AI Tools Every Founder Needs: A Practical Guide to Faster Validation

Hey everybody, welcome to the Founder Institute webinar. My name is Jonathan Grecian, co-founder and CEO of the Founder Institute. We got over 1,000 people that have registered for this event today across a variety of our platforms, so it should be an exciting event, um, a really engaging discussion. And just so that all of you guys know, if this is your first Founder Institute webinar that you've joined, these events are completely live, right? And when you have an event like this that's live, we want there to be interaction. Otherwise, you can just go to YouTube and watch a video. So please, if you're here, tell us where you're joining in from. I can already see we got people from Córdoba, Argentina, Bielefeld, Germany. I'm not sure— people— well, Marcel's saying people say our city does not exist. I've never heard of Bielefeld, Germany, excuse me. But I see Barco from Serbia, people from all over the world here too. Which is awesome. A couple of quick just housekeeping items off the bat, guys. If you do want to participate in this event in the interactive portions of the event, highly recommend that you access this event not on a mobile device, but on a desktop or a laptop. Number 2, you will be getting a video recording of this event as well shortly. It takes us just a minute to get this processed, but we'll be sending out the video recording and the deck from this event too, because I know there's probably gonna be a lot of really amazing tidbits in this deck that you're gonna want to, um, they want to apply later. All right, well, let's get into the topic of the event today. As you guys know, um, at Founder Institute, what we do is we help entrepreneurs launch businesses, and a big part of doing that is helping them validate, uh, that their business is going to work, do the market research, start to reach out to customers, start to do the customer development, kind of all the nitty-gritty components of entrepreneurship. And now we have some amazing new technologies that are coming into the fold that are really leveling the playing field for entrepreneurs in a lot of different ways. But I think even more practically, when it comes to validating your startup, right, a lot of these AI tools can save you a hell of a lot of time and also a hell of a lot of heartache, um, you know, by helping you see if certain strategies may work and certain other strategies may not. So we have a bunch of AI tools within the FI system, and you guys have probably seen there's a lot of AI tools being launched right now to help entrepreneurs. But probably one of the most impressive ones that we've seen is a tool called SparkRocket. So, um, SparkRocket— and we have Scott, I'll introduce you here in a second— but SparkRocket is a tool that organically we've been starting to see a lot of people coming through Founder Institute just start to use this product, right? So as a result, these people, uh, founders started telling us they were using the product. We started looking more into it. Uh, now we have a partnership, uh, with Scott and SparkRocket as well. So Scott is gonna, uh, today kind of come through and talk about some of the different ways that you can use AI, um, in general to help validate your business and kind of de-risk what you're doing as a startup. And then he's going to give a short demonstration as well around how SparkRocket can help you do that. So Scott himself is certainly no stranger to startups. I mean, Scott, I know you've been an entrepreneur now for many years. You also, you were the COO of Techstars for a while as well, an organization I'm sure most people on this line are familiar with, but maybe you want to give another couple words about your background and kind of what brought you to building the SparkRocket solution.

Yeah, fantastic. Thanks everyone for welcoming me today, and hola, guten Tag, ohayo gozaimasu for all those from around the world. Happy to be here and excited to be here. You know, I was introduced to the Founder Institute team through some mentoring in the German cohort. I've since done it in Colorado, I've done it in Silicon Valley, and so I have really good relationship with the team here and excited to be presenting today. And so what I wanted to do, and this is kind of atypical of a presentation because I think of presentations should be all about you as the audience, but today I'm going to start talking about me. And that's what's unusual. And I don't normally do this, but what I wanted to share is—

and just one quick thing, Scott, before— so I'm going to drop off and the floor will be yours, but just For anybody in attendance, guys, if you do have any questions, please don't be shy. Put them into the chat, right, after we go through the presentation. The vast majority of this event is going to be Q&A. All right? So don't be shy. Throw your questions in the chat. We have multiple members of the FI team here collecting the questions. And then I'll come back on screen here a little bit, and we'll, you know, start to get some— have a nice discussion with the audience. Okay. Scott, I'll drop off, floor is yours.

