, to their friends and their colleagues and their students and the teachers. And so I think this is, this is, but it's hard very to put like a, to quantify the impact, you know, of the community on net revenue, if you will. How do you think about scaling the business from here? There's the venture community that put pressure on you, I think, and the entire startup ecosystem during the boom days, 19, 2019, 2020, 2021. Go faster, go faster. Burn money, invest. And now we have austerity. And hey, can you hit profitability? How much runway do you have? How are you growing in a sustainable way? You've always taken the approach to grow sustainably. I get your updates. I'm like, hey, can we? we grow faster, what ideas do you have, where is this all going? And you and I have I always that's dialogue and I've always respected that you are very committed to sustainable growth. So maybe you could talk a little bit about how investors and they're looking at tone base, how they're maybe we're pushing you to go faster and then maybe now appreciating your

methodical approach. Not that you're going slow, but you are going methodical. Yeah, yeah. I mean, I think, you know, it just comes down rather to to, to, to, to, to like, you know, I'd rather build something really. I'd rather grow to $200 million in like 10 years, you know, then crush and burn in three. And so I think that the investors, such as, you know, yourself, you know, supported the business, you know, they understand, you know, the, that every company has its, like every growth trajectory and every growth story is unique

to each business and sort of like shoehorning every business across the board into one, you know, know, exponential, viral, you know, growth expectation may or may not harm to the longer term outcome of the business. The way we think about really scaling the business is that like the music community or the art community is really as fragmented by its definition. You see, like, you have like obviously when we talk about music, we have different genres within different genres. We have different instruments within different instruments. We have different communities. And the way we think about this is that if we can build one, replicable, write one replicable playbook, how we deliver and product experience that's

unparalleled, you know, versus the zero to one kind of, you know, approach. You know, we launched flutes, you know, this month. Within a month, you know, I can, you know, I think it's fair to say where the market leader that sort of. In flutes, yes. And this is the playbook, just like Travis or DoorDash, they went from city to city. you're going instrument to instrument. You've got the playbook.

How do you get the first hundred flute players involved? How do you reach flute players? There's no database of that, or does search-based advertising work? Are there communities on Reddit or other message boards that you can kind of participate in authentically to get users? My co-founder, Chris, did you know and that you met, who runs the marketing team, is obviously the more qualified to answer the question.

But obviously, you know, the way we think about high building is, you know, it's a combination of content. It's a combination of targeting the target community and creating, basically building a lead list. You know, we would run a tripwire. So we call them like, hey, like give us your email and download a free practice tip for a flutist, for instance, you know. And we built an email list of a couple thousand people, you know. And then we obviously recruit really famous flutist ahead of the launch, we then share some of the trailers, you know, out of time that we prepare.

And then we get to those 100 flutists. And actually, in the first day, it was more than 100. I'm curious about people having profiles on the site. Do people have profiles on the site yet? And can they kind of start to feature their work? Because that was the other thing I always thought, as you, as I see the activity in the message boards, I would think some people might want to, you know, say, hey, here's my profile.

You know, I'm tonebase.com slash, you know, Jason. And hey, here are, here's some of my guitar solos and kind of making a profile. Have you thought about that? And then they could be public profiles where they share their passion for their instrument. Yeah, this is where the community comes in. You know, we're talking about GCC. This is where we encourage customers and users to share their progress, share their thoughts, share their ideas.

If it comes to the pedagogical content that we then present, you know, to the customer, this is where we kind of like have to be very selective. And this is where, yeah, usually the, this idea of like, you know, you just see this didn't really work out so far. Oh, so no profiles yet. No, not yet. Not yet. It's got to be in control.

Yeah. It's going to be very interesting if people could start to build profiles and share it. But, I mean, what an incredibly beautiful product. All the product design, all the video work done in-house. you set your own standard, you took your time with that, yeah? Yeah, yeah, in terms of the videos. Well, there's how well produced the videos are, and then there's how well, how great the UX is on the site.

I'm curious, just what you learn there, what your philosophy is there, production inside the video, and then the rapper, the UX on the website, you know, how much you've invested there in terms of your cycles and time. Yeah, absolutely. So, I mean, you're building out our own studio in Pasadena. We're building out a second now in Europe and we have a third one on the East Coast. That was crucial, you know, to really controlling, controlling AV output, controlling the costs and controlling the visuals. And this is, you know, our head of production, Sean Mulholland, he's done an amazing job really to really kind of control and really shape the visual identity of the content.

