Portfolio Confidence Reporting for Executives: How It Works
This page explains how Release Management, Roadmaps & Product Portfolio for Jira rolls up committed dates and their confidence into a portfolio-level view for executive and leadership reporting. It is a factual companion to the article Turn Agile Velocity into Dates Executives Trust and to Keeping a committed release date honest: scope, dependencies, and capacity.
Why velocity and story points don't translate directly to a leadership date
Velocity and story points measure a team's internal rate of work. They are not, by themselves, a date, and they carry no statement of confidence. Release Management, Roadmaps & Product Portfolio separates the two concerns: story-point velocity remains the team-level unit of estimation, while the release date leadership sees is produced separately, as a probabilistic forecast (P50/P85/P95) drawn from real historical throughput.
The committed date, carried with its confidence
At the portfolio level, each release is reported as a committed date plus a confidence percentage (for example, "September 12 · 85%"), rather than a single typed-in date. Portfolio status is derived directly from that number:
On track — the committed date sits at or after the P85 line.
Attention — the committed date sits between P50 and P85.
At risk — the committed date sits before P50.
This status is what rolls up into portfolio and executive views, replacing a static Gantt bar with a number that carries its own honesty about uncertainty.
Why the number changes between updates
A confidence percentage shown once, at planning time, is a snapshot. Release Management, Roadmaps & Product Portfolio re-forecasts continuously as the three inputs most likely to move a release change: scope added after commitment, dependencies that surface late, and shifts in real team capacity. See Keeping a committed release date honest for the mechanics of all three. The practical effect for leadership reporting: a confidence figure that moves from 85% to 74% is visible the week it happens, not discovered after the release date has already passed.
Rolling up multiple teams or releases
Where a program spans multiple teams, the portfolio view blends each team's throughput distribution into a single program-level forecast rather than averaging separate team dates. This is the same capacity-blended forecasting used for multi-team releases generally, applied at the reporting layer so leadership sees one number per program rather than one slide per team.