All right, fantastic. So a little bit about me is, you know, I've kind of worked throughout the entire entrepreneurial ecosystem, you know, from being an innovator as the CEO and co-founder of SparkRocket. You know, I started my first business when I was 7 years old. So entrepreneurship has been in my blood for many, many years. But I've also been an operator. I've grown and scaled several businesses. As Jonathan mentioned, you know, my last company was Techstars, where I helped grow them and scale them from 16, 17 accelerators to the 50+ accelerators they got up to. And so I've done that, but I'm also an educator. I teach entrepreneurship at the University of Colorado Boulder. That's where I live. I'm based in Boulder, Colorado, although today I'm in grayish Stamford, Connecticut at some meetings. So that's where I'm calling in from. So that's a little bit about me. Innovator, operator, educator. And one of the things that I saw during this is there was a need for the ability to help those earliest stage entrepreneurs get up and running quickly. And so that's why I've built SparkRocket. And what we're going to talk about today is really about, you know, we'll start with kind of the founder's dilemma. Like you want to start a business. You have this idea, but you don't know how to do it. And as you, you know, spin your wheels, you know, you're wasting time. You know, how do we speed that up? If you're— if you don't know how to do it, you're burning money because you're spending time maybe building features that don't fit the market. You're in the wrong market completely. You're chasing product-market fit in the dark. And we all know that horrible secret that 90% of startups fail. Because you're building the wrong things. And so that's really the dilemma that, that all these entrepreneurs have. And I wanted to find a way that I could help entrepreneurs get over these hurdles really quickly. And so that's why, you know, where I'm here today, I want to share with you some, some ideas around how to validate your idea, get it off the ground, and then share with you a number of AI tools that you can use at the various stages of, of building out your, your product. So, you know, what we say is the old way, it still works, you know, and when I say the old way, back in the dark ages when I started, I wrote a 40, 50-page business plan for my first business idea, and it was very static. I never wanted to change it because, you know, it kind of, there was a thread weave through the whole thing, and if I made one change, you know, I'd have to change lots of places, and it was really hard and cumbersome. And then about a dozen, 15 years ago, Steve Blank came out with this idea of the Lean Canvas. And that made it much faster. Now I have a business plan on a page, but nevertheless, the validation process still took time. You know, you still were doing manual customer interviews, you were still building your MVP and iterating, you know, based on the feedback and still had the opportunity of running out of money because it took time to capture all this customer data and synthesize that. And what does it mean? And then start building and getting that feedback and things like that. You know, I love this quote on the right from Paul Graham at Y Combinator, you know, release early. I've seen a lot of startups die because they were too slow to release stuff and not because they were way too quick. You know, my version of that is if you're not embarrassed by what you ship, then you've shipped too late. Get something out in the field and start getting that feedback. And so that's where you really start to get that customer feedback and information of what you're building. Are you building in the right direction? Are you building, a product that actually solves the problem. So with that, we've had this shift over the last, you know, 4 or 5 years now with artificial intelligence, and it's been incredible because it's so fast. It's able to synthesize all of that market data that's out there from so many different sources and help you refine your idea really quickly. It helps you catch, you know, flaws that you may have in your concept. But the thing that I get most excited about and, you know, wasn't something that I initially had in the forefront of my mind when I built the application is the democratization. Before, it used to be if I want to build a startup, I need to be in Silicon Valley. Like, I need to be there with, you know, all the smart people. And that's where all the venture capital is. And that's where I have to be. Artificial intelligence takes all of that collective data from all of the great minds in the world and allows you, whether you're in a Caribbean island, you're in an African country, you're in Latin America, anywhere, you could be far away from that Silicon Valley and still get a lot of that same access, a lot of that same feedback that you might otherwise not, not be able to have. And that democratization of access to that same information is really exciting. I know when, when we first launched Bark Rockets in particular, We're in 100+ countries. Like, I thought we'd be in the US. Little did I know that we would expand that way. And, you know, one of the things that we always said at Techstars is that great ideas are everywhere, but opportunity isn't. Well, now opportunity and access really can be. And so that's really a huge thing for me. So where did this all start for me? Well, it started back in 2017, 2018 timeframe. These researchers from MIT and Harvard approached us at Techstars and said, we've built an algorithm that we believe can select companies for investment, and we'd like to test it with live data. So we gave them 3 years of our application data, and they came back relatively quickly and said, hey, we, we can actually train the model to select companies for investment. And based on financial outcomes— so did the companies raise money? How much did they raise? Did they have an exit? We found that our algorithm can perform on par with, with your managing directors, the people we had in the company who were selecting companies for investment. And in some cases, they could actually outperform us. And the, the example I'll give is that if someone submits, you know, a 3-page application, you or I as humans will read the entire thing. But if someone submits a 10-page application, you and I, we're a little bit lazy, we'll scan it. Well, their algorithm read everything. And so it often selected those 10-page longer applications for, you know, investment, and they had better returns. And I thought this would be a great complement to what we're already doing. But unfortunately, this was before large language models were out. This is before the OpenAIs and the Claudes and the Geminis, and it was just too cost prohibitive for us to pursue. But when I left Techstars, OpenAI came out shortly after, and I was mentoring and coaching several startups. And I said, hey, there was that idea. I wonder if we could build something. And so we built a very early prototype. And again, I was solving my own problem. I wanted to become a better investor. And what we found is that entrepreneurs really loved it, because it helped them get from that blank page, I don't know where to start, to something. They had a business plan on a page they could start moving towards. And that was really exciting for me. And so that's That was kind of the genesis of what we built, but it started, you know, close to 10 years ago with this idea out of these researchers. So today what I talk a lot about, and you've probably heard the expression, you know, it's AI-human collaboration or human in the loop. Like, that's really what artificial intelligence is doing. It's not building startups for you, but it's making you faster. It's making it easier for you. And so what I always say is, you know, you as the human are needed to prompt the AI, to prompt it, to tell it what you want to do. That's where I say AI handles that foundational information, the market scans, the customer discovery, business model building, things like that. It gives you those results. But as we all know, AI can have hallucinations. So you have to apply your judgment to it. You have to look at that, you have to analyze, and you have to validate. Is this correct? And then from a mentor perspective, you can now leverage your time with your mentors more strategically because they're not buried in the spreadsheets and the details and things. They're helping you at that higher strategic level. And I think that's really a key thing, keeping the human in the loop. So again, AI doesn't replace mentors and founders, but it can help amplify you. And that's really a critical aspect of this. So what I've developed is this, this 5-step framework, and this is what I use with, with my students at CU, is the first step is you just need to clarify your, your concept, you know, turn it into a testable hypothesis. You know, once you do that, now you can start getting out and talking to real customers and identifying who has the problem. You know, throughout my career, I've seen so many great, great ideas and great solutions but they don't have a problem. And I always say, like, what problem are you solving? That, you know, if you're building technology for technology's sake, that's more of an academic exercise. We're trying to build a business, and so we need to be solving real customer problems. The next thing is to, to map your competitors. You know, every startup thinks that they have their unique niche, which is great, it's your unique differentiator, but there's probably someone in the space who's doing something similar. May not be exactly the same thing, But this gives you a way to quickly scan a much broader landscape to identify any potential competitors you may have in your space. And then the key thing is really to test your assumptions. You know, what beliefs have to be true for, for you to, to move forward on, on your product? Or do you need to pivot? Do you need to kill