What you’ll know by the end of this check

What Enterprise Search is and when to use it versus a regular connected chat
What the two-step setup actually involves (admin first, user second)
The real limits of Enterprise Search — so you know when it will and won’t answer your question

What Enterprise Search is

Enterprise Search is a pre-configured project that appears in your sidebar as “Ask [Your Org Name].” It’s designed for one specific job: finding and synthesizing knowledge scattered across your organization’s tools.

Think of it as a read-only intelligence layer over your company’s existing documents, messages, and records. Instead of hunting through Slack, Google Drive, and Confluence separately, you ask a question in plain language and Claude searches across all of them and synthesizes an answer with citations.

The use cases where it shines:

Catching up after time off — “What happened this week in the Platform project?”
Policy and process questions — “What’s our expense report process?” “How do I request a new tool?”
Cross-source synthesis — “What are the recurring themes in our customer escalations over the last quarter?”
Onboarding — “Who owns the authentication system?” “What deployment tools does eng use?”

These are questions you’d otherwise ask a colleague and wait for. Enterprise Search gives you an answer in under a minute, with sources you can check.

The two-step setup

Enterprise Search requires an admin to enable it before any user can access it.

Admin (Owner): Connects the organization’s core sources — at minimum a Documents connector (Google Drive or SharePoint) and a Chat connector (Slack or Microsoft Teams). Email is optional but recommended. Configures the project name (“Ask Emporia” instead of “Ask Your Org”).

Users: Once admin setup is complete, each person authenticates their own accounts for the connected services. The more connectors a user enables, the more comprehensive their search results will be.

If you can see “Ask [Your Org]” in your sidebar but it says something needs to be set up — that’s the admin step not done yet. If you’re the admin: the setup wizard walks through it.

The limits worth knowing upfront

Enterprise Search is only as good as what’s connected. A few gaps that matter:

GitHub is not an available connector. Engineering knowledge in repos isn’t surfaced by Enterprise Search. If your org’s institutional knowledge lives in Git, a curated knowledge base project is the current solve.
Live connector reliability varies. Each connected service is a live API call at query time. If a connector is having issues, that search leg fails silently or returns an error. This is the “before” state — a distilled internal knowledge base solves the fragility problem by removing live dependencies from the query path.
You can only access what you have permission to access. That’s not a bug — it’s the security model. But it means Enterprise Search for a new employee looks very different from Enterprise Search for a department head.

Things to try right now (5 minutes)

If Enterprise Search is set up at your org, ask it a question you’d normally need to ask a colleague: something about a policy, a project status, or who owns a particular system. Note what it cites. Then ask the same question in a regular chat with no connectors and compare the answers.

If it’s not set up yet: that’s a signal. Ask your Claude admin about the setup timeline.

The canonical version

Full official lesson at anthropic.skilljar.com/claude-101/383398 — includes the admin setup walkthrough and security model details.

Ready to verify this check?

Finish the official lesson, then come back and mark this check verified on your flight log.