Starter Story

I Built a $1M AI App [No Code]

16 min
•Feb 11, 2024over 2 years ago
Listen to Episode
Summary

David Bresler built FormulaBot, an AI-powered spreadsheet tool, to $1M+ valuation in 18 months using no-code development while working full-time. He shares how he launched virally on Reddit, navigated $5K in API costs, and evolved his product to compete against Microsoft and ChatGPT by building irreplicable features.

Insights
  • No-code platforms enable rapid MVP development (2-3 weeks) without coding experience, democratizing SaaS creation for non-technical founders
  • Viral distribution through niche communities (Reddit) can generate thousands of users quickly, but requires monetization strategy to sustain growth
  • Direct customer conversations reveal product gaps that generic feature requests miss, driving strategic pivots from simple tools to comprehensive platforms
  • Competitive moats shift from convenience to functionality when larger competitors enter; building irreplicable features is essential for survival
  • Bootstrapped founders must balance personal life with execution intensity; early traction provides motivation to sustain through high-cost scaling phases
Trends
No-code SaaS as viable path to million-dollar businesses for non-technical foundersAI-powered productivity tools creating new SaaS categories in established software marketsRapid competitive cloning of successful no-code products forcing differentiation through feature depthLarge tech companies (Microsoft, OpenAI) integrating AI features, pressuring standalone tools to build defensible moatsCommunity-driven viral marketing outperforming paid acquisition for B2B SaaS in early stagesChurn driven by free alternatives (ChatGPT) forcing paid SaaS to emphasize convenience and customizationAPI cost management as critical financial challenge for AI-dependent applicationsCustomer support conversations becoming primary product development input for bootstrapped teams
Topics
No-code development platforms and toolsAI-powered spreadsheet and data analysis softwareViral marketing and Reddit community growthSaaS monetization strategies and paywall implementationCompetitive differentiation against large tech companiesOpenAI API cost management and billingProduct-market fit through customer interviewsExcel add-ons and Microsoft partnership opportunitiesChatGPT competition and feature parityBootstrapped SaaS scaling on limited budgetWork-life balance while building side projectsFeature prioritization and product roadmapChurn analysis and customer retentionNatural language processing for data analysisMulti-language and localization strategy
Companies
Microsoft
Reached out twice to partner on Excel add-on and AI integration; announced OpenAI partnership, creating competitive p...
OpenAI
Provided API for FormulaBot's AI features; David consulted them on product naming; API costs became major financial c...
Bubble.io
No-code platform David used to build entire MVP; described as easiest no-code solution with strong YouTube tutorial c...
ChatGPT
Free competitor causing 10% churn; forced FormulaBot to differentiate through convenience, customization, and irrepli...
Google Sheets
One of two primary platforms (alongside Excel) where FormulaBot's AI features operate
Zapier
Compared as functional reference point for FormulaBot's Data Analyzer product capabilities
GoDaddy
Domain registrar used to purchase ExcelFormulaBot.com domain name
Stripe
Payment processor used for initial donation link and subscription billing
Google Ads
Advertising platform David tested for monetization; found insufficient ROI compared to subscriptions
People
David Bresler
Founder of FormulaBot; built $1M+ AI spreadsheet tool with no coding experience while employed full-time
Pat Walls
Host of Starter Story podcast; interviewed David about his entrepreneurial journey and product development
Quotes
"Really took me just a couple weeks. Came down to learning all of his content mostly through YouTube."
Pat Walls (describing David's development process)•Early in episode
"I blew through like $5,000 in days. I had a decision to make, right? It's either shut it down or keep going."
David Bresler•Mid-episode, discussing API costs
"My vision moving forward is to build something that cannot be replicated both in chat GBT as well as Microsoft."
David Bresler•Late episode, discussing competitive strategy
"The power of no code. Because if you were to code it, it didn't take much longer."
David Bresler•Discussing MVP development speed
"Early on, it really was convenience and customization. But that can only take you so far."
