The Search Problem That Confluence Never Solved
Every enterprise knowledge base eventually becomes a graveyard of outdated documents, buried wiki pages, and search results that return the wrong version of the right file. Confluence, Atlassian’s long-dominant workspace documentation tool, built its reputation on organizing team knowledge – but search was never its strongest feature. Teams could store everything, and still find nothing.
Glean, the AI-powered workplace search startup, is now sitting at exactly that gap.
Founded in 2019 by former Google engineers, Glean built a product that connects across an organization’s entire software stack – Slack, Google Drive, Salesforce, Jira, GitHub, and yes, Confluence – and returns unified, context-aware search results that pull from all of them at once. The pitch is simple: stop searching in five tabs, and search once. What’s made that pitch increasingly persuasive is the addition of AI-generated summaries, assistant-style query responses, and workflow-aware ranking that surfaces what’s actually relevant to a given employee’s role and projects.
Why Confluence’s Search Problem Is Glean’s Sales Pitch
Confluence search has a well-documented limitation: it searches within Confluence. That sounds obvious, but in practice it means that when a product manager wants to understand the status of a feature, they’re checking Confluence for specs, Jira for tickets, Slack for context, and email for approvals – and stitching together an answer manually. Glean’s entire value proposition is eliminating that stitching process. It reads permissions, indexes content across connected apps, and returns a single answer with citations to the relevant documents.
For IT and knowledge management teams, this is where the competitive pressure on Confluence becomes visible. When a company deploys Glean, employees often start finding Confluence-stored content through Glean’s interface rather than Confluence’s own search bar. Over time, Confluence becomes a document repository rather than a knowledge navigation tool. The differentiation Atlassian has historically offered through Confluence’s structure – hierarchical pages, nested spaces, organized templates – starts to matter less when an AI layer sits on top and does the retrieval work instead.
Atlassian has responded by building its own AI features into Confluence, including Atlassian Intelligence, which offers in-product summarization and search augmentation. But Atlassian Intelligence is bounded by what lives inside the Atlassian ecosystem. It doesn’t reach into a company’s Salesforce data or ServiceNow tickets. Glean’s advantage is specifically its connectors – the company reportedly supports over 100 integrations – which means the more apps a company uses, the wider Glean’s utility gap over a single-vendor AI becomes.
The Enterprise Wallet Share Glean Is Targeting
Glean completed a funding round in 2024 that valued the company at $4.6 billion, a number that reflects investor conviction that enterprise AI search is not a feature – it’s a standalone product category. The company’s customer base includes large financial services firms, technology companies, and professional services organizations, all of which share a common trait: they have sprawling, disconnected software environments where finding institutional knowledge takes longer than doing the actual work.
The sales motion Glean uses is increasingly centered not on replacing Confluence outright, but on sitting above it. That’s a strategically smart position. IT buyers don’t want to migrate wikis. Procurement teams don’t want to negotiate out of Atlassian contracts mid-cycle. Glean doesn’t ask them to. Instead, it positions itself as an intelligence layer, which means it can be purchased as an addition rather than a replacement – lowering the internal resistance to adoption. Once adopted, though, the usage patterns tell a different story. If employees stop using Confluence’s search and start using Glean’s, Atlassian’s renewal conversations get harder.
This is the quiet part of Glean’s market strategy. The company is not announcing that it’s competing with Confluence. It’s announcing that it connects to Confluence. The distinction matters enormously in enterprise sales, where vendor relationships are political and switching costs are high. Glean doesn’t need to displace Confluence to win budget – it needs to become indispensable enough that, when renewal season comes, IT leaders start asking whether they’re paying for Confluence’s documentation structure, or for search functionality that now lives somewhere else. That question alone is eroding Confluence’s pricing leverage.
The Compounding Problem for Atlassian
Atlassian’s challenge here runs deeper than one competitor. The same AI-layer argument Glean is making against Confluence’s search is being made by Notion AI against Confluence’s document editing, by Guru against its knowledge verification features, and by a growing category of “AI assistant” tools that aggregate workplace data and generate answers rather than presenting links. Confluence was built for a world where the bottleneck was organizing information. The new bottleneck is retrieving and synthesizing it, and that’s a problem where Glean’s architecture – purpose-built for cross-app indexing with LLM-generated responses – has a structural head start over a tool originally designed around page hierarchies and wiki templates. For any company evaluating Glean right now, the honest question isn’t whether it beats Confluence at search – it’s whether Confluence can close that gap before the next budget cycle.