it? Those kinds of things. And then ultimately make that, that go/no-go decision whether you can persevere or, or you need to do something different. And, you know, in my example, I built this from an investor perspective with an investor lens, and I quickly pivoted because I found entrepreneurs liked it. It wasn't— I wasn't building for investors anymore, I was building for entrepreneurs. So this is one thing that I'm always, you know, kind of fascinated in. And these are from, you know, my LinkedIn feed. It seems every week or multiple times a week, there's someone out there sharing their thoughts on different LLMs and which ones are good for what. ChatGPT is good for brainstorming and Claude speaks in more of a language that I speak, or Gemini is great if you are in the Google ecosystem. And these are great. You know, I think they're fairly accurate for, in my experience. And so those are tools, but those are kind of general tools that anyone can use for any problem. But what's unique now is there's lots of companies building things specifically for startups. Specifically for entrepreneurs to build products. And that's what we're going to talk about today. So going back to kind of the, the Clarify concept, I've got kind of this theme running through where, you know, you may come up with an idea like, hey, I just love to build an app for, for fitness enthusiasts. It's, it's pretty generic. And, and what you really need to do is identify, you know, the persona. Who, who's going to buy this? Why are they going to buy it? What's the cost associated with it? And these different AI tools can help you create that clear hypothesis, whether it's you're using ChatGPT or you're using Consensus, which is focused on academic research, or one that I've been using a lot is I, I've been using Notion a lot because I can very easily do Claude or Gemini or ChatGPT, you know, all in one interface. And so that's, that's really nice. But there's all kinds of other products out there that, that do other things that will help you clarify that concept. Really what I say, help you build that first elevator pitch where you can clearly articulate what you're building. The next is identifying, you know, your customers, and this is probably the number one thing I talk to my students about is, you know, customers can't— you have to be solving a problem that they have, and, you know, artificial intelligence can help you do that, and, you know, you can create simulated synthetic users. You know, you can have simulated AI personas. You can analyze demographic data with ChatGPT. You know, one thing that I think is often overlooked is Reddit is a rich repository of information. People can post things in there. You'll get real feedback. You can use artificial intelligence then to actually start analyzing a lot of that data and feedback because what you really want to do is you want to understand the customer's pain points. What problem are you solving? What are the demographics of the people? You know, what are their buying influences? You know, all of those kinds of things. You can use all these different AI tools to do those kinds of things in identifying your customers. The next is mapping your competition. And here's a, we'll call it a dirty secret from the investor world. If you come to me and you say, I have no competitors, that immediately is a red flag. Everyone has a competitor. They can be a direct competitor that, you know, you're competing, you know, Coke versus Pepsi. They can also be an indirect competitor, using that same analogy, Coke versus water. Like, they're both liquids, but they compete very differently. But I have a choice between, you know, drinking a soft drink or drinking water or milk or anything else. And then there's just substitutes, you know, how are other people solving, solving these things. And so you really have to ask yourself, okay, I have these competitors, how do I differentiate myself? And that could be from a monetary perspective. Are they pricing it different? You know, what do customers love and hate about, you know, other products in the market? And I think, you know, some of these in terms of water, like how many different bottles of waters can we have, but they're all somewhat differentiated. And so they're all competing against this for essentially the exact same product, a bottle of water. But it really helps you understand, you know, your unique advantages and your differentiation. So that's a key thing that artificial intelligence can help you do because it scans and synthesizes a much wider swath of information than you may have local to yourself and be aware of. The next is really testing your assumptions. And, you know, the first thing, you know, what we want to do is we want to help you de-risk your idea. And that's one of the things that, that artificial intelligence is really good at because you can game plan with it. You can run different kinds of scenarios through it. And so, you know, there's gonna be certain assumptions you have to make. And in this concept of, you know, a fitness app, you know, some of the things you have to believe or you have to validate are true, or will busy professionals change their workout behavior, or will they pay $15 a month, or will they, can you acquire customers for $50? Now, it could be one or multiple of those things have to be true for you to move forward. The best way to validate some of those things are through some, some testing. And so that can be through a landing page with some ads. Are people signing up for it? Are they joining your waitlist? Using, you know, AI-powered tools, you know, you can generate a list of survey questions. So you can go out and actually talk to real users, or you can create a landing page that you can then test with. So there's all those different kinds of tools that are available to you with the whole idea of I want to de-risk my idea here. And then the last step, you know, what we highlighted is you have to make that go/no-go decision. And, you know, it's, hey, I've built something, I've validated my core assumptions, my customers want it, they're willing to pay for it, I can acquire them, great. But in other cases, I have to pivot. And, you know, one of the things that, you know, we always talked about at Techstars was our investment criteria. And it was in order, team, team, team, market, market size, opportunity, and then idea, because we knew the idea was, was less because there's so many companies have to pivot. And so we, we intentionally knew that that was going to be something that we'd have to— so we invested highly in the team, not necessarily the idea, because we knew it was going to change. And ultimately you may have to decide to kill it. So the real decision framework here, when you're thinking about this, is this something people want? Is this something that people will pay for? Can you acquire them? Can you deliver? And then ultimately, if you're going down the venture path, is it something that has a big enough market that a venture capital would wanna invest? So those are really the kind of the key things to think about and kind of the 5 steps. At the same time, there's also 5 common mistakes you can use, and so I want to highlight these as well. And the first, you know, I've kind of hit on is, you know, trusting the, the output without validating. Like, the human has to prompt the AI, the human has to validate the idea. There's hallucinations that come in, maybe you didn't prompt it correctly. Those are all things that you as the human still have to be human in the loop and, and validating things. You know, the other thing is skipping the real customer conversations. AI can move you along, again, at a very foundational level, but ultimately customers are very nuanced. And I remember early on with my first startup, I had a clipboard and I stood in front of a grocery store and I interviewed people going in and out of the store and getting their feedback and then going back to ask follow-up questions. That took a lot of time. One of the things that I share with my students at CU is if you're building a B2B business, I expect them to talk to 100 potential customers every semester. So before they can present their data to me, I want to know that they've actually gone out and they really understand what the problem is. B2B, it's a little bit tougher. And so in that case, I have them talk to at least, you know, 5 different competitors or potential customers of that space. But nothing beats talking to an actual customer. It saves you so much time in the long run because you start building what a customer actually wants. You know, I mentioned overrelying on generic prompts. Like, you have to know how to prompt it. You know, you can enter one sentence or you can enter paragraphs. You're gonna get richer, better results the more data you can add into that. So make sure that you understand prompting and how to get those best results. But at the same time, it can be analysis paralysis because AI is so fast, it can generate so much information. You want to be able to quickly have a way to analyze the data coming back and make decisions. And so what you really want to do and what we talk about here in this talks about, you know, not perfecting DEXs, you know, don't let perfection become the enemy of the good. Again, it goes back to that, you know, what I started is get something out in the market and start shipping. So if you're not sure, build something, build a low-cost MVP, get out in the market, get that customer feedback really quickly instead of waiting till you have the perfect solution, the perfect scenario. And then lastly is just, you know, forgetting that human element. AI is a, is a tool. It, it will do what you ask it to do, but you as the human have to be the one who's involved in making the decisions of what you're going to do with that. And so those are the key things. But, you know, if anything on this list, you know, trust but verify is one, and talking to the customers, I think, are probably two of the most important things to take away from this. So when we talk about, you know, what this fast validation gets you, you know, I started at the beginning saying, you know, you want to de-risk your business. 