Where Jira Align sits in the delivery stack

Jira Align is Atlassian's enterprise platform for connecting business strategy to technical execution across portfolios, programs, and value streams. It aggregates team-level execution from Jira Software upward into program and portfolio rollups: Jira Fix Versions map to Jira Align "release vehicles," epics map to features and programs, and team workflow states feed the program board and portfolio roadmaps.

That architecture makes Jira Align excellent at alignment and visibility — Program Increment (PI) planning, the program room, portfolio roadmaps, OKRs, and dependency management at scale. What it consumes is your teams' output. What it does not do is turn that output into a probabilistic delivery date. That is the gap a focused forecasting layer fills.

How Jira Align forecasts dates today: deterministic, not probabilistic

This distinction is worth understanding precisely, because it defines exactly where a complementary tool adds value. Jira Align's predicted time-to-complete in the program room is a deterministic calculation, not a simulation. For estimated work it divides the remaining estimate (in team-weeks) by four and then by the number of teams in the program. For story-point work it divides the remaining points by the average velocity of the program's teams. In both cases the shape is the same: remaining work ÷ an average rate = one projected date.

Historically Jira Align exposed a "Monte Carlo Simulations" setting and a "Monte Carlo / Forecast Setup" menu. Per Atlassian's own documentation, that setting is now a legacy option that Jira Align no longer uses, and the menu was renamed to "Estimation Conversions," which performs unit conversions rather than probabilistic simulation. The forecast that reaches your program room is deterministic.

A single deterministic date is a fine planning signal, but it is fragile as a commitment, for two reasons:

• Velocity varies, and an average hides the spread. If your throughput swings sprint to sprint, a date built on the mean will be wrong a large fraction of the time — and the number itself tells you nothing about how wrong, or in which direction.

• Story points are relative, not absolute. A two-point item is not reliably twice the effort of a one-pointer. Summing ordinal estimates and dividing by an average compounds estimation noise directly into the projected date.

The math of probabilistic release forecasting

Monte Carlo forecasting answers the same question in a fundamentally more honest way: instead of one date, it produces a distribution of dates. It samples your team's real historical throughput (or cycle time) many thousands of times, "playing out" the remaining scope repeatedly, and records when each simulated run finishes. Aggregating those outcomes yields confidence levels you can actually plan against.

The result is expressed as probability, not certainty — for example, a 50% date (the coin-flip), an 85% date (a sensible external commitment), and a 95% date (the conservative promise). The underlying flow metrics — throughput, cycle time, and work-in-progress — are linked by Little's Law, so the same data that drives your delivery also drives the forecast.

Release Management, Roadmaps & Product Portfolio for Jira runs this simulation on your live Jira throughput — not on story-point velocity — so the forecast reflects how your teams actually deliver, and it recalculates continuously as new data arrives rather than being a one-time estimate.

Where the forecast layer fits the SAFe cadence

SAFe measures PI Predictability as actual business value delivered divided by planned business value, with a healthy target band of roughly 80–100%. Deterministic, average-based dates make that band hard to hit because the commitments entering PI planning are single-point guesses.

A probabilistic layer improves predictability in three concrete ways:

• It makes PI commitments realistic. Bring an 85% date into PI planning instead of an optimistic mean, and the plan that comes out is one stakeholders can trust.

• It sanity-checks the rollup. When Jira Align's deterministic projection and the Monte Carlo distribution diverge, that divergence is a signal worth investigating before you commit.

• It updates between PIs. Because the forecast re-runs on live throughput, RTEs see risk emerging mid-increment rather than at the retrospective.

None of this replaces Jira Align's role. The portfolio governance, the strategic alignment, and the program board stay exactly where they are. The forecast simply feeds them better numbers.

Beyond the forecast: capacity, dependencies, and scope in one place

A confidence-based date is only useful if you can act on it, so Release Management, Roadmaps & Product Portfolio pairs forecasting with the planning levers RTEs and Delivery Leads reach for:

• Capacity-aware planning that factors in PTO, public holidays, and work-in-progress, so projections reflect real team bandwidth rather than an idealized calendar.

• An interactive drag-and-drop dependency graph with critical-path highlighting and cross-release dependency mapping, so you can see what blocks what before it blocks a release.

• Intelligent prioritization using WSJF, RICE, and MoSCoW frameworks to scope releases deliberately.

• A "what-if" scope sandbox to experiment with scope and timing and commit in bulk — without altering live Jira data — plus undo/redo and audit logs for every board action.

Architecture and data residency

Release Management, Roadmaps & Product Portfolio for Jira is a native Atlassian Forge app that meets the Runs on Atlassian program criteria: all compute and storage stay inside Atlassian's cloud, no customer data is sent to external servers, and the app honors your data-residency region. For enterprises that chose Jira Align partly for governance and control, the forecasting layer carries the same trust posture.

When to use which

The short version: use them together. Jira Align answers "are we building the right things, aligned to strategy?" Release Management, Roadmaps & Product Portfolio answers "and what are the odds we deliver this by the date we promised?"

Try Release Management, Roadmaps & Product Portfolio for Jira

Release Management, Roadmaps & Product Portfolio for Jira is available on the Atlassian Marketplace with a free trial of either the Standard or Advanced edition. Install it alongside your existing Jira Align setup and bring a probabilistic date into your next PI planning session.

👉 Get Release Management, Roadmaps & Product Portfolio for Jira on the Atlassian Marketplace

For the business-stakeholder version of this argument — written for leaders weighing predictability and cost — see the companion article on the Divim blog: How Release Planning for Jira Complements Jira Align.