Case Study: Zillow reduces outage impact to minutes and saves millions with Splunk Enterprise and the Splunk Machine Learning Toolkit

Zillow Helps Customers Find Their Way Home With Machine Learning

Zillow, a leading real estate and rental marketplace and one of the top 30 U.S. websites, faced growing operational challenges as terabytes of unstructured log and customer data made it hard to detect and diagnose website outages and measure release impact. Legacy tools left the team “blind,” causing slow root-cause analysis, risking customer experience, advertising revenue, and contractual obligations, and forcing developers to wait days for analytics.

By deploying Splunk Enterprise and the Splunk Machine Learning Toolkit to ingest and normalize data from web servers, applications, databases, load balancers and devices, Zillow built custom ML models, alerts and outlier detection for real-time visibility. The result: outage impact dropped from hours to minutes or seconds, the company saved millions, customer experience improved, and developers now ship code multiple times per day with immediate feedback.

Zillow

Jerome Ibanes

Data Architect