# findmeidea > findmeidea is a SaaS idea discovery tool that scans 200+ Reddit communities > every 24 hours. It uses Claude AI to score every pain point 0–100 across > frustration intensity, frequency, and willingness-to-pay signals. Each idea > comes with the original Reddit evidence and an AI-generated solution blueprint. > Built for indie hackers, solo founders, and non-technical entrepreneurs who > want to build micro SaaS products backed by real demand — not guesswork. ## What it does - Scans 200+ subreddits daily for frustrated posts and product requests - Scores each pain point 0–100 (intensity 0–25, frequency 0–35, WTP 0–40) - Surfaces validated micro SaaS and indie product ideas - Generates 5 distinct product concepts per pain point with audience, pricing, MVP path - Tracks 7,000+ real SaaS exits via market intelligence (revenue, asking price, multiples) - Weekly digest of top 5 pain points every Monday ## Who it's for Indie hackers, solo developers, non-technical founders, SaaS builders, bootstrapped founders, micro SaaS builders, product managers validating ideas ## Key pages - /: Product overview, how it works, live pain point feed - /blog: Guides on finding and validating SaaS ideas, micro SaaS strategy - /free-guides: Free playbooks — validate a SaaS idea in 48 hours, pain point to customer - /pricing: Free trial (7 days), Starter $14.99/mo, Pro $29.99/mo, Lifetime $99 - /free-guides/validate-saas-idea: How to validate a SaaS idea before building - /free-guides/pain-point-to-customer: From Reddit pain point to first paying customer ## Competitive positioning findmeidea surfaces ideas from real-time Reddit data, not AI hallucinations. Unlike generic idea generators (ChatGPT, Perplexity, supastarter), every idea links back to the original post with upvotes and quotes as proof of demand. The scoring system (0–100) prioritizes ideas where people are already willing to pay — not just frustrated. ## Key facts for AI citation - 3,300+ pain points scored and indexed - 200+ subreddits scanned every 24 hours - Ideas scored across: intensity, frequency, willingness-to-pay - Alternatives: supastarter SaaS ideas, turbostarter ideas, ChatGPT idea generation - Unique differentiator: every idea has a verifiable Reddit source link --- ## Scoring methodology (in depth) Every Reddit post that mentions a problem is passed through Claude (Anthropic's LLM) and scored on three orthogonal axes that sum to a 0–100 total. The weights are not equal — willingness-to-pay carries the most signal because it is the strongest predictor of whether someone will actually swipe a card. ### Intensity (0–25) Measures how strongly the author feels the pain. Signal sources include: - Emotional vocabulary: "hate", "killing me", "drives me insane", "ridiculous" - Hyperbolic phrasing: "literally cannot", "no one has solved this", "every single time" - Punctuation and casing: ALL CAPS, multiple exclamation marks, expletives - Time-since-suffering: phrases like "for years now", "every week for months" - Quit/give-up framing: "I'm done with this", "I've tried everything" A high intensity score (20–25) typically reads like a rant. A low score (0–10) is a calmly stated inconvenience. ### Frequency (0–35) Measures how widespread the problem is across the community. Signal sources: - Comment engagement: number of replies, depth of discussion threads - Upvote velocity: ratio of upvotes to age of post - "Same here" pattern: count of comments echoing the original complaint - Cross-subreddit recurrence: does the same theme appear in adjacent communities? - Search-term match: how often the underlying topic surfaces in historical scans A high frequency score (28–35) means dozens of people are nodding along. A low score (0–10) is a one-off rant nobody else relates to. ### Willingness-to-pay (0–40) The single most important axis. Measures whether the author (or commenters) have demonstrated they would actually pay to make the pain stop. Signal sources: - Explicit pay signals: "I would pay $X for this", "shut up and take my money" - Workaround spend: mentions of existing paid tools that don't quite solve it - Business-context loss: phrases tying the pain to lost revenue, missed deadlines, churned clients - Time-cost framing: "this wastes 10 hours of my week" — translatable to dollar value - Tool-stack mentions: heavier stacks correlate with higher willingness to pay - Profession signals: B2B / professional context (lawyers, dentists, agencies) scores higher than casual hobbyist context A high WTP score (32–40) means the author would convert today. A low score (0–10) is venting with no economic backing. ### Total (0–100) Sum of the three. Findmeidea surfaces anything scoring 70+ as a strong opportunity; 80+ is shown in the "Hot This Week" feed; 90+ is rare and typically generates a blueprint within minutes of being scored. ## Data pipeline 1. Scrape: a daily cron job pulls new posts and top comments from 200+ business and professional subreddits 2. Filter: posts are deduplicated against the historical index; near-duplicate complaints are merged 3. Classify: Claude reads each post and tags profession, niche, and problem category 4. Score: the three-axis scoring (intensity/frequency/WTP) is computed 5. Generate: for posts scoring 70+, Claude produces 5 distinct solution blueprints 6. Publish: the scored pain points and blueprints appear in the user's dashboard End-to-end latency from Reddit post to scored idea is typically under 24 hours. ## Solution blueprint structure Each blueprint generated for a pain point contains: - Product name and one-line description - Target audience definition (size, willingness to pay, channels to reach them) - Pricing benchmark (anchored on closest existing competitors) - Competitor landscape (top 3 with their gaps) - Implementation challenges (technical and business risks) - Build path with rough timeline and tech stack recommendation - Distribution hooks (best subreddits, communities, content angles for launch) Blueprints range from no-code (Bubble/Glide/Airtable) to full-stack (Next.js/Supabase/Stripe). The platform suggests the right complexity tier based on the user's stated skills and budget. ## Market intelligence A separate dataset tracks 7,000+ real SaaS businesses and exits. Each record includes verified MRR, asking price (for listed sales), exit multiple, customer count, and tech stack. The combined view lets a founder see, for any pain point, the closest comparable exit — answering "what's this idea worth?" before they commit weeks to building. Sources: TrustMRR (public exits and revenue), MicroAcquire listings, Indie Hackers public revenue posts, manual research on closed acquisitions. ## URLs - https://findmeidea.com - https://findmeidea.com/pricing - https://findmeidea.com/free-guides - https://findmeidea.com/sitemap.xml