90% of startups fail. You want to catch the flaws before they catch you. Or they cost you. And artificial intelligence allows you to do that because to the, to the box on the right, you can iterate much faster. You know, I have no problem with, with founders failing. I want them to fail fast because it saves them money. It saves me money. So test more ideas, fail faster, because ultimately if you do that, you can extend your runway. You can have, you know, more funding available to test more things in the market and really nail your product market fit. And then lastly, it's founder clarity, knowing what you wanna build. When you're talking to those customers, they're gonna tell you exactly what they need and you can build on that. So those are really kind of the tangible benefits you get of using artificial intelligence. So, you know, the key takeaways, and then we'll get into some of the actual AI tools here, is, you know, you use this 5-step framework where you're clarifying your idea, you're identifying, you're mapping the competitors, you're testing it, and then you're deciding, you know, what to do. You know, start small, test fast, you know, get out there in the market, get that user feedback. Again, remember that AI amplifies, it doesn't replace. You are still the human in the loop. You still have to be part of this process and really make the decisions. And one of the things, you know, maybe a more concrete example is I can introduce you to 100 mentors, and 50 of those mentors are going to tell you to turn right. And 50 of those mentors are going to tell you to turn left. You still have to decide as the founder which mentors are you going to listen to. AI is exactly the same way. AI is going to tell you 50 different things you can do and 50 things you should avoid, but only you know the customer, only you know the product space, and you have to make those decisions. And so it can really help amplify you, but ultimately it's your judgment and your creativity that's really going to solve this. And then lastly, you know, it's evidence over opinion. Build a culture on rapid experimentation. Again, get the product out, start testing it, and get that user feedback from them as quickly as possible. So here's probably the— you know, for the audience, probably one of the more interesting slides in the deck. I realize I saved the best for last here, is there's all kinds of tools out there for founders. So you're not just using a Gemini or a Claude or a ChatGPT or something like that. If you're doing rapid prototyping, there's a whole host of different products out there. Some you've probably heard of, some you probably haven't. You know, there's, you know, marketing and growth hacking and I'll just highlight the one at the top, AMAR. AMAR has a platform where it can build a 90-day go-to-market strategy down to the point where it will say, On Wednesday, you need to post this on LinkedIn, and here is a generated image that will go with it. Like, it can do all of that work for you very, very quickly. You know, market research, you see in some of those like Fireflies and Otter, like a lot of those, you know, will pop up in your Zoom meetings and things like that. Well, think of that as like you're capturing customer data. You can analyze that data. And so that's really the market research piece. You know, the decks, Beautiful.ai has been around for a long time. A lot of people have shared Gamma with me. Like I'm seeing lots more decks made out of Gamma. I haven't used it yet and people keep making fun of me why I haven't used it yet, but that's a very popular product. Canva is probably the number one deck I see out of my students at CU. They love using Canva in building their things. And then ultimately, if you decide to go down the investment path, and raise funds, there are other tools out there that can help you with fundraising and building your deck. And so those are kind of the highlight of the various tools. I'm sure there are many others, but these are the ones that I've used or have been recommended to me in the marketplace. So now I'm going to show a quick demonstration, and what I'm going to show is just kind of the speed and depth of the decisions that can be made here. So one thing I'm going to do real quickly is I forgot to do this earlier. Oops. But I'm going to skip to this idea. So this— let me take it out of presentation mode. I'm going to highlight this section. I'm going to stop sharing this screen. And now I'm going to share the SparkRocket screen. We're going to run this idea through the platform. So this is our platform, SparkRocket. I'm going to simply cut and paste that idea in there. It's one sentence. I could put paragraphs of information in there, but one sentence works great. And I'm going to click Unleash Your Startup Idea. And what's going to happen is in the next 20 seconds or so, I'm going to build a business plan on a page. I'm going to build a lean canvas. And so this is going to be your, your 9-box business plan on a page, and it will have your, your problem, your solution, your unique value proposition, customer segments will all be included in this. And so you see that's, that's done. So a lot of the things I talked about, I've just built in— what is that, 15, 20 seconds? So all of that information is in there. And again, I have this crazy idea, I don't know where to start, here's, here's a quick way to start. And I'll show you a quick couple of examples here. We have these customer segments for it. If I click on the customer segments, you'll see it's built out a persona for each of them, as well as the customer jobs and the customer pain points. So I'm already doing a lot of that work to understand my customers and the customer segments for you. We do the same thing, you know, we have the problem, solution, Remember we talked about competitors? Here are all these different competitors, and these are all links out to those competitive websites. Here are all the different direct, indirect competitors. And then again, the alternative solutions. You don't need to do a fitness app, you can just go walking outside. Well, that's a competitor to what you're doing. And then we also, in a similar way, talk about, you know, kind of the core features of the solution, but we want to foster that creativity in the entrepreneur. And so we give you all these different potential ways to play. Have you considered a partnership or community features, things like that. So all of that information is, is captured in there. The next thing we talked about is validation. Does anyone actually want my idea? So in the validation section, I'll give you 5 different hypotheses and how to test them. So is it desirable? You should go do a survey or do a customer interview or spend the day in the life of or all those different things. And recognizing that a lot of entrepreneurs may not know how to do some of these things, we've built this Tools section, and in the customer discovery, we will actually give you a list of questions that you can, you can start doing your customer discovery to better understand your, your customers' needs. I don't know if this will show up because it opens in a different screen, but what we'll do here is we'll build a landing page. So we'll build a landing page on the fly with your idea. That could be a step to where you go out and start testing your idea. You've built a landing page, start directing people to it? Can you start doing customer capture there? Are people signing up for your waitlist? Are they asking questions? Things like that. And then the last piece I'll show you is kind of the marketing piece, you know, mission, vision, values, brand personality, you know, lots of information here. Again, I don't know what your values are, but I've given you this framework with these 5 values. They seem like the right values for your idea, but you could certainly swap those out and put something else in. And then, you know, the biggest challenge for entrepreneurs, how do I get my elevator pitch right? Well, we're going to give you a start here with this one-liner to get you started on what your elevator pitch is, but then there's all this additional information here below that you can use to supplement your elevator pitch. And then lastly, you know, you have to come up with a startup name. In this case, it's called HomeFit, but we could call it SelectPulse, and we can change the name. Now what's nice is you don't wanna spend all your time putting all your energy into QuickPulse and find out you can't buy the domain. So we'll actually check the different domains for you and what's available. And then some of the other things that we won't go into is Insights. Insights is really understanding the overall size of the market. So think of your total addressable market, your TAM, and you can start building your SOM and SAM out of that, as well as an assessment that we'll do saying, hey, we think this is a good market, Ah, there's some question about this market. Maybe you should do a little more research. And then lastly, a team gap analysis where you can upload the, uh, the CVs or the resumes of your founding team and identify, do you have gaps within your team, uh, to build this idea? And so those are the kinds of things that you can do. SparkRocket does all of this in kind of one shot with, with one idea. But then those other tools that I exposed, like, hey, if you need a pitch deck, you know, you have these other tools, you go pitch it, build a pitch deck, or you need to do deeper market analysis than what we've provided, you can do those kinds of things. So those tools are all available. We've just created one simple framework that entrepreneurs can use. And so with that, I'm happy to jump back in here with John and start answering some questions.

