Honeycomb

Honeycomb, legally Hound Technology, Inc. and stylized honeycomb.io, is an American software company that develops an observability and application performance management platform delivered as SaaS. The product lets software developers investigate the behavior of live applications, particularly those built on microservice architectures, by querying high-cardinality, event-based telemetry in real time rather than relying on sampling or pre-aggregation. Capabilities span distributed tracing and analysis, metrics, log management and analytics, telemetry pipelines, service level objectives, heatmaps, BubbleUp root-cause analysis, and a private cloud deployment option.

The platform ingests telemetry from applications instrumented with OpenTelemetry SDKs as well as structured JSON and custom integrations, with documented instrumentation paths for AWS, Kubernetes and language SDKs including Node.js, Go, Python, Ruby, Java, .NET, JavaScript, Android, iOS and React Native. Underlying it is a proprietary columnar database running on Amazon Web Services; Amazon has cited the company as an early adopter of its Graviton ARM processors. More recent work targets AI systems and AI-assisted workflows: agent and LLM observability, an Agent Timeline for tracing multi-agent conversations and tracking per-model token cost and latency, Canvas investigations triggered by SLO burn, and a Model Context Protocol server that exposes telemetry to coding and investigation agents. Honeycomb is also a contributor to the OpenTelemetry project and publishes the O'Reilly book Observability Engineering, whose second edition adds 27 chapters.

Founding story

Honeycomb was founded in 2016 by Charity Majors and Christine Yen, engineers who had worked together at Parse, later acquired by Facebook. After the acquisition, Yen proposed that the two start a company; Majors wanted to build tooling to diagnose problems in live cloud applications, drawing on the Scuba in-memory analytics tool they had used at Facebook. The company describes its origin as an ops engineer and a developer sharing the same problem and disagreeing about whose fault it was.

Business model

Subscription SaaS observability platform sold to engineering organizations, with a free entry tier, self-serve sign-up and sales-led demos and custom plans; pricing is positioned around unlimited fields and unlimited users rather than per-seat or per-field charges, with a private cloud option for enterprises.

Recurring subscription revenue from platform usage, spanning a free tier, paid self-serve plans and enterprise contracts including private cloud deployments; the company reports metrics typical of a subscription business such as net revenue retention and customer counts.

Traction

Entering 2023 the company reported a second consecutive year of 2x revenue growth, over 160% net revenue retention and more than 600 customers globally. Named customers and reference users include Slack, Vanguard, Vercel, Robinhood, Moov, Frasers Group, Equinix, LaunchDarkly, Duolingo, Booking.com, Intercom/Fin, Mixpanel, Zeta Global, Salesforce, Stripe, HiPages, Gearset, Scribe, Homeaglow, Gem and Depot. Comparably recognized the company in more than ten categories in 2023 and 96% of employees ranked it a great place to work.

Latest developments

Recent product direction centers on the AI era: agent and LLM observability, an Agent Timeline that traces multi-agent conversations and tracks token cost and latency per model, Canvas investigations that start automatically when SLOs burn, encoded Skills playbooks, and a Honeycomb MCP server that connects coding and investigation agents to telemetry. The second edition of Observability Engineering was released with 27 new chapters, and the company ran O11yCon 2026 with talks from Salesforce, Slack, Stripe, Mixpanel and others, alongside a Leading With Observability interview series.