TaxonomyDrift Monitor

TaxonomyDrift Monitor is a web app (with a lightweight browser extension) that continuously checks whether your information architecture is quietly degrading as content grows. It connects to common CMS and knowledge bases (Confluence, SharePoint, Contentful, Zendesk, Notion) and analyzes how tags, categories, page titles, and navigation nodes are actually being used versus the intended taxonomy. It flags “drift” patterns like duplicate categories, inconsistent naming, orphaned pages, overstuffed buckets, and broken parent-child relationships. Instead of producing a one-time IA diagram, it provides ongoing IA health metrics, weekly change reports, and a prioritized cleanup backlog with suggested merges/renames and impact estimates (e.g., which pages and search queries will be affected). The AI component helps cluster near-duplicate labels and propose normalized names, but the app stays realistic: it requires human approval and produces auditable change plans rather than auto-editing your system. The MVP focuses on one or two integrations and a clear “IA debt” dashboard so teams can prove progress and prevent regressions after reorganizations, migrations, or rapid content publishing.