Yeah, so we got a lot of great questions here. Can you just go back to the 5 common AI validation mistakes? Slide that you had. I really just want to point that one out, guys, because it does address, um, a lot of the— some of the, the questions in here, but also some of the biggest issues that we've seen. So yes, look, it's, it's— there was a couple of questions, for example, like who validates what the AI is telling you and this kind of stuff. Guys, nobody. You do, right?

That's right.

Um, I think the example you gave, Scott, was perfect where we have the same thing with, in a lot of Founder Institute programs, right? We get complaints. They're like, oh, well, this mentor told me to do this and this mentor told me to do that. And it's like, guess what, guys? There is no one perfect way to build a business, right? If there was, like, we'd all be successful. You know, there is the human element. So it's really, really dangerous in the same way to just go to one mentor and do everything that one mentor tells you to do, right? It's really, really dangerous to just do every single thing that the AI tells you to do and to just use it as a general shortcut. Okay, um, absolutely. Yeah, so that, you know, it answers a couple questions here too, right? Like, so Matthew in the chat was saying, I think that some of these can be really dangerous for founders, right? Um, you know, there were other, uh, questions again about like who validates this kind of stuff. And guys, the, the answer is you, right? You as the founder. Um, utilize these as tools, right? Don't utilize them, I'd say, necessarily as like shortcuts, um, because it's an AI, right? It's making educated guesses, uh, generally speaking, right? The LLM is looking for patterns and things like that based on a lot of whatever the training data is. So whether it's SparkRocket or ChatGPT or whatever it is that you use, guys, um, you know, you have to make sure that you still have the human element in here.

