CC-06
Outlier Cluster in Recent Window
Control Chart
Default severity: medium
ProcessMetricsOutputRisk
What it detects
Number of items above UCL in the recent window exceeds the historical average by the configured multiplier and meets the configured minimum count.
Detection formula
recent_window = config.cc.outlier_cluster_window_days // default: 30d min_count = config.cc.outlier_cluster_min_count // default: 3 multiplier = config.cc.outlier_cluster_growth_rate // default: 1.5x IF recent_outliers > hist_avg_outliers * multiplier AND recent_outliers >= min_count THEN FLAGExamples in practice
- 3–5 weeks
- Rolling averages, control limits, or band width diverge from expected statistical bounds.
- Recent control-chart windows show drift, clusters, or missing limit configuration.
Suggested response
Treat the metric as a control signal: confirm limits, windows, and cross-team spread before acting.
Coaching playbook
Symptom
Number of items above UCL in the recent window exceeds the historical average by the configured multiplier and meets the configured minimum count.
Why it matters
When "Outlier Cluster in Recent Window" keeps appearing, the team is signalling a repeatable process gap. Left unexamined, the pattern hides where work really stalls and makes improvement metrics harder to trust.
What you can achieve
Treat the metric as a control signal: confirm limits, windows, and cross-team spread before acting.
Facilitation questions
- What system change would stop "Outlier Cluster in Recent Window" from firing again?
- What do the cited issues have in common — same root cause or same workaround?
- Who owns the two-week experiment and how will we verify on the next import?
Run this rule against your own tracker data with Flow Analyzer.