In 2026, the average mid-sized business runs against hundreds of external URLs. Vendor portals to file orders. Client production sites for monitoring. Supplier login pages for pulling invoices. Knowledge bases that hold the answer to the question your support tech is asking right now. Most companies store these links in three places: each person's browser bookmarks, a shared spreadsheet that nobody trusts, plus someone's head.

The cost shows up daily. A vendor changes their portal URL on a Tuesday. By Friday, three people have called support asking why the link in the spreadsheet returns a 404. A client gets a new login domain. The account manager finds out from the client, not from a tracked record. Onboarding a new hire means walking them through fifty URLs from memory because nothing was ever written down in the right place.

The problem is that URLs are never the primary record. They are always attached to something else, a vendor, a client, a department, a project. When the URL gets divorced from that record, it loses meaning. A bookmark named "portal" tells you nothing six months later.

Every URL attaches to an entity record. A vendor's portal link sits on the vendor record, a client's login page sits on the client record, plus a department's wiki sits on the department record where the team actually looks. The same URL can attach to multiple entities when it serves more than one, plus the system tracks each relationship separately. No duplicate links floating loose in some shared folder nobody trusts.

Each link gets a purpose tag and a status flag. The "Support Portal" tag tells your team exactly what the link is for, separating it cleanly from a "Marketing Site" or a "Vendor Login" for the same company. Status flags answer one question fast: does this link still work? When a URL goes dead, the team disables it without erasing the history of when it was active or why it was used in the first place.

The search layer makes the whole thing findable in seconds. Type the vendor name plus "support" and the right URL surfaces immediately. Filter by tag, by status, or by attached entity to pull exactly the slice you want. One query pulls every "Knowledge Base" link across every department.