And if I can just interject there, one of the things that— it's something that comes up frequently, especially with users outside of the United States. Most of these LLMs are US-based, and they have bias in them. And there was a study, I believe it was out of Stanford, that came out last year, and there's this concept that it's called weird. LLMs are weird. And what that means is that most LLMs are westernized, educated, rich, industrialized, and democratic. And so it comes— so that's the one thing you think of. That's the perspective. That's the bias that they come out with. The challenge is that there's more data in these westernized LLMs than there are in LLMs from other countries and stuff. And so they have a richer data set, but you still have to keep in your mind, gosh, it's also got this bias towards, you know, this weird bias that they term with Western-educated, rich, industrialized, and democratic society. So something to keep in mind.

Yeah, exactly. I see a fun question here that I'd like to ask that just popped up. Chris was asking, did you put spark rockets into Spark Rockets to validate the idea?

I get that.

Does that rip a hole in the universe?

I don't know what happens when you do No, about once a month I enter a version of SparkRocket into SparkRocket and see if there are things that are missing because the LLMs are getting smarter all the time. And so I absolutely do that. Again, it's, you know, about once a month or so I absolutely put SparkRocket in SparkRocket and see what the output is because it'll help me identify perhaps a competitor or a new market segment that I hadn't been considering. Absolutely.

Yep. So we had a question here. We had a couple questions like kind of on the customer development process, guys. And, you know, we have something, for example, called the Startup Ideation Bootcamp, right? And again, there is no substitute for talking to customers, right? There is none. And I know Scott kind of prefaced it in the beginning, you have to talk to customers. These tools can help you do it more effectively and more efficiently and get more out of those conversations, identify the right customers to talk to and things like that. A lot of people in here were asking just tactical stuff on how do I get people to respond to me. At least my piece of advice there, guys, is there's an acronym, Very Few Wizards Properly Ask. And I'll put that in here. If you Google it, you know, it's not a super, super common term. So if you Google it, you'll find it's an acronym. It comes from a book called The Mom Test, which is a super, super useful book. And, you know, and I think there's like free PDF versions of this book all over the internet. But it's just a very simple way to actually reach out to people because I can tell you, And I get customer development outreach all the time, right? People are trying to reach out to me and, you know, you don't want to have an email that's like super formal asking for an interview, right? You want it to be casual. You want to make sure that you're, you know, people know that you're not trying to sell them something. You're really just trying to solve a problem that hopefully they're passionate about. So they'll take a little bit of their time, right? Scott, do you have any other, you know, pieces of advice on actually that, that mechanic of it, of of doing the outreach to people so that they'll actually kind of get on the phone with you or get on a Zoom call to talk about the problem?