David Bresler•Discussing competitive differentiation
Full Transcript
This is David. He built an AI app that is now valued at over a million dollars. And the crazy part is he built it all with no code. He invited us into his house to show us exactly how he built this thing with a family, a full-time job, and just working a couple hours a day. Really took me just a couple weeks. Came down to learning all of his content mostly through YouTube. David launched his product using a viral marketing strategy that got him thousands of users in just a few days. But what goes up must come down. I blew through like $5,000 in days. I had a decision to make, right? It's either shut it down or keep going. In this video, he breaks down how he bounced back, how he made his first dollar, and how he plans to build a business that won't be replaced by chat GBT. My vision moving forward is to build something that cannot be replicated both in chat GBT as well as Microsoft. This is the breakdown on how a normal guy turned one idea into a million dollar business on just a few hours a day. I'm Pat Walls, and this is Starter Story. David, thanks for having me. Thanks for coming out ever. I love your story. Tell me a little about who you are and what you built. Sure. Dave Bresler built what was called Excel for RealBot. Now it's just called for RealBot. Microsoft gave me a season to this. I have a few months to go or so. And so it's a software that really helps people work faster and smarter with data. There's really three main products, AI and spreadsheet. It's kind of like chat GBT inside both Excel and Google Sheets. Four-wheel generators that translates text into formulas with the help of AI. And then the last product where a lot of my ton of resources have gotten into this called the Data Analyzer, where it's kind of like Zapier. You can upload your data and then through Natural Language, get answers, analysis, charts, modeling, all done by just typing in some words. The application has been around for about a year and a half. And during that time, there's now roughly about 750,000 users. And of those 750K, there's about 5,000 people that pay for unlimited access to the tool. And during that time, it's reached about $26,000 a month in re-crum. As I understand, you started FormulaBot while you had a full-time job. Tell me a little bit more about that story. It was July of 2022. I had this opportunity and the opportunity was my youngest one was about to be born. And so I knew I was going to have about six weeks to do something. And I've been working a full-time job since I graduated college 2011. And I've always kind of had that entrepreneurial neck. Basically, I said, okay, I've got six weeks. Let's build something. People always say, scratch your own itch when I solve your own problem. And my problem was I was getting so bogged down with all of these junior Adols. They'd come knocking on my office door like, hey, can you help me with this formula? Can you help me with that? I kind of thought, okay, my itch is helping people with Excel. While chat GPT wasn't around then, the actual technology was. I actually asked the OpenAI application. I said, hey, I'm thinking about building this app. What's a good name for it? And one of the names I've suggested was Excel Formula Bot. Same thing you do, of course, is you go to GoDadding. It was available. And then I was like, let me just see if someone's built this yet. And I scoured the internet for hours. Couldn't find it. I'm like, this is it. I'm on the something special. It was tough trying to find that balance between personal life. You know, I can't help but to a certain extent, stay on the grid for my full-time job, especially in the senior position that I was in. And so while I wasn't working, I still had access to my phone. People were still asking me questions here and there. But then in the back of my mind, I'm like, I've got this idea that I have to execute on. I tried to spend as much time with the family, especially early on, but I couldn't help but find myself kind of stepping away for a little bit. Got to coffee shops where it might be and just kind of put my head down and execute. And there were definitely a lot of nights where I only got a couple hours of sleep. My wife at that time thought I was crazy, especially because I didn't really have a clear path. I just don't want to build this thing. So for what I understand, you have no experience coding. How did you build and launch something like this? I'm not a coder by any means. If I see it, I kind of know, okay, that's JavaScript, that's HTML. And I knew for what I wanted to build as an MVP, it would take a pretty significant time to build it with code. And I at that time wasn't making any money. I didn't want to hire anyone because I just had an idea. And so that's for where I was really exposed to the node code industry. I was exposed to bubble.io. That was really the only application that I really dove into to build because it was the most well-known. And based on the research that I had done, that the easiest of all the node code solutions. The first place I went was the bubble official website for just kind of reading some documentation. But there's so much documentation out there, which is a good thing. Kind of where I got hung up was very isolated issues that I was trying to solve. That's really where YouTube University came in. There are some awesome bubble developers out there that produce such high quality content on YouTube. And it's