Case Study: Real-estate software provider AppFolio achieves 80–90% latency reduction and 300% adoption increase with Datadog LLM Observability

AppFolio selects Datadog LLM Observability to monitor and optimize its AI applications

AppFolio, a real-estate management software provider serving 20,000+ customers and managing over 8 million units, needed to deploy and scale LLM-powered features while managing model non-determinism and upholding responsible AI practices. The team also wanted a low-friction integration with Amazon Bedrock and tools to detect issues early during development—especially as they built Realm‑X Messages to reduce time spent on tenant communications.

AppFolio adopted Datadog LLM Observability to monitor usage, latency, token consumption, error rates and response quality (including toxicity and failure-to-answer checks), and to receive real-time alerts and topic-level insights. With this observability, they moved from QA to production in under a week, cut latency by 80–90%, drove nearly 300% higher adoption of Realm‑X Messages, and saved property managers about five hours per week while enabling safe, faster feature expansion.

AppFolio

Teddy Ho

Principal Product Manager