A healthy sprint is one your team can predict and repeat: it delivers roughly 80–90% of what it committed to, absorbs only small mid-sprint scope changes, and pushes little work into the next sprint. For a government PMO, five metrics tell that story — sprint predictability (the say-do ratio), velocity stability, commitment-to-capacity, scope churn, and carryover.

Every program manager knows the question that arrives the afternoon before a status review: “Are we on track?” The leaders asking it rarely read a burndown chart, and in a year of hiring freezes and flat budgets they have less patience than ever for a chart that needs a translator. They want a small set of numbers that mean something — and they want to trust those numbers are honest. This is a guide to the handful of metrics that actually describe a healthy sprint, and how to read them together.

Why isn’t velocity enough on its own?

Velocity — the number of story points a team completes in a sprint — is the metric most agencies reach for first, because it looks like productivity. But velocity alone is easy to misread. A team can post a high velocity by quietly pulling in easy work, or a low one because two people were detailed to another program. On its own, velocity tells you how much got done, not whether the team is predictable — and predictability is what a fixed-deadline, fixed-budget program actually runs on.

The shift that matters is from “how much did we do?” to “did we do what we said we would, and can we count on that next sprint?” That question needs a few metrics read side by side.

Which five metrics show whether a sprint is healthy?

Track these five together. No single one is a verdict; the pattern across them is.

• Sprint predictability (say-do ratio) — completed points ÷ committed points. Healthy range: roughly 80–90%. Consistently near 100% can mean the team is sandbagging; well under 80% means commitments aren’t trustworthy.
• Velocity stability — the rolling three-to-six-sprint average and, more importantly, how tight the band is. A steady 38–42 is far more useful for forecasting than a jumpy 25–55 with the same average.
• Commitment-to-capacity ratio — points committed ÷ realistic capacity for that sprint (after holidays, leave, and part-time allocations). Committing well above capacity is the single most common cause of a “missed” sprint.
• Scope churn — points added or removed after the sprint starts ÷ committed points. Some change is normal; above roughly 15% and the sprint goal is usually at risk.
• Carryover rate — unfinished points rolled to the next sprint ÷ committed points. Under 10% is healthy; a rising trend signals chronic overcommitment.

How do you calculate sprint predictability?

Predictability is the anchor metric, and it’s simple: divide the points a team actually finished by the points it committed to at sprint planning. A team that commits to 34 points and finishes 31 has a say-do ratio of 91% — reliable. The trap is reading it in isolation. A 91% say-do ratio means very different things depending on whether the team committed within its real capacity or padded the commitment down until it was easy to hit. That is why the commitment-to-capacity ratio sits right next to it: together they tell you whether the team is both honest about what it can do and disciplined about not promising more.

A worked example: two teams, one honest scorecard

Two teams on the same benefits-modernization program each average about 40 points over their last three sprints. The upcoming two-week sprint contains one federal holiday and has one developer out for three days of training, so real capacity is lower than usual.

Team A calculates that its reduced capacity supports about 34 points and commits to exactly that. Mid-sprint, a security patch adds 3 points of urgent work (scope churn of 9%). The team finishes 31 points and carries 3 into the next sprint. The scorecard: say-do ratio 91%, commitment-to-capacity 100%, churn 9%, carryover 9%. Every vital sign is in the healthy range.

Team B ignores the holiday and the training days, rounds its average up, and commits to 42 points. Mid-sprint, 8 points of new work slip in (churn of 19%). The team finishes 32 points and carries 10 into the next sprint. The scorecard: say-do ratio 76%, commitment-to-capacity 124%, churn 19%, carryover 24%. To leadership, this reads as a missed sprint.

Here is the part worth underlining: both teams did roughly the same amount of real work — 31 and 32 points. The difference in how healthy they looked came almost entirely from the commitment, not the coding. Team A subtracted the holiday and the training days before it promised anything; Team B didn’t. When you have fewer people and the same mission, that discipline is the whole game.

How do you keep the scorecard honest inside Jira?

These metrics are only useful if they’re captured the same way every sprint, and that’s where spreadsheets quietly fail — capacity gets calculated in a side file, commitments get made in Jira, and the two never reconcile. Moving the calculation into the backlog itself, as part of the broader shift to Atlassian Cloud and Atlassian Government Cloud (AGC), closes that gap. Sprint planning with Capacity Planning for Jira lets a team set commitments against real capacity on one screen — pulling past velocity, subtracting days off, and showing per-person allocation before anyone commits to a number. Because the plan and the work live together, the say-do ratio, the commitment-to-capacity ratio, and the carryover are all reading from the same source of truth, sprint after sprint.

That repeatability has a second benefit for public-sector teams. A capacity-based commitment leaves a record: who was allocated, which days off were subtracted, and why the team committed to N points. When a reviewer or an auditor later asks how a delivery decision was made, that documented decision trail is the answer — not a certification, but evidence of disciplined planning. The app is Cloud Fortified and runs on Atlassian’s SOC 2 Type II / ISO 27001 platform, so that record lives on infrastructure your security team already recognizes.

A practical next step

Before your next sprint planning, write your team’s last three say-do ratios and this sprint’s real capacity on a single line. If the commitment you’re about to make is above that capacity, change the commitment — not the estimate. That one habit moves most teams into the healthy range faster than any other change.

FAQ

What is a good say-do ratio for a government agile team? Roughly 80–90% completed-to-committed, held steadily over several sprints. Higher isn’t automatically better — a team that always hits 100% may be committing to less than it can actually do.

How many sprint metrics should a PMO report up the chain? Three to five is plenty. Predictability, velocity stability, and carryover answer most leadership questions; add commitment-to-capacity and scope churn when a program is under deadline pressure.

Is velocity a bad metric? No — it’s a useful input, especially as a rolling average for forecasting. It’s only misleading when it’s the only number you show.

---

Try it: Install free · Help docs

More from the Government Agile Delivery series