Across 500+ YC companies, the pivot is the single most common growth event. Not fundraising. Not launches. The decision to throw away the original idea. The Master Pivot Database tracks every documented pivot in YC history; below are the ones every founder should know.
Nine pivots that worked
| Company | Originally | Pivoted to | Trigger |
|---|---|---|---|
| Twitch (Justin.tv, W07) | Lifecasting one person 24/7 | Live streaming platform | Users only watched gaming streams |
| Brex (W17) | VR headsets for Brazil | Corporate cards for startups | Couldn't get hardware traction; saw founder pain firsthand |
| Reddit (S05) | Mobile food ordering (Snoo) | Link-sharing community | PG asked them to build the 'front page of the internet' |
| Airbnb (W09) | Air mattresses for conference attendees | Full-home short-term rentals | Users kept asking for non-conference dates |
| DoorDash (S13) | Palo Alto Delivery (1 restaurant) | National last-mile logistics | Demand from non-Palo-Alto restaurants in week 6 |
| Coinbase (S12) | Bitcoin wallet (consumer) | Exchange + wallet + custody | Realized buying was the harder problem |
| Razorpay (W15) | Generic payment SDK | India-specific payment stack for SMBs | Indian banks blocked international gateways |
| Cruise (W14) | Aftermarket self-driving kit for cars | Full autonomous vehicle stack | Aftermarket retrofit was a regulatory dead end |
| Flexport (W14) | Customs brokerage software | Full digital freight forwarder | Customers wanted the whole shipment handled |
The shape of a successful pivot
- Triggered by user behaviour, not investor feedback (every single one above).
- Decided in under 2 weeks once the trigger was clear (Brex, DoorDash, Twitch all flipped fast).
- Kept at least one of: team, distribution channel, or technical advantage. None of them started from scratch.
- Was announced internally before externally — co-founders aligned, board told, then users.
When NOT to pivot
The Pivot Database also tracks the failed pivots — companies that pivoted too early, too often, or for the wrong reason. The single largest pattern in failed pivots is 'we pivoted because we couldn't raise.' Pivoting to a story that fundraises rarely produces a company that compounds.
Key takeaways
- Every YC unicorn pivoted at least once.
- Successful pivots are triggered by user behaviour, not investor pressure.
- The decision usually takes <2 weeks once the trigger is clear.
- Keep at least one asset (team, distribution, or tech). Don't restart.