Eppo runs trustworthy, data warehouse-native A/B tests. World-class statistical rigor, feature management, and experiment reports - all in one tool.
Founded
2021
Employees
501-1000
Funding
Series B ($28M)
AI Classification
DataAIAnalytics/BI
Eppo is a warehouse-native A/B testing and experimentation platform acquired by Datadog in May 2025 for approximately $220M, offering feature flagging, statistical experimentation, and AI model evaluation integrated directly with enterprise data warehouses (Snowflake, BigQuery, Redshift, Databricks). It serves data-mature technology organizations seeking to scale product, marketing, and engineering experimentation with statistical rigor.
Deep Intelligence
Buying Signals
Acquired by Datadog in May 2025 for ~$220M, with active integration underway into Datadog's Product Analytics, Real User Monitoring, and Session Replay offerings — signals major platform investment and expanded distribution
Raised $28M Series B in August 2024 led by Innovation Endeavors and Icon Ventures, with Snowflake Ventures and Datadog as investors — demonstrating investor conviction and data ecosystem alignment
AI model evaluation launched as a distinct product capability in 2024, targeting AI-native engineering teams evaluating LLM and model performance — new use case expanding TAM
Customer roster includes enterprise technology brands (Coinbase, Twitch, DraftKings, Perplexity, ClickUp), indicating successful enterprise-level contract wins and expansion of customer base
Global operations across five continents with GCP-native, SOC2 Type 2 certified infrastructure — signals readiness for enterprise procurement requirements and international expansion
Account Scoring
Tier BMedium Churn RiskExpanding
Pain Signals
Post-acquisition product roadmap uncertainty: existing customers may face consolidation or migration pressure as Eppo is integrated into Datadog's platform under the 'Eppo by Datadog' brand, creating potential churn or renegotiation moments
Estimated ARR of ~$15M as of late 2025 relative to $51M in total funding suggests the company was still in a scaling phase at time of acquisition, indicating revenue-per-employee efficiency was a challenge pre-exit
Standalone pricing model not publicly visible, which may reflect a high-touch sales motion that limits self-serve adoption and creates friction for smaller or mid-market prospects