Risk register · entry
Q1 · PredictableZillow Offers
An AI home-pricing model overpaid at scale and forced Zillow to exit.
Documented, foreseeable risks that were ignored anyway. The failure is attention, not information.
Why this quadrant
Each individual home purchase looked like a simple, thin-tail payoff (a bounded renovate-and-flip margin priced by a formula), but scaling that formula across thousands of correlated bets on one housing cycle turned the aggregate payoff complex and fat-tailed, since a single market turn hit the whole inventory at once rather than being smoothed out by diversification.
The record
- $304 million Q3 2021 inventory write-down in the Homes segmentcertain
- Guidance of an additional $240-265 million in Q4 2021 write-downscertain
- Homes segment pre-tax net loss of $421.6 million in Q3 2021certain
- Total write-downs cited at roughly $540-569 million across the wind-downlikely
- About 2,000 jobs cut, roughly 25% of Zillow's workforcecertain
- Announcement date of exit: November 2, 2021certain
- About 8,000 homes acquired by Zillow Offers in Q3 2021; more than 3,800 in Q2 2021likely
- Roughly 7,000 remaining homes marketed to institutional investors for about $2.8 billion (~$400,000 average)certain
- Internal sample of 650 Zillow-owned homes: two-thirds priced below what Zillow paidlikely
- Stock fell roughly 25% the day after the announcementuncertain
- Zestimate launched 2006; Zillow Offers launched 2018likely
- Average loss cited at roughly $30,000 per home in inventoryuncertain
Sources
The newsletter
One risk story a week, taken apart the way this one was: what was known, what was ignored, and which quadrant it really belonged to.
No spam, one email a week.