Absolutely. Like, I always, you know, talk about, like, use the resources you already have. So as an example, if you are a student, you know, it's a much easier conversation because you're non-threatening to go to someone and say, hey, I'm a student at XYZ school. I'm working on a project. I have this idea. Would you be willing to spend a few minutes with me? You know, another is leverage LinkedIn, your LinkedIn connections, and say, hey, you know, I see we both follow this same person, or we're commenting on similar things. Like, you seem, you know, pretty intelligent. Like, compliment people, or leverage, like, hey, I see you went to the same college I went to, or we both worked, you know, at some company. Hey, I'm building this thing. Would you have a few minutes? To ask, like leverage those mutual connections. So I'm an alumni of the same job or the company. I am currently a student. Leverage those connections as much as you can for sure.

Yep, definitely. And a couple people here too, and some of it was related to like SparkRocket, right? Like who owns access to the IP you put in SparkRocket. I'm pretty sure the answer to that question, but I, But it's also the larger question that people are saying of like, what about, you know, the ideas you're talking to LLMs and stuff? Like, what kind of protection do you have, if any?

Yeah. And so for us at SparkRocket, it's two things. One is that we call it at the SparkRocket level, we're very clear in our terms and conditions and things like that, that you own the IP, you own the idea. We are not taking that idea. We're not stealing that idea. And then our relationship with, with OpenAI, which is, which is the backend, we find OpenAI provides the best results for us, is they, they do not train their models on any of the ideas that we send them, and then they delete them within 30 days. So not only are they not training it, they actually delete it out of the system. But yeah, no, number one question I get from a security and IP perspective, for sure.

And I know you mentioned Reddit in there, and there were a couple people who asked also about Reddit, you know, like who it was of the Nunca, and hopefully I'm saying that name correctly. Apologies if I'm not, you know. And I know that this is a strategy that a lot of people have used also, is to— because Reddit is essentially this, it's a public forum, right, of everybody. So like you said, it's not just the rich educated, like, you know, LLM, right? Reddit is actually the voice of the people in a lot of areas. How have you seen people leverage, you know, AI tools and then going into Reddit to kind of identify, you know, maybe some problems or validate some problems? And because I know I've heard a lot of people talking about it, but to be honest, I haven't myself seen kind of that process in action?

Yeah. And so the first thing I would say is on Reddit, go into, you know, there's a variety of different channels around startups, around product-market fit, you know, go into those channels and you can learn from how others are using it. But, you know, one of the biggest things is use it as a sounding board. Go in and ask a question like, hey, I'm starting this business, I have this idea. This is what it is. I would love feedback on that and be able to capture that feedback. And depending on how much you capture, it's literally you just cut and paste, you throw it into an LLM, and you can do that in, in multiple different channels because you may be, um, you know, uh, uh, talking to, to parents in one channel, um, you may be talking to, to dog owners in another channel. You know, I want to make a dog walking app. Okay, well, let's go to, to the dog walking people. Let's go to young busy professionals. There's, you know, some group about young busy professionals who want to exercise. Go in all the different groups, pose your questions, and then wait till the feedback comes in, and then you use the large language models to start analyzing that data.

Right. Okay, and guys, we have a couple more minutes here, so please don't be shy. Throw any other questions that you have into the chat here, and I'm trying to sort through these, sort these, sort through these as well. Um, let me see. So, okay, so, uh, Ricardo and a bunch of other people, right? It's like the, the perfection is the enemy of progress kind of, um, mentality, right? Like, when would you say is it too early, like, to, to launch something? Um, is it ever too early? And, you know, especially if people are just looking, trying to validate like one specific feature or one specific, you know, use case, you know, I know you guys have like a landing page builder on there. Is that sort of the standard process that you recommend is just, hey, start with the landing page, then validate it, then maybe decide to build from there?

Yeah. I mean, my, my kind of short glib answer is it's, it's never too early. Get, get something out there as fast as you can, but in reality, You want to have something that is viable enough that you can get legitimate feedback on. And that can be a mockup. Like, literally, it can be on pen and paper. You can build something like that. You don't have to actually build software. You know, in this last semester when I was teaching, you know, I had all these teams, and I said, you know, you need to build a mockup. Like, one was— a newfangled butter dispenser. Like, you know, it's butter, it's pretty basic. Don't overthink this. Literally get some cardboard, fold it up into the format you want, you know, color it yellow or put construction paper on it. Like, just get the idea out there. 'Cause what happens is a lot of people that you're talking to are very concrete. They need to see something, you know, in detail. They struggle with an abstract idea, especially when it's new and novel. So the sooner you can get something in their hands, literally or figuratively, that they understand. One of the things we can talk about are kind of analogies, and you've probably heard, oh, we're the Uber of, or we're the Netflix of, or something like that. Using those things that make it a little bit more tangible. But the sooner you can get that feedback, and like, that can be a conversation as part of your elevator pitch. Hey, we're the Uber of water bottles. You know, what does that mean? Oh, I can call up and order water bottles to my house. People understand that. And then they're like, That's the dumbest idea ever. Oh, that's the greatest idea ever. Why is it the greatest idea? Oh, because of this. Oh, and so earlier is better is what I say.

