Getting Started
This guide covers initial administrator setup, then shows end users how to create and interpret Delivery Trajectory's roadmap timeline and probability views. For a reference to individual controls and options, select the ? button in the app.
1. Install from the Marketplace
Install Delivery Trajectory from the Atlassian Marketplace on your Jira Cloud site. On install, the app pre-configures sensible defaults, so it is usable immediately with little or no configuration.
2. Configure (Jira admin)
Configuration is completed by a Jira administrator. Open Settings (gear icon) > Apps > Delivery Trajectory Settings.
You are not entering delivery data here. Instead, you tell Delivery Trajectory which existing Jira fields, statuses, and link types represent information it can use in a forecast. Delivery Trajectory is pre-configured with sensible defaults. Review and adjust these settings to reflect how your organization tracks work in Jira and the information you want included in the forecast.
- Team fields — Groups issues into delivery tracks. Issues in different groups can be scheduled in parallel, while issues in the same group share a delivery pace. Defaults to Project. Select another field if it better represents your teams or work streams.
- Estimation methods — Defines the Jira fields that can be used to represent work size. Add the sizing fields used by your organization, such as Story Points. These appear in the forecast's estimator selector. The built-in Card Throughput method is always available and needs no field configuration. It forecasts from the time between a team's completed work items rather than from a sizing field.
- Done statuses — Identifies completed work used to calculate historical delivery pace and forecast variability. Add statuses when you need to define completed delivery work more precisely, such as excluding terminal statuses like Abandoned or Cancelled. This setting starts empty. When no statuses are selected, Delivery Trajectory treats work in Jira's Done status category as completed.
- Dependency link types — Identifies issue links that mean one work item must wait for another, including the dependency direction. The list is pre-populated with Blocks. Add or adjust link types and direction to match your Jira workflow.
- Work item types — Defines which issue types represent the atomic unit of scheduled work. When a selected work item has children, it is scheduled as one unit and its children are not scheduled independently. By default, Delivery Trajectory schedules the lowest-level work items.
- Card type colors — Sets colors for issue types when using the card type color mode. Optional. Affects presentation only.
- Contention fields — Defines shared-resource constraints that prevent items with the same value from overlapping. This is optional. Configure the field and hierarchy level that define the contention scope.
- Earliest-start field — Identifies a date that prevents an item from being scheduled before a specified point in time. Optional.
For a detailed description of every option, see the inline notes on the settings page or the app's built-in help.
3. Use the tool
End users open the app from Apps > Delivery Trajectory in the Jira top navigation. Inside the app, the ? button opens a full help reference for everything below.
3a. Choose what to forecast (JQL)
At the top of the page, tell the app which issues to include, then click Run. You can either:
- Pick a saved filter from the dropdown (the simplest path — no JQL knowledge needed), or
- Type a JQL query. A good starting point is a single project, e.g.
project = YOURPROJECT.
The query defines the scope of the forecast. Completed items are not placed on the projected schedule. The projection starts from today and shows remaining work, while historical completed work is used to measure delivery behavior. The app fetches the matching issues, follows their dependencies, computes team velocities, and builds the projection (usually a few seconds).
Use the Estimator dropdown to switch between configured estimation methods (and the built-in Card Throughput method). Different methods can produce different projections because they measure work size differently.
3b. Timeline tab — the roadmap
The first tab is a deterministic schedule: "based on current historical delivery behavior, here is when each item is projected to start and finish." Each row is a team; work items appear as horizontal bars over time, and dependency links appear as lines connecting them.
- Hover a card for details and to highlight its dependency chain; click to open the Jira issue.
- Group by adds a second level of grouping (e.g. by epic). Toggling this is a good way to see the feature-level roll-up.
- Summarize / Detail All collapses each parent's items into a single summary bar (showing active vs. idle periods) or expands them.
- Color by switches card coloring between On Time (red/green versus due date), Card Type, and parent instance. A legend is shown for each mode.
- Due dates — Delivery Trajectory automatically uses Jira's built-in Due date field when one is present. Due dates do not drive the forecast. They show whether projected completion is likely to be on time.
- Scroll horizontally to see further out; Ctrl+scroll to zoom; use the minimap to navigate large timelines.
3c. Completion Forecast tab — probability of finishing by a date
The second tab runs a Monte Carlo simulation (1,000 iterations by default) and shows, for each item, a distribution of likely completion dates rather than a single date. A narrow distribution means the simulations produced similar outcomes, while a wide distribution shows more variation in likely completion dates. A narrow range is not always high confidence. When a team has little or no historical delivery data, simulations may rely on the same fallback values and produce an artificially narrow result. Parents can be expanded or collapsed to show or hide their children.
Enable Show target dates to overlay due dates on the distributions, and use On Time color mode to shade items by their probability of hitting the deadline. This view answers "how likely are we to finish by date X?"
3d. Duration tab — how long work will take
The third tab is the same Monte Carlo simulation shown as duration (how long) rather than calendar dates (when). For a parent, duration is the total elapsed time from its first child starting to its last child finishing — including gaps where the team is on other work — not a sum of individual efforts. Use this to answer "how long will this take?"
4. Keeping forecasts current
The projection is derived from your Jira data at the moment you run it; it does not refresh on its own. As priorities, dependencies, teams, and completed work change in Jira, simply re-run to get an updated forecast. Velocity also shifts naturally as new work completes, so projections evolve over time even without deliberate changes. There is no separate plan to maintain — the forecast comes from the work you already track. For the reasoning behind this, see the methodology.
5. Limitations & support
The projection is only as good as the data in Jira. A few things to keep in mind:
- The app can't account for work that isn't in Jira.
- A narrow JQL may omit important context (for example, the full backlog of a team you depend on), which can understate how long that team takes.
- Teams with little or no historical delivery data provide limited evidence for the forecast. Delivery Trajectory uses fallback values to keep work schedulable, but the resulting range may appear narrower than the available evidence warrants.
- The timeline shows one possible outcome; the probability views show the range.
If results look wrong, it is usually a configuration matter (team mapping, estimation fields, dependency link types) handled on the settings page. For anything else, see Support or email support@deliverytrajectory.com.