Frequently Asked Questions About Greg Isenberg Startup Opportunity Scanner
22 answers covering everything from basics to advanced usage.
// Basics
What is the difference between a niche and a product in this framework?
The niche is the audience you serve — GERD patients, creative entrepreneurs, lonely 50-year-old dads. The product is the specific thing you build for them. The framework treats the niche as a long-term marriage and the product as a first date. You commit to the audience and iterate the product, because a great niche survives multiple product pivots but a bad niche kills a great product.
What inputs do I need to run the Startup Opportunity Scanner?
You only strictly need a category of interest — a broad market or trend like health tech, elder care, or pet products. Optional inputs that sharpen the output include your unfair advantages (skills, networks, lived experience), a target audience hypothesis, and any willingness-to-spend signal showing the audience already pays for related problems.
What are the main product categories this framework maps niches to?
Five main ones: Action Apps (agent-first mobile for repetitive digital workflows), Community/IRL Third Spaces (for loneliness and shared-interest pain), Elder Tech (for underserved 65+ users), Creator Media (AI Native Media Company or live unscripted shows), and Vertical Personalized Health (for specific conditions). One niche may support multiple product types — the framework tells you to note all of them.
What are Third Spaces and why are they an opportunity?
Third Spaces are environments that are neither home nor work — communities, clubs, Discord servers, event spaces — that solve the loneliness epidemic by giving people somewhere to gather and do things together. They're an opportunity because loneliness is a sharp, growing pain, and niche-specific third spaces for underserved audiences (dads, hobbyists, entrepreneurs) can monetize richly through the Free + Premium Stack.
// How To
How do I start when I only know a broad category like 'health tech'?
Start wide and enumerate without evaluating. List every sub-niche: gut health, elder mobility, pet health, longevity, women's hormones. Use the CVS Shelf Heuristic to surface pain-point-dense verticals. Only after listing all obvious sub-niches do you apply the 'Fish Where the Fish Are' filter and the three-question Niche Qualification Test to narrow down.
How do I find the acquisition wedge for my first 1,000 users?
Match the wedge to the product type. For Action Apps, reverse-engineer a TikTok or Reel dramatizing the pain — someone doing the task manually and suffering. For community/IRL, start a niche Discord, subreddit, or Facebook group for the exact persona. For Elder Tech, run underpriced Facebook ads. For AI media, study a successful AI-native creator in an adjacent niche and replicate the format.
How do I apply the Huberman's-sponsors shortcut?
Identify what's already working for humans — supplements, blood testing, smart devices, communities — then apply the same model to an underserved adjacent group like pets or elders. If blood testing works for biohacking humans, ask what a personalized health product for aging dogs or 70-year-olds looks like. It's a reliable ideation shortcut for spotting proven demand in new audiences.
How do I know if a pain point is sharp enough to build on?
A sharp pain point is something the audience already spends money to patch — visible in pharmacy shelf density, existing subscriptions, doctor visits, or competitor revenue. Vague discomfort isn't enough. If people are only mildly annoyed, they won't pay. Look for evidence of active spending, which is the third question in the Niche Qualification Test.
// Troubleshooting
My first product failed — should I abandon the niche?
Not necessarily. Ask: could you pivot to a different product for the same niche? If yes, your niche selection was sound and you should iterate the product. Abandoning a great niche because the first product flopped is the most common pitfall — marrying the product instead of the niche. Only reconsider the niche if the business only ever works with that one specific product.
I'm worried a fully automated app will feel low-value to users. Is that a real problem?
No — this is a misread called mistaking set-and-forget for low engagement. Humans always seek the path of least resistance, so convenience to the max is the goal, not a weakness. An Action App that quietly does work on the user's behalf and only surfaces the 2–3 decisions needing human judgment is more valuable, not less.
I don't think Facebook ads work — my peers aren't on Facebook. Should I skip them?
No, that's a peer-group blind spot. The 50–65+ demographic is highly active on Facebook and the ad inventory is underpriced because most builders assume it's dead. If your niche is older audiences or Elder Tech, Facebook ads are one of the strongest and cheapest acquisition channels available. Don't judge the channel by your own social circle.
