Weekly cross-space sweep
A weekly background job enumerates every space, counts pages per quality tag and per stale-age bucket, and stores raw, config-independent numbers. An hourly trigger only resumes an in-flight run or restarts a weekly run that has not fired for more than 8 days. Phases are enumerate → spaces → totals → complete. Spaces whose content has not changed since the last completed run are skipped; spaces with zero pages are dropped from the rollup; progress is checkpointed every 250 spaces; rate limiting (429) pauses and resumes rather than failing.
#Where to find it
Background (no UI beyond the Dashboard freshness line and Refresh button)
#Where you do it
Task | Where |
|---|---|
Turn Quality on or off for one space | Suite: Space Settings → Capable → General → Quality. Standalone: Capable Quality space page → Settings. |
Configure which rules run and how serious they are | Capable Quality → Manage → Rules |
See the overall state of content quality | Capable Quality → Dashboard (headed "Reports" on screen) |
#Settings that affect this
Setting | Where | What it does |
|---|---|---|
Inappropriate language | Rules → Language | - |
Non-inclusive language | Rules → Language | - |
Not updated in (Stale content) | Rules → Stale content → Not updated in | - |
#Worth knowing
Rules are configured once for the whole site. Opening Quality from a space scopes the counts to that space but edits the same global rule set, so a space administrator changing a rule level changes it for every space. There is no per-space rule configuration. The only per-space control is the Auto/Always/Never visibility switch.
Four gates decide whether Quality is visible, in this order: the site-wide Capabilities Visible/Hidden switch; the space set to "Never"; the space set to "Always" (Suite byline only. Ignored standalone); and the Advanced edition (Suite only).
On a space-scoped Quality page, "Issues over time" still shows site-wide numbers even though the header names the space.
Cross-space figures are up to a week old, and Refresh queues a background sweep rather than repainting the charts. A brand-new install shows "Space data not yet collected" with empty trend and top-spaces charts until the first sweep completes.
The weekly sweep skips spaces whose content has not changed since the last completed run, drops spaces with zero pages from the rollup, and silently skips a space that errors. None of which is surfaced to the user.
Only the last 60 weekly and 36 monthly trend points are kept; older points are dropped.
#Limits that apply here
Top spaces by issues
Cross-space data freshness
Weekly sweep pacing and budget
Space enumeration page size
#Related
How it actually behaves.
