DEV Community

Alan Descure
Alan Descure

Posted on Originally published at smtm-store.pages.dev

Finding Real Opportunities: A Practical Guide to Opportunity Discovery

Finding Real Opportunities: A Practical Guide to Opportunity Discovery

The Problem in Concrete Terms

You're staring at a blank notebook or an empty Notion page. You want to build something—product, service, business—but every idea feels either obvious and crowded, or vague and unvalidated. You've read about "scratching your own itch," but your daily frustrations seem too small for a business. You've tried "looking for problems in forums," but Reddit threads and Twitter complaints blur into noise. You've considered "trend riding," but by the time a trend hits your feed, the window has closed.

The concrete reality: you have limited time, limited capital, and zero tolerance for building something nobody pays for. You need a repeatable method to surface opportunities that are specific enough to act on, painful enough that people pay, and accessible enough that you can actually reach the buyers. Most advice gives you starting points ("talk to customers!") but no decision framework for choosing which customers, which problems, or when to stop researching and start building.


Why the Usual Fixes Fail

1. "Scratch your own itch" assumes your problems are market-sized. Your workflow annoyances are real, but they're often too niche, too low-value, or too idiosyncratic to support a business. You also lack objectivity—you can't tell if you're the only one with this problem.

2. "Mine forums for complaints" produces signal-to-noise ratios of 1:1000. People complain about everything. Distinguishing "I'd pay $50/month to fix this" from "this is mildly annoying" requires context you don't have. You also can't tell if complainers are decision-makers or just venting.

3. "Follow trending topics" puts you in crowded markets with established incumbents. By the time a trend appears in newsletters or Product Hunt, early movers have captured distribution. You're competing on features in a defined category instead of defining a new one.

4. "Interview 50 customers" without a selection framework wastes months. Talking to random people in a broad industry yields contradictory feature requests. You need a hypothesis about who has the most urgent problem before you schedule calls.

5. "Build an MVP and iterate" burns runway on the wrong problem. Iteration optimizes a solution. It doesn't tell you if you're solving a problem worth solving. You can perfectly execute a product nobody buys.


Step-by-Step Solution

Phase 1: Define Your Search Boundaries (Week 1)

1.1 List your unfair advantages. Write down every asset you control: technical skills, domain expertise, audience access, distribution channels, capital, network connections, proprietary data, regulatory knowledge. Be specific: "3 years selling to dental practices" beats "healthcare experience." "12K newsletter subscribers in DevOps" beats "tech audience."

1.2 Define your constraint envelope. Maximum time to first revenue (e.g., 90 days). Maximum upfront cash (e.g., $5K). Minimum monthly revenue to continue (e.g., $3K). Willingness to do sales, support, ops, coding. These constraints eliminate entire categories of opportunities before you waste time evaluating them.

1.3 Pick 3-5 verticals where you have some advantage. Not "SaaS" or "B2B." Verticals: "independent insurance agencies," "commercial HVAC contractors," "specialty coffee roasters," "boutique law firms under 20 attorneys." Specificity lets you find the right people and speak their language.

Phase 2: Map the Value Chain (Week 1-2)

2.1 Draw the money flow for each vertical. Who pays whom, for what, how often, and why. Example for independent insurance agencies:

  • Carriers → Agencies: commissions (renewal + new business)
  • Agencies → Producers: split commissions
  • Agencies → Vendors: AMS, comparative raters, lead gen, marketing, E&O insurance
  • Clients → Agencies: premiums (agency keeps commission)
  • Agencies → Staff: salaries for CSRs, account managers, processors

2.2 Identify where money gets stuck, leaked, or delayed. Look for: manual re-entry between systems, spreadsheet bridges, delayed payments, high error rates, compliance fire drills, seasonal crunches, knowledge trapped in one person's head. These are high-value problem zones.

2.3 Rank verticals by problem density × your access. Score each vertical 1-5 on: (a) how many stuck-money spots you found, (b) how easily you can reach 10 decision-makers this month. Multiply scores. Top 2 verticals move forward.

Phase 3: Run Targeted Discovery (Week 2-4)

3.1 Build a decision-maker list, not a user list. For each top vertical, find 30-50 people who control budget for the problem zones you identified. Titles: Owner, Principal, VP Operations, Office Manager, Controller. Use LinkedIn Sales Navigator, Apollo, industry directories, association member lists, conference attendee lists. Export to spreadsheet with: name, title, company, email, phone, problem-zone hypothesis.

3.2 Craft a cold outreach template that earns a 15-minute call. No "pick your brain." No "quick chat." Script:

"Hi [Name], I'm researching how [vertical] firms handle [specific problem zone, e.g., 'policy renewal data entry between AMS and carrier portals']. You're a [title] at a [size] firm—likely close to this workflow. Not selling anything. 15 minutes, 3 questions, I'll share aggregated findings. Worth a brief call?"

3.3 Run the discovery interview with a rigid script. Three questions only:

  1. "Walk me through the last time you dealt with [problem zone]. What happened, step by step?"
  2. "What did that cost you—in time, money, errors, or missed opportunity?"
  3. "What have you tried to fix it? Why didn't it stick?"

Take verbatim notes. Record if permitted. Do not pitch. Do not suggest solutions. Do not ask "would you buy X?"

3.4 Score each interview immediately after. Rate 1-5 on:

  • Urgency: How actively are they trying to solve this right now?
  • Budget authority: Can this person say yes to a $500-5000/month tool?
  • Frequency: Daily/weekly problem vs. monthly/quarterly?
  • Workaround pain: Are they paying people, tolerating errors, or using duct-tape software?
  • Specificity: Can they describe the workflow in detail, or is it vague?

Discard verticals where average urgency + budget authority < 7/10 across 8+ interviews.

Phase 4: Validate Willingness to Pay (Week 4-6)

4.1 Create a one-page problem/solution brief. One page: problem description (their words), current cost of problem, proposed solution concept, pricing model, implementation timeline. No design. PDF or Notion link.

4.2 Run 5-10 "concierge validation" conversations. Same decision-makers. Show the brief. Ask: "If this existed exactly as described, what would you need to see to commit to a 3-month pilot at $[price]?" Note objections precisely. "Too expensive" ≠ "no budget"—ask "compared to what?" "Need integration with X" = product requirement. "Need SOC2" = dealbreaker or enterprise signal.

4.3 Collect letters of intent (LOIs). Non-binding, one-page: "We intend to pilot [solution] at $[price]/month for 3 months starting [date], contingent on [specific conditions]." Target 3-5 LOIs from different companies. This is your green light to build.

4.4 If <3 LOIs, return to Phase 3 with a refined problem zone. Use objection data to narrow or shift. Do not build without LOIs.

Phase 5: Build the Smallest Billable Version (Week 6-10)

5.1 Define the "first paid workflow." One end-to-end flow that delivers the core value proposition. Example: "Upload renewal list → auto-match policies to carrier portals → flag discrepancies → export correction file." Not dashboards, not settings, not onboarding flow. Just the workflow.

5.2 Build for the LOI companies only. Hardcode their specific AMS, carrier portals, file formats. No configuration. No multi-tenant. Deliver to their environment (shared server, VPN, local install). Charge from day one per LOI terms.

5.3 Weekly 15-minute check-ins with each pilot customer. "What broke? What's missing? What would make you renew?" Ship fixes within 48 hours. Document every request. Pattern-match after 3 weeks.

Start here

Top comments (0)