Analyst(s): Nick Patience
Publication Date: April 3, 2026
Glean achieved a $7.2 billion valuation after doubling its ARR to $200 million in nine months, driven by an enterprise AI platform that has fundamentally moved beyond search. The platform now centers on a permissions-aware knowledge graph, leveraging its deep retrieval capabilities to power new features such as the Agentic Engine 2 and the Canvas co-authoring UI. Glean is strategically positioning itself with model neutrality, supporting more than 15 LLMs to counter hyperscaler lock-in and by offering Glean Protect Plus, a paid SKU that addresses the paramount enterprise concern of AI governance. The company’s future success depends on whether its knowledge graph depth can hold off an improving and natively integrated Microsoft 365 Copilot.
What is Covered in This Article:
- Glean doubled ARR to $200 million in approximately nine months, reaching a $7.2 billion valuation following a $150M Series F in June 2025.
- The platform’s third-generation Assistant, Agentic Engine 2, and Canvas co-authoring surface represent a material step beyond keyword search toward a multi-modal knowledge layer.
- Glean’s adoption of MCP as both server and host positions it as an integration hub, though competition for that role is intensifying.
- The model strategy has broadened to a 15+ LLM hub, with customers able to bring their own model keys – a deliberate neutrality positioning against hyperscaler and SaaS lock-in.
- Glean Protect and the new Glean Protect Plus SKU address the trust and governance gap that is a practical barrier to enterprise agent deployment.
The News: Glean has released a series of platform updates since our last coverage of the company, spanning its assistant, agentic engine, model hub, governance tooling and ecosystem integrations. Key developments include the launch of Agentic Engine 2 with adaptive planning and parallel sub-agent orchestration; a unified model hub supporting 15+ LLMs across Amazon Bedrock, Azure OpenAI, and Google Vertex; the Canvas co-authoring UI; proactive data governance via Glean Protect Plus; and expanded MCP support operating as both server and host. These changes accompanied the company’s revenue reaching $200M ARR – approximately double the figure from its fiscal year ending January 2025 – and a June 2025 Series F that valued Glean at $7.2 billion.
Glean Doubles ARR to $200M. Can Its Knowledge Graph Beat Copilot?
Analyst Take: Glean’s product trajectory since its founding has followed a consistent path: start with enterprise search, establish a permissions-aware knowledge graph as the core asset, and build progressively more capable reasoning and automation on top of it. The latest updates from the company confirm that this progression is accelerating. The third-generation Assistant now incorporates deeper behavioral personalization, such as writing style inference and task continuity across connected data sources, while Agentic Engine 2 introduces adaptive planning and parallel sub-agent orchestration for complex workflows. Canvas, Glean’s co-authoring surface, fills a functional gap that had made Microsoft 365 Copilot’s document-generation capabilities a routine customer objection.
Glean’s underlying premise, that the quality of enterprise search directly determines the quality of everything built on top of it, remains its organizing principle. Glean’s debug view of its knowledge graph signals illustrates the engineering investment in search scoring: click signals, dwell time, anchor scores, data source affinity, and user behavior data are all weighted before a document surfaces in results. The company’s argument is that this approach to relevance is what distinguishes its agents from those that sit on top of lower-quality retrieval.
Model Neutrality as Strategic Positioning
Glean’s model hub now supports 15+ LLMs accessible via Amazon Bedrock, Azure OpenAI, and Google Vertex, in addition to direct provider relationships with Anthropic, Google, and OpenAI. For most assistant interactions – roughly 70% of usage according to Glean – users default to auto-routing and the platform selects the model based on task context and personal graph signals. At the agent level, administrators and builders can specify models per agent and per workflow step, enabling cost control by routing simpler tasks to cheaper models.
The practical effect of this architecture is that large enterprise customers can use models they have already contracted for or that satisfy their data residency requirements, without requiring Glean to host or negotiate those relationships. This directly addresses a procurement-level concern in heavily regulated industries. Glean’s consumption-based pricing units, layered on top of its per-user base tier, create a parallel governance question. The introduction of consumption-based pricing units on top of the per-user base tier creates a parallel governance question: as agent workloads become event-driven and more complex, infrastructure costs are rising even if application-layer token consumption looks stable, per Glean’s own observation.
Governance Is a Differentiator – For Now
Glean Protect, the security and governance layer included for all customers, has been supplemented by Glean Protect Plus, a paid SKU that adds proactive event-driven notifications, AI-based policy enforcement on agent-generated responses, and integration with third-party security tooling such as Palo Alto and CrowdStrike. The practical pitch is that customers can give employees broad access to Glean’s knowledge graph without the anxiety of over-sharing or PII exposure, because the platform is scanning continuously and can suppress non-compliant content before it reaches the end user.
Glean’s timing could be good here, as the enterprise concern across the agentic AI market is not primarily whether agents can execute tasks; it is whether those tasks can be authorized, audited, and constrained at scale. Indeed, Futurum Research’s recently published AI Decision Maker survey revealed that security and data privacy are cited as the number one agentic concern across all five maturity stages in the survey – from those just experimenting to those that claim to be running autonomous ecosystems; security is an ongoing operational challenge that scales with deployment complexity.
