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 to invest. What you don't have is a clear signal on what to build that people will actually pay for.

This shows up in specific ways:

  • You've started three side projects this year. All abandoned at 40% completion when the "is this even a real problem?" doubt crept in
  • You spend evenings scrolling Product Hunt, Indie Hackers, and Reddit looking for "gaps" but everything feels either saturated or trivial
  • You've interviewed five potential users who said "that sounds interesting" but none would commit to a pre-sale or even a follow-up call
  • You keep circling back to "scratch your own itch" ideas, but your personal problems are either too niche or already solved by tools you already use
  • The vague advice—"talk to customers," "find a painful problem," "look for workflows held together by duct tape"—feels unactionable when you sit down Monday morning with a blank notebook

The core issue: you're treating opportunity discovery as inspiration rather than research. You're waiting for a lightning strike instead of running a systematic process that produces viable candidates.


Why the Usual Fixes Fail

"Just Talk to Customers"

This advice assumes you know which customers to talk to and what to ask. Without a defined segment and hypothesis, conversations drift into polite feature requests for things that don't exist yet. You end up with a list of "nice to haves" from people who will never buy.

"Scratch Your Own Itch"

Works if your itch is shared by thousands of people with budget. Fails when your problem is idiosyncratic, already solved, or belongs to a market that doesn't pay for software. Most developers' itches are terrible business opportunities.

"Look for Complaints on Reddit/Twitter"

Complaints are noise. People complain about things they'll never pay to fix. They complain about pricing of tools they'll never leave. They complain about missing features in products they got for free. Signal-to-noise ratio is abysmal without a filter.

"Build an Audience First"

Audience-building is a separate skill set that takes 12-24 months. It delays validation. You can validate without an audience. You cannot validate without a specific hypothesis about a specific group.

"Use a Framework (Jobs-to-be-Done, etc.)"

Frameworks are analysis tools, not discovery tools. They help you structure what you've already found. They don't tell you where to look or how to generate initial candidates.


Step-by-Step Solution: The Opportunity Discovery Loop

This process takes 2-3 weeks of part-time work. It produces 3-5 validated problem hypotheses ready for solution validation. Do each step completely before moving to the next.

Week 1: Define Your Search Space (Days 1-3)

Step 1.1: List your unfair advantages
Write down every asset you bring that isn't "I can code." Examples:

  • Domain expertise: "I worked in commercial real estate lease administration for 4 years"
  • Network access: "My sister manages 12 dental practices; I can get intro calls with 8 office managers"
  • Distribution channel: "I moderate a 3,000-member Discord for independent insurance adjusters"
  • Data access: "I know how to pull and clean FDA 510(k) clearance data"
  • Regulatory knowledge: "I understand HIPAA business associate agreements cold"

Step 1.2: Define your constraints
Be honest. Write down:

  • Hours/week you'll actually commit (not aspirational)
  • Runway: how long before you need revenue
  • Technical ceiling: what you won't build (e.g., "no ML," "no hardware," "no mobile-first")
  • Market constraints: "B2B only," "US only," "teams of 10-500," "annual budget >$5k"

Step 1.3: Generate 20 initial segments
Combine advantages + constraints. Format: [Role] at [Company Type] who [Context].

Examples from a former CRE lease admin:

  1. Lease administrators at retailers with 50-500 locations managing renewals in spreadsheets
  2. Property managers at private equity-owned multifamily portfolios tracking CAM reconciliations
  3. Tenant reps at boutique brokerages handling lease abstraction for clients
  4. Asset managers at REITs reviewing portfolio-level lease clauses for risk
  5. Facilities directors at healthcare systems managing compliance deadlines across sites

Stop at 20. Don't filter yet.

Week 1-2: Rapid Segment Filtering (Days 4-10)

Step 2.1: The 30-minute desk research per segment
For each segment, answer these using only public sources (LinkedIn, company websites, job postings, Glassdoor, industry publications, Reddit, forums):

Question Pass Criteria
Do they spend money on software? Evidence of existing tool budgets (job posts mentioning tools, pricing pages, case studies)
Is the problem recurring? Workflow happens weekly/monthly, not annually
Is it measurable? Time lost, error rate, compliance risk, revenue leakage quantifiable
Can I reach them? Conferences, communities, LinkedIn groups, newsletters, cold email viable
Is the segment growing/stable? Headcount trends, industry tailwinds, regulatory pressure increasing

Step 2.2: Score and cut
Score each segment 1-5 on each criterion. Cut anything below 15/25. You should have 5-8 segments left.

Step 2.3: The "day in the life" test
For each remaining segment, write a 150-word narrative of a typical workday focusing on the workflow you suspect is broken. Use job descriptions, YouTube "day in the life" videos, forum threads. If you can't write this convincingly, cut the segment.

Week 2-3: Problem Hypothesis Generation (Days 11-17)

Step 3.1: Conduct 3-5 discovery calls per segment
Goal: not to sell. Goal: map the workflow and find where money/time leaks.

Outreach template (LinkedIn/email):

"Hi [Name], I'm researching how [role] at [company type] handle [specific workflow]. Not selling anything—just trying to understand the current process. Would you spare 15 minutes to walk me through how you handle [specific task] today? Happy to share what I find across the industry."

Step 3.2: Use the same interview script every time

  1. "Walk me through the last time you did [workflow] start to finish."
  2. "What tools or spreadsheets do you use at each step?"
  3. "Where do things break, get delayed, or require manual cleanup?"
  4. "What happens when it breaks? Who notices? What's the cost?"
  5. "How do you handle [specific edge case: regulatory change, vendor error, volume spike]?"
  6. "If you could wave a wand and fix one part, what would it be?"
  7. "What have you tried already? Why didn't it stick?"

Record (with permission). Transcribe. Do not take notes during—listen.

Step 3.3: Extract problem hypotheses
After each batch of 3-5 calls, synthesize. Format each hypothesis as:

[Segment] loses [quantifiable metric] because [root cause], and current workarounds [fail because].

Examples:

  • Lease admins at 100-location retailers lose 12 hrs/month per location on CAM reconciliation because landlord statements arrive in 47 different formats, and current workarounds (outsourcing to $200/hr consultants) fail because turnaround exceeds audit deadlines.
  • Tenant reps at boutique brokerages lose 30% of billable hours to lease abstraction because PDFs lack structure, and current workarounds (junior staff manual entry) fail because error rate triggers client disputes.

Keep only hypotheses where:

  • The metric is credible (you heard it from multiple sources)
  • The root cause is specific and technical/process, not "people are lazy"
  • Existing workarounds are expensive, slow, or error-prone
  • You can imagine a software solution addressing the root cause

Week 3: Solution Validation Prep (Days 18-21)

Step 4.1: Build the "minimum credible demo"
For each surviving hypothesis (aim for 3-5), create a one-pager:

  • Problem statement (from 3.3)
  • Proposed solution: 3 bullet features max
  • Target user: exact title, company size, trigger event

- Pricing hypothesis: per seat/usage/outcome, dollar amount

Start here

Top comments (0)