US-based software company Datadog has acquired Eppo, a platform for feature flagging and experimentation, to enhance its cloud monitoring and security offerings.

Eppo will continue to operate under the brand name “Eppo by Datadog.”

The financial terms were not disclosed, but Upstarts Media estimated the acquisition at US$220 million.

Eppo enables developers to run randomized application experiments, using AI and machine learning to assess model performance in real-time.

Founded in San Francisco, the startup raised US$47.5 million from investors like Innovation Endeavors, Menlo Ventures, and Amplify Partners.

🔗 Source: TechCrunch

🧠 Food for thought

1️⃣ Datadog’s acquisition strategy focuses on completing its observability stack

Datadog has consistently expanded its capabilities through strategic acquisitions, with Eppo representing the latest move in this pattern.

The company previously acquired Logmatic.io in 2017 to enhance its log management capabilities, demonstrating a methodical approach to building out its platform through targeted purchases 1.

This acquisition strategy has been instrumental in Datadog’s growth from a pure infrastructure monitoring company in 2012 to a comprehensive observability platform with diverse offerings spanning application performance, log management, and now experimentation 2.

The company’s expansion through acquisitions has supported its impressive growth trajectory, helping it reach a valuation of nearly $11 billion by 2019 after launching with skepticism from investors just seven years earlier 2.

Datadog’s approach demonstrates how cloud-native companies can rapidly expand their technology portfolios by acquiring specialized startups rather than building all capabilities internally, accelerating their ability to address evolving customer needs.

2️⃣ Experimentation platforms become critical infrastructure for AI deployment

Eppo’s value proposition focuses on helping companies measure the performance of AI models in production, addressing a growing challenge as organizations deploy multiple competing models.

With AI automation expected to reach $600 billion and projected to save companies an average of 22% on process costs, organizations increasingly need sophisticated tools to determine which models deliver the best results 34.