Glean’s governance-first positioning resonates, but Salesforce and ServiceNow are making similar claims with comparable force. What will distinguish governance stories in competitive evaluations is specificity: which policies are configurable, at what granularity, and how audit trails are surfaced to compliance teams.
MCP: Useful Integration Mechanism, Not Yet a Moat
Glean’s support for MCP as both server and host is notable. As a host, Glean can coordinate external services, such as Canva, Box, Salesforce Agentforce, and others, without ingesting them into the knowledge graph, reducing crawling and synchronization costs while enabling real-time action. This broadens Glean’s potential surface area in a customer’s agent mesh and allows it to participate in the orchestration layer rather than simply being one of many data sources.
The caveat is that MCP adoption is broad. The question of whether Glean’s MCP host capability represents genuine orchestration leverage or simply parity with market practice depends on how customers weigh its knowledge graph differentiation versus pure connectivity. If retrieval quality is the moat, MCP is additive plumbing. If customers primarily want an integration hub, Glean is competing with a wider set of platforms that lack Glean’s search depth but have stronger workflow automation histories.
Business Context: Revenue, Funding, and Scale
Glean surpassed $100M ARR in its fiscal year ending January 2025 and reported $200M ARR approximately nine months later – a growth rate that places it among the faster-scaling pure-play enterprise software companies by that metric. A $150 million Series F led by Wellington Management in June 2025 valued the company at $7.2 billion, up from $4.6 billion at the Series E less than a year earlier. The company now employs more than 1,000 people, serves customers across 27 countries, and has seen its $1 million+ contract segment grow nearly threefold over the past year.
Customers are consuming more than 20 trillion tokens annually on the platform, with token consumption doubling in the prior quarter. New enterprise logos include Booking.com, Comcast, eBay, Intuit, LinkedIn, Pinterest, Samsung, and Zillow.
The Microsoft Copilot Question: Narrowing, But Not Closed
Microsoft 365 Copilot remains the most frequently cited competitive reference for Glean. The narrative Glean’s sales team leads with – vendor neutrality across LLMs, hyperscalers, and data sources – is a credible structural differentiator. Microsoft’s knowledge grounding is architecturally weighted toward its own ecosystem; SharePoint search remains a known weak point. Glean’s Knowledge Graph treats every connected SaaS application as a first-class data source, and the company invests in retrieval science accordingly.
That said, Microsoft is iterating. Copilot’s enterprise context grounding has improved, and it benefits from native integration with the productivity layer that most knowledge workers already occupy. Glean’s addition of Canvas and PowerPoint export is a pragmatic concession: the company knows it cannot simply point customers toward deeper search quality if Copilot appears to satisfy visible workflow needs. The more interesting question is whether Copilot’s broad deployment creates a ceiling on Glean’s penetration or, as Glean contends, whether the DAU/MAU gap in Copilot deployments leaves room for specialists. Evidence suggests both effects are real, and which dominates will likely depend on the degree to which a given organization’s knowledge work is multi-system versus primarily Microsoft-stack.
What to Watch:
- Microsoft’s cadence of improvement in Copilot’s enterprise context grounding is the most important competitive variable to track. If Microsoft closes the retrieval quality gap materially over the next 12 months, Glean’s core moat becomes harder to defend in Microsoft-heavy accounts.
- Consumption-based pricing is new to Glean’s model, and customers with complex agentic workloads are already seeing infrastructure cost increases. How Glean manages the CFO conversation around AI costs at renewal will be a test of commercial maturity.
- Glean Protect Plus, as a paid governance SKU, is a reasonable commercial decision, but enterprise procurement teams will compare it against both native governance features in Microsoft Purview and standalone DLP/DSPM (Data Security Posture Management ) vendors. The differentiation story needs sharper articulation.
- The $7.2 billion valuation reflects strong investor sentiment, but also sets a high bar for the IPO optionality that most late-stage enterprise software companies are building toward. Revenue growth will need to be sustained, and the shift from seat-based to consumption pricing adds revenue predictability risk.
See the press release on Glean’s most recent funding on the Glean website.
Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.
Other Insights from Futurum:
Is Workflow AI Now Native After Microsoft Embeds Copilot in Power Platform?
Glean Launches Enterprise AI Agent Platform – Report Summary
Author Information
Nick Patience is VP and Practice Lead for AI Platforms at The Futurum Group. Nick is a thought leader on AI development, deployment, and adoption - an area he has researched for 25 years. Before Futurum, Nick was a Managing Analyst with S&P Global Market Intelligence, responsible for 451 Research’s coverage of Data, AI, Analytics, Information Security, and Risk. Nick became part of S&P Global through its 2019 acquisition of 451 Research, a pioneering analyst firm that Nick co-founded in 1999. He is a sought-after speaker and advisor, known for his expertise in the drivers of AI adoption, industry use cases, and the infrastructure behind its development and deployment. Nick also spent three years as a product marketing lead at Recommind (now part of OpenText), a machine learning-driven eDiscovery software company. Nick is based in London.
