DEV Community

Alan Descure
Alan Descure

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

Opportunity Discovery: A Practical Guide for When You're Stuck

Opportunity Discovery: A Practical Guide for When You're Stuck

The Problem in Concrete Terms

You have the skills to build things. You have time (or can make it). You want to start a project, a product, or a business—but you keep circling the same question: what should I actually build?

This isn't "I have too many ideas." It's the opposite. You sit down to brainstorm and come up with nothing viable. Or you generate a list, but every idea feels either:

  • Too small to matter
  • Too crowded to enter
  • Too vague to execute
  • Something you'd abandon in two weeks

You've read about "scratching your own itch," but your daily problems feel trivial or already solved. You've tried "looking for problems in forums," but Reddit threads and support tickets feel like noise, not signal. You've considered "talking to customers," but you don't have customers yet, and cold outreach feels like spam.

The result: weeks or months of paralysis. You build nothing. You learn nothing. The gap between "person who could build something" and "person who has built something" keeps widening.


Why the Usual Fixes Fail

1. "Scratch your own itch" assumes you have painful, unsolved problems you're qualified to solve. Most developers' personal itches are either already solved (note-taking, task management, boilerplate generators) or too niche to sustain a product. Your itch is rarely a market.

2. "Browse Indie Hackers / Product Hunt / Reddit for ideas" trains you to see solutions, not problems. You see finished products and reverse-engineer the opportunity. But the opportunity existed before the product. By the time it's on Product Hunt, the easy entry point is gone. You're learning what worked then, not what works now.

3. "Talk to 50 customers before building" is a catch-22. You need an idea to know who to talk to. You need conversations to refine the idea. Most people pick a vague direction, talk to three people, hear "nice to have," and quit. The method works after you have a hypothesis, not before.

4. "Follow trends" (AI, Web3, no-code, etc.) puts you in the most crowded lanes. By the time a trend is legible enough to follow, the asymmetric opportunities are taken. You're competing with funded teams and insiders on their turf.

5. "Just start building something" produces portfolio projects, not products. Building teaches you how, not what. Without a thesis about why someone would pay, you optimize for technical elegance instead of viability.


Step-by-Step Solution

Phase 1: Map Your Asymmetric Advantages (Days 1–2)

Before looking outward, inventory what you bring that others don't. This isn't "passion." It's leverage.

Do this:

  1. List every job, freelance gig, side project, hobby, and community you've participated in for 6+ months.
  2. For each, write down:
    • What workflows, tools, or jargon are second nature to you but opaque to outsiders?
    • What manual workarounds did you see people accept because "that's just how it works"?
    • What data, access, or relationships do you have that a cold starter wouldn't?
  3. Cross-reference: which items appear in 2+ contexts? Those are your real advantages.

Example: You worked in a dental office for three years, then built a scheduling tool for a friend's salon. You know how appointment-based businesses handle no-shows, insurance verification, and staff handoffs. That's not a generic "scheduling problem"—it's a specific cluster you understand.

Output: 3–5 "advantage clusters" with concrete evidence.


Phase 2: Hunt for Expensive Problems in Those Clusters (Days 3–7)

Don't brainstorm. Investigate. Go where your advantage clusters live and look for expensive, recurring, manual work.

Do this:

  1. For each cluster, identify 3–5 places where practitioners congregate:
    • Specialized Slack/Discord communities (not general ones)
    • Industry-specific forums or subreddits with active practitioners
    • Trade association directories, conference attendee lists, newsletter subscriber bases
    • LinkedIn groups where people ask operational questions
  2. Spend 2 hours per cluster reading only questions/complaints about workflow. Ignore feature requests. Ignore "wouldn't it be cool if." Look for:
    • "I spend X hours/week manually doing Y"
    • "We pay $Z/month for [tool] but still have to [workaround]"
    • "Every time [event happens], we scramble to [manual fix]"
    • "Our biggest cost/error source is [process]"
  3. Capture verbatim quotes in a spreadsheet with columns: Source, Problem Description, Frequency Mentioned, Current Workaround, Estimated Cost (time/money/error rate).

Filter ruthlessly: Keep only problems that are:

  • Recurring (daily/weekly, not quarterly)
  • Expensive (≥5 hrs/week or ≥$200/month or causes lost revenue)
  • Manual (human judgment or copy-paste involved)
  • In your advantage cluster (you understand the context)

Output: 5–10 validated problem statements with evidence.


Phase 3: Test Willingness to Pay Before Building (Days 8–14)

You don't need 50 conversations. You need 3–5 qualified ones that prove budget and urgency.

Do this:

  1. From your problem list, pick the top 2–3. Write a one-sentence hypothesis for each: > "Dental office managers will pay $150/month for a tool that automatically verifies insurance eligibility 24hrs before appointments, because they currently spend 8 hrs/week on phone calls and lose $3k/month on denied claims."
  2. Find 10–15 people per hypothesis who match the buyer profile. Use your advantage access: former coworkers, community members, LinkedIn connections with relevant titles. Cold DM if needed, but warm is better.
  3. Request a 15-minute call. Script: > "I'm researching [specific problem]. You've mentioned this before. Not selling anything—just want to understand how you handle it today. 15 mins?"
  4. On the call, ask only:
    • "Walk me through the last time this problem occurred."
    • "What did you do? How long did it take? What did it cost?"
    • "What have you tried to fix it? Why didn't it work?"
    • "If a tool solved this for $X/month, what would need to be true for you to buy it?"
  5. Score each conversation:
    • Budget confirmed: They name a number or reference existing spend
    • Urgency confirmed: They describe active pain this week
    • Authority confirmed: They can decide or directly influence the purchase
    • Fit confirmed: Their workflow matches your hypothesis

Threshold: Proceed only if ≥3/5 conversations hit all four signals for the same problem.

Output: One validated problem with buyer evidence.


Phase 4: Define the Minimum Viable Solution (Days 15–18)

Don't build the product. Build the smallest thing that proves the workflow works.

Do this:

  1. Map the current manual workflow step-by-step (from your calls).
  2. Identify the single highest-leverage step to automate—the one that saves the most time or prevents the most errors.
  3. Define a "concierge MVP": you do that step manually for 3 pilot users, delivering the result via their existing tools (email, spreadsheet, Slack, PDF).
  4. Set a 2-week pilot goal: each pilot must complete ≥5 cycles of the workflow using your delivery.
  5. Charge a nominal fee ($50–200 for the pilot). Free pilots don't prove willingness to pay.

Output: A pilot plan with 3 committed users, defined success criteria, and a 2-week timeline.


Phase 5: Run the Pilot, Then Decide (Days 19–32)

Do this:

  1. Execute the concierge delivery. Document every edge case, exception, and complaint.
  2. After 2 weeks, interview each pilot with:
    • "What would you do if this stopped working tomorrow?"
    • "What's the most you'd pay for this to continue?"
    • "What's missing that would make this a no-brainer?"
  3. Decision matrix:
    • All 3 would pay ≥$100/mo and say "can't live without it" → Build the automated version.
    • 2/3 would pay, 1 churns → Fix the gap

Start here

Top comments (0)