When & where we visualise our Atlassian data
You’ve got Jira. You’ve got Confluence. You’ve probably got a few apps, a handful of automation rules, and a small army of helpful page templates. But how healthy is your ecosystem really? This was exactly the topic for the latest ACE Solent event. Walking through the history of running Atlassian health checks. Moving from a reactive to proactive task, understanding current state of play, looking at future metrics and eventually how to automate the process.
What do we mean by a “health check”?
A health check is a structured review of your Jira and Confluence configuration, content, usage and security posture. It aims to answer three questions:
Is it safe?
Is it fast?
Is it useful?
“Safe” is your governance layer, from access policies to audit-ability. “Fast” is performance and operational hygiene. “Useful” is whether teams can actually find things, trust the data, and get work done without wading through clutter.
If you only remember one thing, remember this: a small set of repeatable checks, automated where possible, will outperform heroic one-off cleanups every time. This is why you will always hear long term architects talk about standardisation, automated processes and continuous improvements.
Pillar 1: Governance that people can live with
Start with the organisational view. Atlassian Guard gives you identity, policy and audit controls across products, which helps you spot risky behaviour, reduce shadow IT, and enforce sensible rules like preventing public links where they’re not appropriate. Running health checks against Guard can ensure your policies are constantly being adhered to as well as provide new opportunities for improvements.
Three quick wins:
Turn on organisation audit logs and export them regularly so you can track material changes, prove control, and feed SIEMs if you need to.
Review discovered apps and decide whether to bring them under management or leave them separate with eyes open. It keeps budgets honest and surprises low.
Define data security policies for sensitive projects or spaces. Block public links where required, and consider app access rules if you have strict data handling needs.
Pillar 2: Main app hygiene that actually improves performance & trust
Jira
Jira gets messy slowly, then all at once. Your check should cover:
Projects and lifecycle: archive inactive projects so they stop cluttering search, simplify boards, and reduce confusion.
Issues and attachments: archive rather than delete when content needs to step away from day-to-day work, then restore if needed.
Auditability: use product audit logs to see who changed what and when, and to speed up incident or config troubleshooting. Then correct your documentation to ensure the future you can easily understand your ecosystem.
Guardrails & Standards: Understanding common configuration can help limit the number of individual configuration items significantly. Being able to reuse and reduce helps with overall speed and can improve user experience.
If you want extra visibility, there are admin-focused apps that surface health signals, duplicates and unused config, with dashboards and clean-up helpers. Configuration Manager for Jira, Optimizer and Instance Auditor are some great additions worth evaluating.
Confluence
A healthy knowledge base is maintained knowledge, not just more pages. Your check should include:
Space and page analytics: see which spaces are actively used, what content people search for, and where engagement drops.
Archiving strategy: archive outdated content so it stops polluting search. Bulk archiving helps when you’re doing a quarterly sweep.
Review and approvals: page status is a good visual cue, but many teams also benefit from lightweight approval workflows for compliance or high-risk content. Comala Document Management is a brilliant app that helps track and manage knowledge lifecycles.
If your content model is growing, add dashboards that highlight “stale but important” spaces, and schedule reviews. Your future self will thank you.
Pillar 3: Metrics and automations
Metrics
Jira configuration footprint: number of custom fields by type, fields unused in the last 90 days, workflows and statuses in use vs. total defined. Use this to target clean-ups.
Project and space lifecycle: last updated date, ownership, and archive candidates.
Automation health: rules with the highest run counts, most failures, and most throttles. Use the audit logs and service limits to catch noisy rules before they bite. If you are running into limits, don’t forget to also take a look at marketplace apps that are also not limited and can provide the options you need.
Access and changes: number of policy exceptions, public link attempts blocked, and high-risk changes from the audit log over time.
Adoption and findability: top searches with zero results, top viewed pages vs. top updated pages. These often reveal missing docs or duplicated content.
Automate
Quarterly Jira cleanup: schedule a report of inactive projects, unused fields, and top failing automation rules. Refactor noisy rules before they hit service limits.
Confluence review nudges: every month, notify page owners of content older than your chosen threshold, then auto-archive after a grace period if no action is taken. Bulk archiving helps here.
Policy drift alerts: watch for public link usage or third-party app access on sensitive data and raise a review task for the owner.
If you’re pushing into heavier governance, consider a simple “content lifecycle” playbook for space owners: create, review, re-certify, archive. Keep it visible, keep it light.