all free. At that time, the whole app was in bubble. It was 100% node code, probably up until maybe four or five months ago. It went from 100% node code to now it's about, I would say like five or 10%. As you get more experience with node code, you start to understand some of the limitations. There's not a lot, but there are some limitations. David is the perfect example of how someone with zero experience turned one idea into hundreds of thousands of dollars. But that comes with knowing the right information and finding the right problems to solve. Now, imagine there's a place that gave you all this. The problems to solve, the blueprints to solve them, and the strategies that turn civil ideas into million-dollar businesses. Well, at Starter Story, we have a library of over 4,000 case studies and business idea breakdowns where you can access this, all backed by data from real entrepreneurs. So if you're serious about building a profitable side project and joining a tight knit community of founders, head to the first link in the description and we're going to give you 52 micro sass ideas, just like David's, so you can get started on your journey. Much love. Enjoy the rest of the video. Peace. All right, so you have this MVP that you built from scratch. How do you get your first customer and how do you get a ton of people to use it? The first people I told were the people that were at my full-time job at that time and they're like, oh my god, this is amazing. They thought it took me months to build this thing. It really took me just a couple weeks and that really kind of speaks to the power of no code. Because if you were to code it, it didn't take much longer. And so my next step was I need to tell more people about this tool. And so naturally the next place I went was Reddit. Went to the Excel subreddit where there's probably several hundred people that follow it. Reddit title was AI Excel Formula Generator. I think I mentioned it was free and it was just a title and the description was just a link. Minutes later, it was like the top post of the day. Turned into the top post of the week. It blew up as much as like something can blow up within the Excel community. A couple of days after that, someone in the cut it section said you should post this on the internet as beautiful subreddit. And so I posted it there. That was like the thing that put the app over the edge and made it go viral for months. And so I posted it there. It got some 10,000 upvotes, thousands of comments. You know, I got thousands and thousands of people visiting the site using the app. All right, so you have thousands of users. How do you start making actual money from this? The next thing I did was I went to the OpenAI website. What's the billing section? It was just like, oh God. I blew through like $5,000 in OpenAI API costs in days. I had a decision to make, right? It's either shut it down or keep going. And so I accrued this huge bill over that time. I recouped some of it back. I put up a link, like a stripe link where it's just like, hey, my name's David. I built this app. It went viral. Give me a deal. Feel free to donate. So I made a few thousand dollars back. I was super scrappy of like trying to make something back at that time. It was pretty poetic, especially probably at its peak of virality. The host from the eSports ESPN Excel competition. I didn't say it correctly, but something like that. But it's basically like extreme sports, but it's for Excel. The host reached out to me and said, hey, we have our competition on ESPN in a couple of weeks. Can we run a beta ad? I'm like, sure. I'm like, and at that time it was like, maybe that's how I'm before even thought about subscriptions. I'm like, maybe that's how I can make money, at least in the interim, is just running ads. So I ran some Google ads, made a few bucks here and there. It just definitely wasn't worth the hassle. But I ran an ad for ESPN for like a week and it drove a ton of clicks. And it was actually pretty funny because the guy asked me to design the creative. And like, look, dude, like you've seen the website, especially at that time, it was a single page piece of crap. It wasn't creative at all, but I mocked something up for him. He said, okay, works for me. And so some of the donations that came in, a couple of sponsorships, it gave me the motivation to kind of keep going and not just completely shut this thing down. And so then a few months later, lunch with a paywall, logins, and the whole nine yards. Let's talk about product, talking to customers. How have you figured out what features to build? Early on, just kind of scratched my own itch and what would bring me value. And what started off as just an Excel formula generator turned into different generators. I was really just the first product. It was actually interesting because at that time my site was blowing up and I took a few phone calls with VCs. They would ask me, you know, what's your vision? What's your broader vision? And I would just say, Lord generators. And they're like, okay, I never got a call back. I didn't have a broader vision. That's where I started talking to people that I work with who it's required to work with data and spreadsheets, started asking them and see what they liked, what they didn't like. But occasionally, and this is where I think I would get the most value, talking with customers. A lot of them email me basic Excel questions that couldn't necessarily be answered through an Excel formula