Short answer: Predictable delivery against a fixed government deadline comes from planning backward from the date using real available capacity (people, minus holidays, leave, training, and support duty) multiplied by an evidence-based velocity rate, then splitting scope into must-have and should-have tiers before the first sprint starts. Re-forecast every sprint and record why each commitment was made.
A statutory go-live date does not move when your team shrinks. Many agencies are now running the same mission with fewer people: a hiring freeze, a contract that was not renewed, two developers reassigned to an urgent program. The deadline written into a regulation, a grant agreement, or a legislative mandate stays exactly where it was. What changes is whether anyone notices the gap in February or in June.
Why do fixed-deadline government programs miss their dates?
Most missed dates are not caused by slow teams. They are caused by a plan built on headcount instead of availability. A program office sees six developers and eight sprints and assumes 48 developer-sprints of output. In reality, those six people share state holidays, annual leave, mandatory training, production-support rotations, and partial allocation to other programs.
The second cause is treating scope as one undivided commitment. When everything is a must-have, there is nothing to trade when capacity drops, and the first honest conversation with leadership happens after the date has already slipped.
How do you plan predictable delivery against a fixed deadline?
- Count sprints backward from the date. Start at the go-live date and walk back. Remove the sprints that are not feature capacity: user acceptance testing, the agency security review window, data migration rehearsal, and a final stabilization sprint. What remains is your real delivery runway.
- Convert people into available person-days. For each sprint, take working days per person and subtract holidays, planned leave, training days, and any percentage of time allocated to operations, support, or another program. A developer at 50% allocation is half a person for planning purposes, every sprint.
- Use a rate per available day, not raw team velocity. Divide recent completed points by the person-days that were actually available in those sprints. That rate survives staffing changes; a team-level velocity number measured with eight people does not describe a team of six.
- Plan a range, then tier the scope. Multiply available days by a low and a mid rate taken from your recent sprints. Rank scope into must-have (required by the mandate) and should-have (valuable but deferrable). If the must-haves do not fit inside the low case, escalate now.
- Re-forecast every sprint and record the decision. After each sprint, update actual availability and the rate, re-run the range, and note what changed and who agreed to any scope move. That record is what lets a program manager explain the plan to an oversight body months later.
What should the decision record include?
When leadership or an incoming program manager asks why the team committed to a given scope, the answer should already exist:
- Who was on the team, and at what allocation, for each sprint
- Which holidays, leave, and training days were subtracted
- The velocity rate used and the sprints it was drawn from
- The must-have and should-have lines, and every change to them with date and approver
This is not a compliance product or a certification; it is ordinary planning discipline. But in a climate where agencies are increasingly asked to show what happened, who acted, and what evidence supported a decision, a planning trail that already exists is far easier to defend than one reconstructed from memory.
Worked example: a mandated go-live with two fewer people
A state benefits-modernization team has a legislatively mandated go-live date. After removing UAT and a stabilization sprint, seven two-week sprints remain. Two contractor positions were not renewed, so the team drops from eight people to six.
Rate per available day. Over its last four sprints the team completed an average of 64 points per sprint with about 70 available person-days per sprint. That is roughly 0.91 points per available person-day. The slowest of those sprints ran near 0.80.
Available capacity for the runway.
- Nominal: 6 people × 10 days × 7 sprints = 420 person-days
- Two state holidays in the window: 6 × 2 = 12 days
- Planned leave already on the calendar: 18 days
- One mandatory training day per person: 6 days
- One developer allocated 50% to production support: 35 days
- Available: 420 − 12 − 18 − 6 − 35 = 349 person-days
Forecast range. Mid case: 349 × 0.91 ≈ 318 points. Low case: 349 × 0.80 ≈ 279 points.
Scope. Remaining must-have work is 300 points; should-have work is 110 points. The mid case covers the must-haves with about 18 points of margin. The low case falls about 21 points short. The 110 should-have points do not fit in either case.
Decision. The program manager moves the should-have tier to a post-go-live release, asks the product owner to split or trim roughly 20 to 25 points of must-have work, and reports the range to leadership in the first sprint, not the last. The decision, the numbers, and the approver are recorded alongside the plan.
Making this repeatable inside Jira
The method above breaks down when it lives in a spreadsheet that one person maintains and nobody else trusts. The backlog is in Jira, the leave calendar is somewhere else, and allocation is in a third place, so the forecast drifts out of date within two sprints. As agencies consolidate work on Atlassian Cloud, the natural fix is to keep capacity next to the backlog it constrains.
Sprint planning with Capacity Planning for Jira puts sprint and team capacity on one screen inside the Jira backlog, using past velocity, days-off, and allocation, so you see whether a sprint fits before you commit. Each sprint’s plan, and the inputs behind it, stay with the work instead of in a side file. The app is Cloud Fortified and runs on Atlassian’s SOC 2 Type II / ISO 27001 platform.
FAQ
How many sprints of history do I need to set a velocity rate?
Three to five recent sprints is usually enough to set a low and mid rate. Exclude sprints with unusual disruption, or note them, so the range reflects normal conditions.
What if the must-haves do not fit even in the mid case?
Escalate with numbers immediately. The options are the same in every agency: reduce must-have scope, add capacity, or change the date. Raising it early keeps all three options open.
Should support and operations time be planned as capacity?
No. Subtract it before you forecast. Treating support duty as feature capacity is one of the most common reasons fixed-deadline plans look achievable on paper and slip in practice.
Try a 30-minute deadline readiness check
Pick your next fixed date. Count the feature sprints left, list every holiday, leave day, training day, and support allocation, and compute your rate per available day from the last four sprints. If the must-haves do not fit in the low case, you have found your most important conversation of the quarter.
Try it: Install free from the Atlassian Marketplace · Help docs




Leave a Reply
Your email is safe with us.