My idea is a generic 'AI employee platform' — why isn't it getting traction?
Because it's horizontal when the opportunity is vertical. 'AI employee platform' competes with everyone and speaks to no one. Go deep on one job title — an 'AI junior YouTube producer' or 'AI junior podcast editor.' Build the Jobs-to-be-Done Stack for that role and position as 'Juniors,' handling menial work rather than claiming to replace senior talent. Specificity is the moat.
// Comparisons
How does this framework compare to Y Combinator's 'make something people want'?
YC's advice is directionally correct but leaves niche selection to intuition. This scanner operationalizes it: it tells you where to look (underserved 45–65+ audiences, dense pharmacy shelves), how to validate (three-question test), how to position (verticalization), and how to monetize (Free + Premium Stack). It's a repeatable filtering system rather than a mindset, especially strong for solo builders without a network of design partners.
How does an Action App differ from just adding AI to an existing app?
An Action App is agent-first — the core UX is agents doing things on your behalf, and the interface shows what was already done, surfacing only decisions needing human judgment. Bolting AI onto an existing app keeps the click-and-stare UI and adds a chatbot feature. The framework flags bolt-on AI as a pitfall; the real opportunity is reimagining the whole category around agents.
How does AI-native media differ from AI content slop?
AI slop is high-volume, low-quality content pumped out for reach. An AI Native Media Company bets on top-1% quality in a specific niche, with a human in the loop and the AI-assisted nature disclosed to the audience. The goal is a loyal, high-quality audience you later monetize with products or apps — not maximizing raw output volume.
How does this compare to building for trendy young demographics?
Building for the 13–35 crowd feels exciting but means competing in crowded water with lower disposable income per user. This framework's 'Fish Where the Fish Are' principle points you toward 45–65+ adults, niche hobbyists, dads, and condition-sufferers who have more money, less competition, and deeper loyalty. The contrarian audience choice is often the entire edge.
// Advanced
How do I use the Jobs-to-be-Done Stack to build a product roadmap?
Pick the winning niche and the specific job title within it, then ask Claude or ChatGPT to list the 30–50 jobs-to-be-done for that persona. That list becomes your roadmap. Build agents to cover them incrementally — start with the 2–3 highest-pain jobs, expand to 10, then 50, until you've built a true digital employee for that role.
How can a small community generate serious revenue without ads?
Niche selection determines margin, not format. Even 1,000–2,000 engaged members in the right niche can generate €400–500K/year through events, merch, and memberships. The same painting retreat that earns $5,000 for the general public earns $90,000–$110,000 when the niche is burnt-out entrepreneurs. Pick a high-disposable-income niche and layer the Free + Premium Stack.
What is 'Anti-AI as differentiation' and when should I use it?
As AI sanitizes and automates content, live, unscripted, human, messy output becomes scarce and valuable. Betting on authentic long-form human presence is a deliberate contrast to the trend, and that contrast becomes your differentiator. Use it when your niche craves authenticity and connection — a live unscripted show or IRL event can outperform polished AI content precisely because it's human.
How do I stress-test for builder-market fit?
Ask whether you have an unfair advantage in this niche: lived experience of the pain, an existing audience, or a relevant skill like event organization, a health background, or coding. A great niche with no builder-market fit is still hard to execute. If there's a mismatch, either close the gap by partnering, or pivot to a niche where your advantages actually apply.
Can one niche support more than one product?
Yes, and the framework encourages noting all viable product types for a single niche. A GERD niche could support a vertical health app, a community, and a personalized nutrition referral business. Since the niche is the marriage and the product is the experiment, having multiple product options for one audience is a strong signal the niche is durable and worth committing to.
What is the 'Juniors' positioning for AI employee companies?
'Juniors' is positioning your AI employee product to replace junior-level, repetitive, non-creative work rather than claiming to replace senior talent. It lowers threat perception, makes the sale easier, and honestly matches current AI capability. Instead of 'we replace your team,' it's 'we handle the menial work at a fraction of the cost of a junior hire' — a friendlier, more credible pitch.