generator. They would literally email me their Excel files. You're like, hey, your Excel formula generator couldn't do this. Can you help me with this? And it wasn't even to like build a formula. It was to like build a report where I have no domain knowledge of what they're trying to accomplish. I would probably get one email a week of someone just sending me a file. And so I thought, okay, you know what, even if they're not a paying customer, I'm going to take this. But guess what? I'm going to do it with you on a phone call and I'll do your work for you. But every time I do it, they have to hop on a call with me and they have to ask questions about what they like and what they don't like about the product as well as some ideas that they have. And so the culmination of kind of scratching my own itch, bleeding on my network and having these kind of impromptu phone calls kind of revolutionized the product away from just being an Excel formula generator to just kind of being this all in platform for anything that you're doing with data analytics. Building a SaaS product, software product, there's probably going to be a lot of competitors coming in. Tell me about how that works for you. Took me a few weeks to build this app and a few weeks after that, I think there was probably 10 other Excel formula generator sites. And the frustrating part was they all sounded the same, right? At that time, Excel formula bot, Excel formula generator, and then other what was called Excel formula later, they all sounded the same. And when you'd go to Google and you'd search mine, you'd see the other couple ones right there. So at that point, it was really a race to the bottom. And my prices, especially early on, it was $3. I was getting pretty frustrated. Just all of the competition that was coming in and literally like carbon copying my exact application, not only like what it does, but also the front end, the name. And then I just kind of reached a point where I knew I was going to have to build something that isn't just the simple input output box. A few months later, Excel formula bot was getting more and more popularity. Microsoft actually reached out to me. They reached out to me twice, one for a system business. And the other one, it was someone that was on their add-ons team. Excel has thousands and thousands of partners that they work with to build an add-on that supplements what you can do with Excel. They reached out to me and said, hey, we want to take your Excel formula bot and add it within our add-on store and let us build it for you. I was like, sure, how much does it cost? They're like, we'll do it for free. I'm like, this is awesome. And then as they were building it, I was like, wait a second, they're just going to hijack this idea and build it as one of the many features in their suite. A couple of weeks after that, they had announced their partnership with OpenAI. Everyone's so excited for AI being integrated in the Office suite. And I'm like, I am so gross. And so that's where it kind of went back to me being my own competition, where I had to really take a step back and think through what is this product going to be once Microsoft embeds this product within their suite. I've even talked about ChatGPT. ChatGPT, especially at this time as well, it was free. And it still is free for unlimited usage. At that time, you had to pay $299. Now you got to pay $9. It's going to work its way up $299 to $699 to $9. And you have to pay for unlimited usage, whereas ChatGPT has always been free. And so I have about 5% churn. 10% of the reason as to why people churn is because of ChatGPT. And so early on, before a lot of the newer products, especially the data analyzer, you can upload your data in top with it. The competitive advantage I had over ChatGPT was really convenience. Generate formulas within Excel. You don't have to leave your spreadsheets. You don't have to open up your browser. You don't have to open up ChatGPT. And right at this long prompt, you can do it all there. Convenience and customization. When you sign up, you can put in your language, your speaking language. You can put in your language for what you prefer for Excel. Because there's a lot of people out there that speak in Spanish, but they prefer the Spanish version of Excel, and then also vice versa. Again, Excel is used by a billion people in the world. Excel for the bot is used by every single country and thousands of languages. So early on, it really was convenience and customization. But that can only take you so far. And so more so recently with the addition of the new features and new products. It's moved from that to you can no longer do that in ChatGPT. And eventually, when Microsoft really just this past week builds AI into their own tool. Right now, my vision moving forward is to build something that cannot be replicated both in ChatGPT as well as Microsoft. Hey, Pat from Starter Story here. Thank you guys so much for watching. I really hope David's story inspires and motivates some of you to get started on your own thing. If you're curious about doing something similar, but you're still looking for an idea, well, right now you can download our deep dive SaaS report for free. It breaks down 52 different Microsoft ideas and tons of other stuff you'd want to know. Just click the first link in the description. And if you're serious about building it, then consider joining Starter Story and we'll help you do that much love. And I'll see you guys in the next one. Peace.