The funding signal

Glean announced a $150 million Series F at a $7.2 billion valuation in June 2025, saying the capital would support AI security, enterprise search, agents, partnerships and international growth. The round followed a larger Series E and reflected intense demand for systems that can make scattered workplace information usable by AI.

A funding announcement is the company's account of its momentum, not an independent product evaluation. Still, it identifies where investors see leverage: the platform that knows what an organization has, who may access it and which source is authoritative can supply context to many assistants and agents.

Why workplace search is difficult

Company knowledge is distributed across chat, drives, tickets, code and line-of-business systems. Federating live searches across every application creates inconsistent ranking and rate limits; building a central index introduces a difficult permission and freshness problem. An answer is only useful if it reflects the user's current access and links back to evidence.

Generative models add synthesis but also a new failure mode: a fluent summary can blend stale, duplicated or unauthorized material. Vendors therefore compete on connectors, ranking, identity, citations and governance as much as model quality.

How buyers should evaluate the category

Choose a set of real questions with known sources and users with different permissions. Measure retrieval accuracy, citation quality, freshness and whether access changes propagate quickly. Include questions the system should refuse because the user lacks access or the evidence is insufficient.

Then measure adoption against a workflow outcome: onboarding time, support resolution or hours spent finding policy. A compelling demo and a large funding round can open the conversation. Only permission-safe answers and repeated use prove that enterprise search has become infrastructure.

Sources & further reading

Social-media activity is treated as a signal of attention, not proof. Product claims are attributed to the linked publisher or announcement.