Jira and Confluence Governance

Overview

Led Jira and Confluence governance across multiple departments — designing custom fields and screen schemes, building lifecycle automation for content archiving, and running the cleanup of what had accumulated before that structure existed.

Problem

Atlassian tools degrade by accretion. Every team that needs “just one more field” gets one, every project spawns its own scheme, and every page written for a project that ended three years ago stays in search results forever. Nothing individually is a mistake, and the cumulative result is a system where issue creation forms are full of irrelevant fields and Confluence search returns confidently wrong answers from abandoned pages.

The administrative cost compounds too: more schemes means more objects to maintain, and every small change requires checking what else it touches.

Approach

Treated it as a schema problem and a lifecycle problem, separately.

Schema consolidation — rebuilt custom fields and screen schemes so departments share structure where their needs are genuinely the same and diverge only where the work actually differs. Fewer objects to maintain, and issue forms that ask for what a team needs rather than the union of what every team has ever needed.

Content lifecycle automation — built automation to archive stale content on defined rules, so aging out is the default state rather than a cleanup project someone has to volunteer for. This is what stops the problem from re-accumulating.

Backlog cleanup — worked through the existing accumulation rather than only fixing the forward path, since automation applied to a system already full of stale content just preserves the mess more efficiently.

Architecture

Shared field and screen schemes across departments with documented divergence points, plus rule-based archival automation that acts on content age and activity rather than requiring manual review. Governance decisions are documented, so the next administrator inherits reasoning rather than a set of unexplained configuration objects.

Outcome

Reduced administrative workload by 35% while improving platform scalability — fewer schemes to maintain, cleaner issue creation, and search results that surface current documentation instead of abandoned pages.

Lessons Learned

Cleaning up an Atlassian instance without changing what let it get messy just resets a clock. The archival automation mattered more than the cleanup did — one is a project with an end date, the other is the thing that means there is no second cleanup project in two years.