Startup Pitch Teardown

Slide 1 – “Reads your brainwaves to predict what you want to type before you think it.”
• Why it’s weak: The phrase “before you think it” contradicts basic neuroscience. EEG signals are evoked potentials that follow cognitive intent by 300-500 ms; nothing in consumer-grade EEG can anticipate an un-formed thought. The claim invites immediate skepticism from both scientists and investors who have followed BCI research.
• How to strengthen: Replace the over-statement with an empirically defensible value prop:
“MindMeld AI detects motor-imagery and language-planning signatures in EEG 150 ms after intent forms, letting you type at 120 WPM—4× faster than thumbs.”

Slide 4 – “TAM: $180 B” (derived from 3.5 B smartphone users).
• Why it’s weak: Top-down multiplication of every smartphone owner by a presumed price point is not a serviceable obtainable market (SOM). The average consumer will not pay $50–$100 for an accessory unless it becomes a platform standard (like AirPods). Comparable BCI consumer products (NextMind, MUSE) have sold low-six-figure units at best.
• How to strengthen: Break the market into reachable slices with bottoms-up math:
• Phase-1: high-mobility-impaired users (ALS, spinal-cord injury) → 3 M addressable × $500 = $1.5 B SAM.
• Phase-2: prosumer AR/VR creators → 25 M headsets × $200 attach-rate = $5 B SAM.
• Provide clear regulatory path and attach-rate assumptions for each slice instead of the blanket smartphone TAM.

Slide 5 – Traction (“$200 K ARR, 12 enterprise pilots…partnership discussions with Apple and Samsung”).
• Why it’s weak: $200 K ARR across 500 beta users implies either < $400 annual revenue per user or that most users are still free pilots. “Enterprise pilots” are not contracts; conversion rates in B2B hardware are 10-20 %. “Discussions” with Apple/Samsung are non-binding and common in hardware startups. These items read like signal-boosting rather than durable traction.
• How to strengthen:
• Convert pilots into LOIs or paid letters-of-intent: “8 of 12 enterprises signed 1-year LOIs worth $2.4 M contingent on FDA 510(k) clearance.”
• Replace “discussions” with concrete milestones: “Apple MFi team has requested 50 units for MFi certification lab; Samsung’s C-Lab has issued term-sheet under NDA.”
• Break out ARR by vertical to show paying vs. free users and attach pipeline with conversion probability.