FI-14
Team-Level Flag Concentration
Flags
Default severity: medium
ProcessOutputRisk
What it detects
Analyses whether flagged issues are disproportionately represented in sprint carry-over compared to non-flagged issues — confirms that flags are predictive of delivery failure. A high flag-to-carryover correlation validates the flag mechanism; a low correlation suggests flags are raised on non-blocked items or that carry-over is caused by something the flag mechanism does not capture. flagged_issues = issues WHERE flagged=true AT any point during sprint S non_flagged_issues = committed_issues - flagged_issues flagged_carryover_rate = COUNT(carried_over, flagged) / max(1, COUNT(flagged_issues)) non_flagged_carryover_rate = COUNT(carried_over, non_flagged) / max(1, COUNT(non_flagged_issues))
Detection formula
IF flagged_carryover_rate \< non_flagged_carryover_rate * config.fi.min_flag_predictive_ratio // default: 1.5 (flags should 1.5x more likely to carry over) THEN FLAG flags_not_predictive — team may be flagging wrong issues or flag mechanism is cosmetic IF flagged_carryover_rate > non_flagged_carryover_rate * config.fi.strong_predictive_ratio // default: 3.0 THEN INFO high_flag_predictiveness — flag mechanism is working well (positive signal) Report: flagged_carryover_rate, non_flagged_carryover_rate, ratio, interpretationExamples in practice
- A team shows analyses whether flagged issues are disproportionately represented in sprint carry-over compared to non-flagged issues while min flag predictive ratio is set to 1.5 (flags should 1.5x more likely to carry over).
- Example signal: Analyses whether flagged issues are disproportionately represented in sprint carry-over compared to non-flagged issues — confirms that flags are predictive of delivery failure.
Suggested response
Close the loop on flags—link blockers, set owners, and clear impediments with traceability.
Coaching playbook
Symptom
Analyses whether flagged issues are disproportionately represented in sprint carry-over compared to non-flagged issues — confirms that flags are predictive of delivery failure. A high flag-to-carryover correlation validates the flag mechanism; a low correlation suggests flags are raised on non-blocked items or that carry-over is caused by something the flag mechanism does not capture. flagged_issues = issues WHERE flagged=true AT any point during sprint S non_flagged_issues = committed_issues - flagged_issues flagged_carryover_rate = COUNT(carried_over, flagged) / max(1, COUNT(flagged_issues)) non_flagged_carryover_rate = COUNT(carried_over, non_flagged) / max(1, COUNT(non_flagged_issues))
Why it matters
When "Team-Level Flag Concentration" 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
Close the loop on flags—link blockers, set owners, and clear impediments with traceability.
Facilitation questions
- What system change would stop "Team-Level Flag Concentration" 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.