Committing Twenty-Seven Items to a Date You Cannot Hit
A team has completed six two-week Sprints with velocity of 34, 41, 38, 52, 29 and 44 story points; the 29 came from a Sprint in which the office was closed for a week. The Product Backlog contains 27 items that sales has already promised to a customer for 30 November, five Sprints away, and the estimates total 260 points. Two of the seven developers will be on parental leave for two of those Sprints. When the Scrum Master asked for a forecast, the Product Owner said, 'Just commit to it - we always find a way, and the customer needs certainty.' The team has tracked throughput only in story points, never in items, and nothing below epic level has been decomposed except the top eight items.
1. What forecast would you give, and how would you present it?
Read the worked answer
Do the arithmetic honestly first. Mean velocity is 39.7 points per Sprint over six Sprints, with a range of 29 to 52. Five Sprints at the mean gives roughly 198 points against a 260-point scope - a shortfall of about 62 points, or more than one full Sprint of work. Hitting 260 would require 52 points per Sprint, which has happened once in six Sprints and will be even less likely with two developers away for two of the five. The 29-point Sprint should not simply be discarded; it is evidence about what disruptions do to this team.
That answers the question 'can we do all 27?' with high confidence: no. The useful forecast is therefore about how much, not whether. Present a range with a confidence level the Product Owner can act on - for example, 'we are about 85% confident of completing roughly 190 points, and about 50% confident of 200' - and translate it into items rather than points, because the customer buys items. With eight items decomposed and only coarse epics elsewhere, the range on item count is wide; decomposing the next two Sprints' worth of work will sharpen it.
Also state what the forecast assumes: no further disruption, stable team, no urgent production incidents. Forecasts presented as ranges with assumptions invite better decisions than a single number presented as a promise.
2. Sales has already committed the date to the customer, and the Product Owner wants the team to commit. What should happen now?
Read the worked answer
The date was committed outside the team, so the team cannot honestly 'commit' to it - but refusing to engage is not helpful either. The productive move is to make the gap visible and put the levers on the table, because with a fixed date something else must flex.
Lever one is scope: identify the smallest set of items that delivers the value the customer actually bought, ordered by the Product Owner, and forecast how much of the remaining 27 fits in five Sprints. Twenty-seven items are almost certainly not all essential on day one. Lever two is date: show what the same scope looks like at seven Sprints. Lever three is capacity: contractors, delaying the parental leave overlap, removing other commitments from the team - and note that overtime is not a capacity lever, it is a debt with a predictable repayment date and a quality cost the Definition of Done should not absorb.
The Scrum Master's job here is to facilitate that conversation with the data, not to shield the team from it or to broker a private deal. 'We always find a way' works when scope is negotiable; when scope and date are both fixed and capacity is not, the only remaining variable is quality, and that choice should be explicit rather than accidental. Sprint Reviews become the checkpoints where the forecast is refreshed against real progress.