Because this is also something, you know, that a lot of customers are kind of responding to it positively. If we are charging $50 per month or $300 per year, that screams premium product. And if we are a premium product, we have to deliver this premium experience. So if somebody opens the video, they're like, oh, wow. Like this is where this you just see thing kind of like, you know, it's a little bit difficult. Janky. It's janky. I mean, let's be honest.

I feel like you're a niche version of Masterclass. where a masterclass you can, I'm sure, you know, Yo-Yo Ma, somebody's on master class doing a masterclass for cello, but they're not going to do the next thousand cello videos, the next thousand flute videos or 100. So, you know, the master class, what Steasy does in video, what you're doing, it's all just amazing in terms of helping people get an education. And you had a nice bump during COVID and you used to stay in that bump. It didn't go away, right? Or did you go away a little bit? Yeah, I mean, it's interesting. Now, if we now look at, now that we have, what is it, three years at this point, if we look at those cohorts versus other cohorts,

they didn't retain that well, I think it's, it was a bump certainly, you know, it was interesting. But, I mean, again, you know, I think at that point, even we knew it's like, it's a hopefully it's a temporary recurrence, you know, it turned out to be longer than we all hoped for. But, I mean, you know, I think we're focused on, on, Because I feel like building a sustainable business means also like targeting the right type of customers that are in there for. Your ideal customer profile, the drive by, you know, transient users, they eat up cycles.

They don't use the product. It's not product market fit. So, yeah, it's kind of a waste of everybody's time. I mean, there's no harm, no foul if they want to sample it. But then you having to service them as opposed to putting time into finding that people who are truly passionate. How many instruments per year are we going to release, do you think? Is it one per year? And in 10 years, we've got, you know, the 15 major instruments or two per year.

How do you think about what's the right pace? Yeah, yeah. So, I mean, I think it's interesting. I mean, last year we released one new instrument. This year, released two new instruments. Next year, we're planning on doing three to five. So it's really about optimizing the more efficient we become, the more profitable. we become the more content, the more revenue we can reinvest into scaling more verticals

and doing that more effectively. And then it becomes about the process, really. And so. What's what I hope for you is, you know, you just keep investing in the product. You cannot go wrong. Every dollar that goes into the product, you can see it on the screen. You're really, really produced something that was always quality from the start. You always had quality, but you have scaled quality production.

For that, you have my great respect. I'm so proud of the effort the team's put in and the passion comes out. Every time I see an update from you and you've been great keeping the investors updated. So I'm just so happy for you and the team. And just kudos on staying relentlessly focused on something you're passionate about. It's very easy to get distracted. Keep investing every dollar you can into the business. I'm fine with the business losing a little bit of money, breaking even whatever you choose. But I just love the fact that every time you make another video, it just adds to the value for those users. It makes it

harder for them to churn. And, you know, it's, I look at what you're doing, like I look at my Epic Pass for skiing. I'm like, you know, okay, this thing's $600. When I get to day four or five, I've now, day three or four, actually, in America for skiing, I've broken even on the cost of my thing. And then I'm like, oh, I've hit

20 days. I've hit 30 days, 40 days, 40 days, 50 days. It's cost me $10 a day to go skiing. That's literally how I look at it. And it's the same thing with your beautiful servants. Every time you had those videos, the cost per video for the user goes down, right? Yeah. And they get to just keep enjoying it for the same low price. And this is why the majority of people are in yearly plans and another third is on lifetime plans,

just because people invest in their long-term development. I love the lifetime plan. I know accountants lose their mind over a $700 one-time transaction. But listen, if it gets a bunch of passionate people to help underwrite the business with that lifetime.com.com did it a number. People have done it. You know, you don't have to sell a million of them, but if you sold a thousand of them, it's not the end of the world. Those people become your super advocates, and you could always discontinue it or keep raising the price of it. Yeah. But I love the lifetime. It's 700 bucks lifetime. 700, yeah. Wow. So amazing. All right,

listen, continued success. Everybody go check out Tonebase.com, tonebase.com. Go to Tonebase.com. Go to Tonebase. com and see all the magic. All right. Well done. And we'll see you all next time on this week and startups.