Please wait a moment...
5 articles in this tag
Five practitioner lessons from building a ~6,900-company venture-outcome graph — identity is the real bottleneck, the dangerous errors are the plausible ones, survivorship hides in track records too, the founder is the unit not the company, and consensus isn't correctness.
Almost every startup-scoring model is trained on companies that survived — because failures leave no structured trace. The fix isn't a cleverer architecture, it's the negative class: label the dead ones and weight them equally. Field notes on wiring a real failure class into a training corpus, and the two things that broke.
64.8% of VC deals return 0-1x. Only 4% return 10x+. The data from PitchBook and Cambridge Associates shows why deal screening volume is the most underrated edge in venture capital.
Our AI analysed 298 decks from companies worth $116B combined. Here's what the data says about detecting outlier potential, conviction archetypes, and the signals that predict which startups become unicorns.
We scored 298 pitch decks from companies that raised $116 billion combined. Here's what the data says about what separates funded decks from the rest.
Browse all tags
Want to know how your pitch deck scores?
Upload Your Deck — Free