Sprint Health Check

Assess sprint status mid-cycle and intervene early before problems compound.

---

When to Run

• Standard: Mid-sprint (day 4-5 of a 10-day sprint)
• On-demand: When you sense something is wrong
• Signals that trigger an emergency check: Blockers open > 2 days, scope added after day 2, 3+ tickets not started by mid-sprint

---

The Five Health Signals

1. Scope Integrity

Did we add or change work after sprint planning?

Status	Condition
🟢 Healthy	0-1 items added; scope stable
🟡 Watch	2-3 items added; team aware
🔴 At Risk	4+ items added; or any item added without removing something

Questions to ask:

• What was added, and why?
• Did the team agree to the addition, or was it assigned top-down?
• Was anything removed to compensate?

---

2. Progress Rate

Are we on track to complete committed work?

Simple burn-down check (by mid-sprint, you should have completed ~50% of story points or tickets):

Committed points: [X]
Completed by mid-sprint: [Y]
Expected by now: [X * 0.5]

Status:
  Y >= X * 0.5: 🟢 On track
  Y >= X * 0.35: 🟡 Behind, watch
  Y < X * 0.35: 🔴 At risk — intervene

Note: Don't treat story points as precise. Use as a signal, not a measurement.

---

3. Blocked Work

Is work stuck waiting on something?

Status	Condition
🟢 Healthy	0 blockers, or blockers resolved same day
🟡 Watch	1-2 items blocked < 2 days; owner identified
🔴 At Risk	Any item blocked > 2 days, OR blocker has no owner

Questions to ask:

• What is blocked and why?
• Who owns unblocking it?
• Is this a dependency problem we need to escalate?

---

4. Work in Progress (WIP)

Are people focused, or is everything partially done?

Too much WIP = context switching = nothing gets done.

Healthy WIP ceiling: Each engineer should have at most 2 items In Progress simultaneously.

Status	Condition
🟢 Healthy	Most engineers at 1-2 items in progress
🟡 Watch	1-2 engineers at 3 items; discussing in standup
🔴 At Risk	Multiple engineers at 4+ items; tickets aging in "In Progress"

---

5. Team Morale Signal

How is the team feeling?

Quantitative morale check (1-5 scale, anonymous):

• Energy: How energized do you feel about the work this sprint?
• Clarity: How clear is the goal and what "done" looks like?
• Confidence: How confident are you we'll complete the sprint?

Thresholds:

Avg Score	Status
4.0+	🟢 Strong
3.0–3.9	🟡 Worth discussing
< 3.0	🔴 Needs immediate attention

---

Sprint Health Report Template

## Sprint Health Check: [Sprint Name/Number]
**Date**: [mid-sprint date]  
**Team**: [name]  
**Reported by**: [EM/TL name]

### Overall Status: 🟢 / 🟡 / 🔴

| Signal | Status | Notes |
|---|---|---|
| Scope Integrity | 🟢/🟡/🔴 | [items added: X, removed: Y] |
| Progress Rate | 🟢/🟡/🔴 | [X/Y points done at midpoint] |
| Blocked Work | 🟢/🟡/🔴 | [# blocked, longest-blocked item] |
| WIP Level | 🟢/🟡/🔴 | [avg WIP per engineer] |
| Team Morale | 🟢/🟡/🔴 | [avg score if surveyed] |

### What's at Risk
[Specific items or work streams that may not complete]

### Interventions

| Action | Owner | By When |
|---|---|---|
| [descope / unblock / pair on / re-plan] | [name] | [date] |

### Decisions Made
[Any scope cuts, priority changes, or re-assignments agreed in this check]

### Outlook
[ ] Sprint will complete as committed  
[ ] Sprint will complete with agreed scope reduction  
[ ] Sprint is off track — escalating to stakeholders

---

Intervention Playbook

If scope was added

1. Make it visible: add to board, count the scope change
2. Ask: what comes out? (LIFO — last item added is first to descope)
3. If stakeholder-driven: loop in PM; protect team from silent scope growth

If progress is behind

1. Identify the largest in-flight item — is it actually too big?
2. Split it: can we deliver partial value and carry the rest?
3. Check for hidden work: undocumented tasks, context switching, technical debt payments

If work is blocked

1. Name a blocker owner by end of standup
2. If blocked on external team: manager escalates immediately, not "let's see"
3. If blocked on technical unknowns: timebox the investigation (1 day max before re-planning)

If WIP is high

1. Stop starting, start finishing — team pull policy
2. Ask each engineer: "what would it take to close one of these today?"
3. Consider pairing to unblock the stalled item

If morale is low

1. Don't ignore it — low morale predicts low output within 1-2 sprints
2. 1:1s within 24h with each team member
3. Ask: is the work unclear? Is the work meaningful? Is the pace sustainable?
4. Consider: sprint scope cut to reduce pressure, or retrospective pulled forward

---

References

• Henrik Kniberg: Scrum and XP from the Trenches
• LeadDev: Sprint anti-patterns
• Daniel Vacanti: Actionable Agile Metrics (WIP and flow)

MCP Integration

Query in-progress items and their age:

jira_search_issues(jql: "project = PROJ AND sprint in openSprints() AND status in ('In Progress', 'In Review') ORDER BY updated ASC")

Count total sprint scope and completed items:

jira_search_issues(jql: "project = PROJ AND sprint in openSprints()")
jira_search_issues(jql: "project = PROJ AND sprint in openSprints() AND status = Done")

Identify blocked items:

jira_search_issues(jql: "project = PROJ AND sprint in openSprints() AND labels = 'blocked'")

Pull sprint details (dates, goal):

jira_get_issue(issue_key: "PROJ-sprint-board")  -- or query via sprint boards if available

Graceful fallback: Ask the team to paste the sprint board snapshot (stories, statuses, remaining days) as plain text. Apply the same Sprint Health Score analysis to the text input.