Yeah, and Jules was saying in here too, the quote is, if you're not uncomfortable with what you sent out, you sent it too late. I think like, I think it was, what is it? Reid Hoffman, you know, super famous investor, one of the founders of LinkedIn. I think he said, you know, if you're not embarrassed by your first product, then you've waited too long.

100%.

Um, it is all a learning experience, guys. Trust me, embarrassment is, is, is a part of entrepreneurship. So, um, it's all parts of, of learning, um, and, and figuring out all of this stuff, uh, along the way. And failure is also the most effective way to learn, right? When I look at mindset shift that most successful entrepreneurs have, they're not afraid of failure. They, they reframe failure and they leverage it as a utility, right? Because they understand that failure is the most kind of valid— you know, it's a valid way to learn, right? As you start moving forward. So another question here too, and again, this kind of gets back to before, you know, the topic that we were talking about before where people are saying, hey, again, isn't this stuff kind of dangerous? Can— so let me, let me see this— could the creation of this canvas limit the creativity of a founder where, and, you know, kind of put the founder more in the, in the position of an editor instead of being a creator, right? Um, and it's interesting because I, I was at an AI conference not too long ago and they were starting to talk about how, at least in like the B2B realm at large enterprises and stuff, that people are going to have start thinking of themselves more as creative editors, you know, managing different agents and things like that throughout their business. But, you know, I guess, and yeah, maybe just to wrap things up here, the reason why I had you bring up that slide was because I thought that was the biggest thing, right? I see a lot of entrepreneurs when they start to use AI tools, they they, they essentially kind of use it almost to like cheat, right? They're not using it as a utility to help them get the right insights, collect more data so that they themselves can then make the right decision based on AI as a data point, this conversation I had yesterday as a data point, right? All these other data points, right? They're almost using it as like a cheating tool, as if they're like in university or something, right? And they're like submitting a paper. Yeah. Do you have any feedback too on just how to, how to try to get entrepreneurs out of that mindset? Because that, if people treat it like that, then I do think it becomes dangerous. Right. But, but again, it's like that's not the AI's fault. That's the founder's fault.

100%. And that's one of the things we try to build into our platform in particular is to foster additional creative ideas because that's the last thing we want. Is for you to get locked into an idea and believe it's true. And that's really where, you know, I just, I keep coming back to you as the founder. It's AI-human collaboration. Don't believe the AI out of the gate. You have to do your research. And that's really where the nuance of talking to actual customers comes in because they're gonna tell you, you know, as, you know, as horrible as this sounds, like no one wants to hear that their baby's ugly, but sometimes someone needs to tell you your baby's ugly. And that's exactly what it is like building a startup. The startup is your baby. And sometimes people just say, you know, this doesn't make any sense. An Uber for water doesn't make any sense. Why would you do something like that? Well, that's great feedback. That's great data. So definitely I would say it's twofold, is keep in mind that you are in control. The AI doesn't control you. You control the AI. And get out and talk to people, talk to customers, because they will tell you the truth.

Yeah, exactly. It's still— it'll reduce some time, I think, to start a business, but you got to get out there. There's no— there's— No, there's no substitute for talking to customers. And it's funny because I even see now, especially like in the B2B sales realm, now you're starting to see these tools where people can like talk to ICPs, right? So it's like, oh, I have a product and I'm going to sell to, you know, a buyer at this type of company, right? So they're really trying to take the humans out of the equation as much as possible. But I would just for everyone listening today, resist to that urge. Okay, um, you know, maybe in a couple of years or so, maybe they will, but at least for what I'm seeing now, and especially as an entrepreneur who— your superpower as an entrepreneur, when you're— especially when you're first starting out, you're trying to build something where there's a lot of incumbents, right? And you're trying to find your niche that you can really service really well, right? Your competitive advantage is going to be your ability to understand your customer better than anybody else. And if you're just trying to outsource that to AI, then that's not going to be an advantage that you have. AI is a tool to help you do that. And definitely, guys, as I've said, you know, we don't get any kind of rev share or anything like this off of our partners. There is a discount code there for SparkRocket just because I've seen a lot of these tools come out, these AI tools to help entrepreneurs during like the ideation and validation part of the process. This is the most impressive one that I've seen so far, and we have actually integrated it into our Startup Ideation Bootcamp as well because we saw so many founders in FI programs starting to use it. Okay, well, thank you so much, Scott, for taking the time to join us today. To everybody there, we— you will be getting a copy of this deck. I know there was, like you said, Scott, that slide was very popular and people are like, give me the list, give me the list. You'll get a copy of this deck this video within the next 48 hours, just as soon as we can do all of the, uh, the process around it. Um, the, the discount code, maybe it's pinned to the top of the chat, but depending on what interface you're using, maybe you don't see it, so I just pasted it in there as well. All right, thank you, Scott. Thank you everybody for joining. I hope you all have a great rest of the day.