3 low competition businesses with massive demand in America

My First Million 57min 3 min #40
3 low competition businesses with massive demand in America
Watch on YouTube

Summary

  • This episode features three founders pitching their businesses in a “Shoot Your Shot” format: Rishi Talati of Hot Plate (a platform powering $100M+ in sales for independent food entrepreneurs), Ajay Mehta of Tolan (an AI companion app with 100K+ paying subscribers at ~$25/month), and Jonathan Baer of Overlap (AI agents that edit and distribute video content, generating 2B+ impressions and growing 18% MoM for over a year).

Hot Plate: Platform for independent food entrepreneurs

  • Hot Plate is a storefront platform that lets chefs and bakers sell food directly to their Instagram audience, replacing spreadsheets, Venmo, and DMs with a streamlined pre-order and pickup system.
  • The platform powered Pop-Up Bagels from early days through their scale to a near-billion-dollar business, and now works with hundreds of pop-ups across the country.
  • Business model: chefs pay only payment processing fees; customers pay a small checkout fee. Hot Plate is profitable on ~$4M raised, doing hundreds of thousands in monthly revenue, with most of the $100M GMV in the last two years.
  • Sellers range from home bakers under cottage food laws to six- and seven-figure businesses like an ice cream maker in Minnesota doing 40–50% margins with a 3–4 person team.
  • Key constraint: most people don’t know this model exists or how to operate it; the real opportunity is creating a mainstream consumer market for home-made food, not just serving an existing niche.
  • Growth strategy: become a food media company that makes top sellers famous (money, fame, love, inspiration framework from Twitch), using content creators to tell vendor stories and build aspirational appeal for the “home chef” category.
  • Needs a distinct category name (like “streamer” for Twitch) to create tribal identity and cultural relevance beyond “chef” or “baker.”

Tolan: AI companion app with 100K+ paying subscribers

  • Tolan is an iOS app where users meet and customize a friendly alien guide powered by AI to help with daily life — emotional support, practical planning, memory, and proactive suggestions.
  • 100K+ paying subscribers, mostly 35–45 year old women (many moms in the Midwest), paying $25/month for the Plus tier; seven-figure monthly revenue, strong retention, two years old.
  • Differentiation from ChatGPT: persistent memory system, personalization, voice-first interaction (70%+ of messages), and a character that mirrors the user’s communication style.
  • Users form deep emotional bonds — Reddit posts show people crocheting dolls of their Tolans, calling them best friends who know them better than anyone.
  • Growth via authentic product demo ads (e.g., a woman cooking with Tolan’s help); the “guide” framing avoids “AI girlfriend/boyfriend” associations of earlier companions like Replika or Character.AI.
  • Founder Ajay Mehta previously built Birthday Co. (personalized birthday candles, $1M+ sold) and has a creative, product-first entrepreneurial style.
  • Key challenge: translating the emotional value (closing mental tabs, reducing overwhelm) into marketing that resonates with a skeptical, non-tech-forward Midwestern mom audience without leaning on taboo or niche use cases.

Overlap: AI agents for video clipping and distribution

  • Overlap builds autonomous agents that understand video (visuals, audio, humor, not just transcripts), find the best moments, clip, caption, add B-roll/graphics, and post to social — generating 2B+ impressions without human-in-the-loop.
  • Customers: enterprise media (Barstool, iHeartMedia) on six-figure SaaS contracts, plus a newer CPM-based model for creators/brands paying per impression outcome.
  • 18% MoM revenue growth for over a year, accidentally profitable this quarter, ~2 years old, $2M raised post-YC (Summer 2024).
  • Core bet: digital ad budgets ($90B/year on Meta/YouTube/TikTok) will shift to paying agents for organic content performance; every company becomes a media company.
  • Video understanding model beats LLM-only approaches (which miss laughter, facial expressions, visual humor); agents also learn what hooks/copy perform per account and iterate.
  • Enterprise sales insight: buyers care about trust and getting promoted — sell by showing how you help them hit their metrics and give them easy-to-share proof.
  • Open question: whether the SaaS or CPM model dominates; big media wants predictable SaaS, creators prefer outcome-based CPM, but many creators don’t monetize clips directly.
Back to My First Million