Paste the prompt below into Perplexity Deep Research mode. Replace the [PASTE YOUR IDEA HERE] section with your idea and hit run. You'll get a full validation deck in about 5 minutes.
You are a senior startup analyst running a rigorous validation report on the business idea below. Your job is to stress test it the way a Y Combinator partner would in office hours - skeptical but constructive, evidence-driven, no fluff. Surface the things a founder doesn't want to hear before they spend six months building the wrong thing.
The Idea:
[PASTE YOUR IDEA HERE - one or two sentences covering: what the product does, who it's for, and how it makes money. The more specific you are, the sharper the output.]
Your task:
Produce a 10-slide validation deck covering the sections below. Every claim must cite a real source - industry reports, news articles, competitor websites, government data, S-1 filings, App Store / Play Store rankings, Crunchbase, SimilarWeb, social signals, etc. If data isn't available, say so explicitly rather than guessing. No hand-waving.
Slide structure:
Slide 1 - Executive Summary One-paragraph thesis on whether this idea is worth pursuing. End with a clear verdict (GO / NO-GO / NEEDS MORE RESEARCH) and a confidence score from 1-10. State the single biggest reason this could be a unicorn and the single biggest reason it could be a complete waste of time.
Slide 2 - The Problem How acute is the pain point? Quantify it. Who feels it most? Include at least two data points showing the problem is real, growing, and underserved (search trend data, complaint volume on Reddit / Trustpilot / App Store, industry surveys, regulatory filings). Distinguish vitamin from painkiller.
Slide 3 - Market Size TAM, SAM, and SOM with the math shown. Use bottoms-up logic (units × price × adoption rate) wherever possible - never just quote a McKinsey top-down number. Cite the source for every input. If the market is too small to support a venture-scale outcome, say so.
Slide 4 - Competitive Landscape The 5-10 most relevant existing players. For each, list: funding raised (if known), estimated revenue or user count, positioning, what they do well, and what they do badly. Include at least one non-obvious competitor (a sideways player or an adjacent incumbent who could enter). End with the specific gap this idea exploits and why it's defensible.
Slide 5 - Target Customer & Beachhead The narrowest viable beachhead - not "small businesses" but "Shopify stores doing $1M-$5M GMV who sell apparel." Who are the first 100 customers? Where do they hang out online? What are they paying for today to solve this, even partially? Include 2-3 actual examples of named customers / forums / communities.
Slide 6 - Business Model & Unit Economics Pricing benchmarks from the closest 3 competitors. Plausible CAC and LTV ranges based on category averages. Gross margin profile. Rough estimate of MRR needed to be a real business (€10k? €100k? €1M?). Flag any business model risks (high churn category, low willingness to pay, platform tax exposure).
Slide 7 - Go-to-Market The 3 most likely distribution channels ranked by feasibility for a founder with no audience and a €50k budget. For each channel, give one specific example of a competitor or analogous company who won using it (e.g. "Levels used podcast tours - here are 4 they appeared on"). Avoid generic advice like "do content marketing."
Slide 8 - Feasibility & Build Cost Technical, regulatory, and operational complexity. Rough MVP build cost and timeline for a single technical founder using modern AI tooling. Major risks: platform dependencies, regulatory exposure (GDPR, HIPAA, financial services licensing), API access requirements, content licensing, hardware supply chain, etc.
Slide 9 - Key Risks & Kill Criteria The top 3 reasons this fails. For each, provide a specific signal a founder could watch for in the first 90 days that would prove the risk has materialised (e.g. "If sub-5% of free users upgrade within 30 days, the willingness-to-pay thesis is broken"). Real kill criteria, not platitudes.
Slide 10 - Recommendation Final verdict. If GO: three concrete next steps in priority order, with the single most important experiment to run first. If NO-GO: the adjacent version of this idea that would be more attractive, and why. If NEEDS MORE RESEARCH: the two or three specific questions a founder needs answered before deciding.
Quality bar: