Finding Real Problems — Evidence-Based Research — From Layoff to Launch: Build Your First Micro-Business

## Finding Real Problems — Evidence-Based Research There's a hierarchy to evidence. Not all signals are equal. Understanding which sources of evidence actually predict willingness to pay — and which ones are noise — is one of the most valuable skills a founder can develop. This lesson is about learning to read the internet like a researcher, not a consumer. --- ## The Evidence Hierarchy Here's the hard truth about where founders look for validation versus where they should look: ``` HIGHEST VALUE | | Someone paid money for something (even if it wasn't yours) | Someone complained publicly, repeatedly, with specifics | Someone created content asking how to solve the problem | High search volume for problem-framed queries | Surveys of target customers (well-constructed, n > 50) | Surveys of friends and colleagues | Your gut feeling based on experience | "Everyone I know says it's a problem" LOWEST VALUE ``` Most founders operate at the bottom of this hierarchy. They ask people they know, interpret enthusiasm as validation, and start building. You're going to operate at the top. --- ## Platform-by-Platform Signal Reading Different platforms produce different types of evidence. Here's how to read each one: ### Reddit — Your Most Valuable Research Tool Reddit is where people complain with specificity. Not "this industry has problems" — "my software crashed during payroll processing at 11pm on a Friday and our vendor's support line goes to voicemail." That specificity is gold. **What to look for:** - **Complaint posts with 100+ upvotes** — Community consensus that this problem is real and shared - **"Is there a tool that does X?" posts** — Direct market demand signal - **Repeated questions across multiple threads** — The same question appearing 5+ times means many more people have it but didn't ask - **Workaround threads** — When people describe elaborate DIY solutions, they're telling you they'd pay for something better **How to search effectively:** - `site:reddit.com "I wish there was" [your domain]` - `site:reddit.com "why is there no" [your domain]` - `site:reddit.com "does anyone know a tool" [your domain]` - Search inside specific subreddits for "problem" or "frustrated" or "annoying" **Engagement quality over quantity:** A post with 300 upvotes and 80 comments is more valuable than a post with 1,000 upvotes and 5 comments. Comments with specifics ("this happens to me every time I...") are primary research. **The signal you're looking for:** Real people describing a specific pain in their own words, with other people confirming they have the same experience. Not general frustration with an industry — a specific, actionable problem. --- ### YouTube — Behavior Evidence YouTube is where you learn what people *do*, not just what they *say*. Video behavior reveals priorities. **High-value signals:** - **Tutorial videos with high view counts** — If 200,000 people watched "How to do X manually in Excel," X is probably worth automating - **Comment sections on problem-framing videos** — Search for the problem, not the solution. "Why is [thing] so hard" comment sections are research gold. - **"Rant" videos with engagement** — Creators who rant about industry problems and get thousands of likes are confirming the pain is widespread - **Channels with small subscriber counts but consistent views** — Niche content that keeps getting watched means sustained demand, not a trend spike **What to look for in comments:** - "Same thing happened to me" (validation of the pain) - "I've been looking for a solution to this for years" (market timing) - Questions about alternative approaches (what workarounds exist?) - Mentions of other tools/approaches they've tried (competitive intelligence) --- ### App Store Reviews — Competitive Intelligence If there's a competing product in your space with reviews, those reviews are one of the most valuable datasets you have access to — and almost nobody reads them systematically. **The 2-3 star review goldmine:** - 1-star reviews are often rants from people who had a bad day - 5-star reviews are marketing material - **2-3 star reviews are where customers articulate exactly what's missing** Read 50 reviews in the 2-3 star range for your top competitors. Cluster the complaints. The most repeated complaints are your product spec. Do this before you write a single line of code. It's free research that took your competitors' customers months of frustration to produce. --- ### Google Trends — Timing and Trajectory Search trend data tells you whether interest is rising or falling — which affects your timing strategy dramatically. **Patterns and what they mean:** - **Steadily rising for 24+ months** — Strong signal. You're catching a wave, not chasing one that already broke. - **Flat with seasonal spikes** — Real demand, but plan for seasonality. Your busiest months matter. - **Sharp spike then collapse** — Trend-chasing territory. Someone made it viral; it's probably over. - **Declining** — Either the problem is being solved (bad), the market is consolidating (maybe okay), or it's genuinely fading (bad). Understand which before proceeding. - **Flat and low** — May be a niche not yet named/labeled. Check adjacent terms. **Pro tip:** Compare your core term to adjacent terms. If "AI writing tools" is plateauing but "AI writing tools for lawyers" is spiking, that's a micro-niche moment worth noticing. --- ### MNB Evidence Rows — 208,000+ Real Signals MicroNicheBrowser.com has collected and structured over 208,000 evidence rows from 11 platforms. Every row is a real data point: a Reddit post, a YouTube video, a trending keyword, a social signal. When you click into a niche in MNB, the Evidence section shows you the actual sources behind the score. This is the work our system did so you don't have to spend 40 hours doing manual research: - **Platform** — Where this evidence came from - **Title and snippet** — What was actually said - **Engagement metrics** — How much the community responded - **Relevance score** — How closely it maps to the core niche problem - **Published date** — Is this evidence current or 3 years old? Learn to read evidence rows the way a lawyer reads case files. Each one is a data point. The pattern across 50 evidence rows tells you something the score alone can't. --- ## The Pain-Point Roulette Approach Sometimes the best research strategy is serendipitous. You're not starting with a hypothesis — you're exposing yourself to a broad stream of real pain and seeing what resonates with your experience. The Pain Point Roulette tool pulls random high-engagement complaint posts from across our evidence database. It's deliberately unfocused — designed to surface pain you weren't looking for. **How to use it productively:** 1. Spend 20 minutes browsing without trying to find something specific 2. When something resonates — when you read a complaint and think "I know exactly why this happens" or "I dealt with this for years" — stop and dig deeper 3. Follow that thread: what subreddit? What related complaints? What solutions do people mention? Your gut reaction to a pain point is itself data. If you read a complaint and immediately understand the root cause, you have domain expertise that matters here. --- ## Building Your Evidence Dossier Before you build anything, you should have a structured collection of evidence for your niche. This doesn't have to be elaborate — a simple document with: - **3-5 Reddit threads** with the specific complaint language (link + summary) - **2-3 YouTube videos** showing the behavior (link + view count + notable comments) - **Top competitor App Store complaints** (10-15 clustered by theme) - **Keyword data** — search volume, trend direction, CPC - **MNB evidence rows** — link to the niche page, note the evidence count and top signals This dossier serves two purposes: it gives you confidence before you build, and it becomes the first version of your marketing copy. The words your customers use to describe their pain are the words you use in your headline.