Opportunity Discovery: A Field Guide for When You’re Stuck Guessing
The Problem in Concrete Terms
You have capacity—time, capital, or technical skill—but you don’t know where to point it. You aren't lacking ideas; you're drowning in them. Or worse, you have zero ideas that feel viable. The stuckness looks like this:
- The "Shower Thought" Trap: You get excited about a problem you personally have (e.g., "I need a better way to organize my Notion templates"), build a solution, and realize the total addressable market is you and three friends on Discord.
- The "Keyword Research" Mirage: You spend weeks in Ahrefs or Semrush finding "low competition, high volume" keywords. You build a niche site or micro-SaaS for "best ergonomic keyboard for left-handed accountants." Traffic arrives. Revenue doesn't. The intent was informational, not commercial.
- The "Trend Chasing" Lag: You see AI wrappers, vertical SaaS, or "boring businesses" trending on Twitter/X. By the time you validate, build, and launch, the arbitrage is gone. You’re competing on features against funded teams or established incumbents.
- The "Validation Theater": You post a landing page, get 50 email signups (mostly friends and bots), call it "validated," build for three months, and launch to crickets. Signups $\neq$ Purchase Intent.
The core issue: You are looking for ideas when you should be hunting for evidence of unmet demand. Ideas are cheap. Evidence is expensive to fake.
Why the Usual Fixes Fail
1. "Scratch Your Own Itch" (Without Market Sizing)
Your itch is statistically anomalous. You are a builder. Most people are not. If you solve a builder problem, you compete with every other builder who had that same itch. If you solve a non-builder problem, you have to leave your desk to find it.
2. "Customer Interviews" (Done Wrong)
Most people ask: "Would you use a tool that does X?" The answer is always "Yes" (politeness) or "Maybe" (non-committal). This generates false positives. You aren't measuring demand; you're measuring agreeableness.
3. "Competitor Analysis" (Feature Comparison)
You make a spreadsheet of features. Competitor A has X, Y, Z. You plan to build X, Y, Z, and Q. You miss why customers churn from Competitor A. Features don't retain customers; workflows do. You end up building a "better" product nobody switches for because switching costs are too high.
4. "Build an MVP in a Weekend"
Speed is useless if the vector is wrong. A fast car driving off a cliff hits the bottom sooner. An MVP tests viability, not feasibility. You already know you can build it. You need to know if they will pay.
Step-by-Step Solution: The Evidence-Based Hunt
This process takes 2–4 weeks of part-time work. It requires zero code. It requires talking to strangers.
Phase 1: Define Your "Unfair Advantage" Constraints (Days 1–2)
Don't start with the market. Start with you. Constraints focus the search space.
Action: Write down your specific constraints. Be brutally honest.
- Capital: $0? $5k? $50k? (Dictates B2C vs B2B, Ads vs SEO, Sales-led vs PLG).
- Time: 10 hrs/week? 40 hrs/week? (Dictates complexity ceiling).
- Distribution: Do you have an audience? A network in healthcare? Zero connections? (Dictates Go-To-Market).
- Skill: Can you build complex backend? Only no-code? Only design? (Dictates technical moat).
- Risk Tolerance: Need cash flow in 30 days? Can wait 18 months? (Dictates business model: Service -> Productized Service -> Micro-SaaS -> Platform).
Output: A one-page "Founder Fit Matrix." If an opportunity requires $20k ad spend and you have $0, discard it now. Stop wasting cycles on impossible fits.
Phase 2: Source "Expensive Problems" Not "Ideas" (Days 3–10)
Ideas are solutions. Problems are market gaps. Look for Money Already Moving.
Tactic A: The "Job Board Archaeology" (B2B Focus)
- Go to Indeed, LinkedIn Jobs, niche boards (e.g., YC Work at a Startup, Wellfound, industry-specific boards).
- Search for roles: "Operations Manager," "RevOps," "Salesforce Admin," "Project Coordinator," "Compliance Analyst."
- Read the Responsibilities section, not Requirements. Look for manual repetition: "Manually compile weekly report from 5 sources," "Copy-paste data from CRM to spreadsheet," "Chase vendors for missing invoices."
- Signal: If a company pays $80k/yr for a human to do a repetitive digital task, that task is a software opportunity. The salary is the budget.
Tactic B: The "Agency/Service Provider Shadow" (Service -> Product)
- Find 10–20 specialized agencies/consultants (e.g., "Shopify CRO Agency," "SOC2 Compliance Consultants," "Medical Billing Services").
- Analyze their Case Studies and Service Pages. What is the "Done For You" deliverable?
- Look for "Productized Services" (fixed price, fixed scope). Example: "$3k/mo for 4 blog posts + SEO audit."
- Signal: The process inside that productized service is the software. The agency has already validated demand and pricing. You are automating the middle.
Tactic C: The "Review Mining" (B2B & B2C)
- Go to G2, Capterra, TrustRadius, Chrome Web Store, App Store.
- Filter reviews for the market leaders by "2 Stars" and "3 Stars." (1-star is usually bugs/outages; 5-star is fanboys. 2-3 stars are power users with specific workflow gaps).
- Search text for: "export," "workaround," "manual," "spreadsheet," "API," "integration," "clunky," "missing."
- Signal: "I love Tool X, but I have to export to CSV every Friday to do Y in Excel." -> Build the "Y" integration/automation as a micro-tool or standalone workflow.
Tactic D: The "Government/Regulatory Trigger" (High Barrier, High Moat)
- Search: "New [Industry] regulation 2024 compliance," "[State] new law requires [X] reporting."
- Examples: SOC2, HIPAA, GDPR, CCPA, ASC 606 (Revenue Recognition), ESG reporting mandates, BOI reporting (FinCEN).
- Signal: Compliance is mandatory, urgent, and boring. Budgets are pre-approved. Founders ignore it because it's unsexy. That is the moat.
Output: A list of 20–30 Raw Problem Statements.
Format: [Role/Persona] struggles to [Specific Workflow Step] because [Root Cause], resulting in [Quantifiable Pain: Time/Money/Risk].
Example: RevOps Managers struggle to reconcile Stripe revenue with Salesforce closed-won deals because Stripe metadata doesn't map to SFDC fields, resulting in 10 hrs/mo manual cleanup and audit risk.
Phase 3: Ruthless Filtering (Days 11–12)
Apply the Founder Fit Matrix (Phase 1) and Market Reality Filters.
Filter 1: Budget Authority. Does the person feeling the pain hold a corporate card or influence the buyer? (RevOps Manager = Yes. Junior SDR = No).
Filter 2: Frequency. Daily/Weekly pain > Monthly/Quarterly pain. (Habit formation requires frequency).
Filter 3: Workaround Exists. Are they already paying for a hack? (Zapier, VA, Excel, Agency). No workaround = No proven demand.
Filter 4: Switching Cost vs Value. Is the pain > the pain of adopting a new tool? (Data migration, training, security review).
Filter 5: Your Edge. Can you build the core hard part better/faster/cheaper than the 3 competitors you'll find?
Output: **Top 3
Top comments (0)