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

David Bressler

• Built an AI spreadsheet and data app originally called Excel Formula Bot and later called Formula Bot.
• Started the product while working a full-time job and helping prepare for the birth of his youngest child.
• Used a six-week window to build something around a problem he personally understood.
• Had no coding background and learned most of what he needed from Bubble documentation and YouTube.

Formula Bot

• Helps people work faster and smarter with data.
• Includes AI inside Excel and Google Sheets.
• Offers formula generators that translate text into spreadsheet formulas.
• Includes a data analyzer where users can upload data and ask for answers, analysis, charts, and modeling in natural language.
• Grew to roughly 750,000 users in about a year and a half.
• Reached about 5,000 paying users for unlimited access.
• Reached about $226,000 in monthly recurring revenue.

Origin And MVP

• Got the idea after repeatedly helping junior analysts with Excel questions at work.
• Asked OpenAI for name ideas and found the Excel Formula Bot domain available.
• Searched online for similar tools and could not find the same product.
• Built the MVP in Bubble because coding it would have taken too long and hiring someone was too expensive for an unproven idea.
• Launched the first version as a no-code app in a couple of weeks.
• Later moved from fully no-code to about 5% to 10% coded as the product hit no-code limitations.

Launch And Early Users

• First showed the tool to people at his full-time job, who were surprised it had only taken weeks to build.
• Posted it on the Excel subreddit with a simple title and link.
• Became the top post of the day, then the top post of the week.
• Posted it on the Internet Is Beautiful subreddit after a commenter suggested it.
• Got around 10,000 upvotes, thousands of comments, and heavy traffic from that post.

Early Monetization

• Quickly spent about $5,000 on OpenAI API costs after the viral launch.
• Added a Stripe donation link and made a few thousand dollars back.
• Tested ads and sponsorships, including an ESPN Excel competition ad.
• Found ads were not worth the hassle as a long-term model.
• Added logins and a paywall a few months later.

Product Evolution

• Started as an Excel formula generator and expanded into additional generators.
• Initially struggled to explain a broader vision to venture capitalists beyond more generators.
• Talked with coworkers who worked with data and spreadsheets.
• Learned from users who emailed Excel files and asked for help beyond formula generation.
• Offered to help those users on calls in exchange for feedback about the product.
• Used those calls to shift the product toward an all-in-one platform for data analytics work.

Competition And Positioning

• Saw many similar Excel formula generator sites appear within weeks.
• Faced competitors that copied the product, interface, and naming.
• Realized a simple input-output generator would become a race to the bottom.
• Heard from Microsoft about building an Excel add-on, then worried Microsoft could turn the idea into a built-in Office feature.
• Recognized ChatGPT as a churn driver because it offered free unlimited usage.
• Initially competed on convenience and customization inside spreadsheets.
  • Users could generate formulas without leaving Excel.
  • Users could set language preferences for spoken language and Excel language.
• Shifted the long-term vision toward features that could not be easily replicated by ChatGPT or Microsoft.

Lessons And Advice

• Started by solving his own problem.
• Used no-code because it let him move quickly without outside funding or a developer.
• Turned customer support requests into product research calls.
• Learned that viral usage can create urgent infrastructure costs before monetization is ready.