Unilever · Google Cloud
15-35% increase — retailer sales
Independently reportedOther · Other · Classical ML · Scaled
Other
Vendor-reportedTechnology & softwareAbandonedEnterpriseClassical MLLive 2008
Vendor-reported. The customer is named and the numbers are quoted from the source page, but the account comes from the vendor. No independent confirmation.
Reported by publichealthaihandbook.com (unclear). Checked against the source page on 2026-09-05. 3 of 3 figures below appear on that page word for word.
The problem
Predicting flu outbreaks faster than traditional surveillance methods.
What was deployed
- Products
- Google Flu Trends
- Technique
- Classical ML
- Build or buy
- Built in-house or to order
- Deployment
- SaaS
- Data used
- Google search queries (e.g., “flu symptoms”, “fever medicine”)
What changed
Each row is quoted from the source. Figures we could not find on the page in those words are marked — they are kept, not deleted, so you can judge them.
r² > 0.90Correlation with CDC data
“Strong correlation with CDC data (r² > 0.90)”
Quoted word for word from the source
1-2 weeks fasterSpeed of estimatesfrom CDC
“Provided estimates 1-2 weeks faster than CDC”
Quoted word for word from the source
Consistently overestimated for over 100 weeksOverestimation duration
“Consistently overestimated for over 100 weeks”
Quoted word for word from the source
Significantly overestimatedOverestimation of flu prevalence2012-2013 season
“Significantly overestimated flu prevalence in 2012-2013 season”
Quoted word for word from the source
more than twofold overestimationPeak error
“Peak error: more than twofold overestimation”
Quoted word for word from the source
Minimal costCost vs traditional surveillance2008-2011
“Minimal cost compared to traditional surveillance”
Quoted word for word from the source
Difficulties and limits
- Algorithm dynamics: Search algorithms changed, affecting what terms people saw and clicked
- Media attention: Increased flu media coverage drove searches independent of actual flu cases
- Overfitting: Model fit historical quirks rather than true flu-search relationships
- No validation: Lack of ongoing validation and model updating
- Black box: Google did not share methodology, preventing external scrutiny
Sources
- publichealthaihandbook.com — unclear, 2025-10-12
Similar deployments
ANZ Bank · NVIDIA
0.82 — Gini coefficient for assessing risk
Independently reportedBanking & finance · Other · Classical ML · Pilot
BHP Billiton
zero — accidents due to driver drowsiness
Independently reportedOther · Other · Classical ML · In production · 2022
DSST: College View · Playlab, Class Companion, Manta AI
Vendor-reportedEducation · Other · NLP · In production · 2024
Cubo Ai · Google Cloud
more than 10X — user growth supported with same IT workforce
Vendor-reportedOther · Other · Computer vision · Scaled · 2019
Precision HVAC & Cooling · AutoScale My Biz
15+ — High-Value After-Hours Leads Captured
Vendor-reportedOther · Other · LLM · In production