Team backlog estimation inside Jira Cloud — user guide, privacy policy and support.
App: Planning Poker for Jira Cloud
Last updated: 5 August 2026
Planning Poker is an Atlassian Forge app. It runs entirely on Atlassian’s infrastructure. It has no servers of its own, makes no outbound network calls, and sends no data to any third party. Everything the app stores stays inside the Forge platform, within your Atlassian instance’s data boundary.
The app declares no external permissions (no egress), which is what makes this guarantee verifiable rather than a promise.
To run an estimation game, the app stores the following in Forge storage:
| Data | Why |
|---|---|
| Game settings — name, card deck, estimation field, privacy flag, options | To run the game as configured |
| Participants — Atlassian account IDs and display names of the facilitator, estimators and spectators, plus the facilitator’s avatar URL | To show who is in the game and who may vote |
| Work items — Jira key and summary of the items being estimated | To show the backlog of the game |
| Votes — account ID, display name, the card chosen, the optional note, and the time the vote was cast | To reveal the votes and compute results |
| Timestamps — when a game was created, last changed and finished | To sort and filter the list of games |
| Final estimates | To show results and reports |
| The last voter list you used | To offer it when you create the next game |
The app does not store passwords, email addresses, payment details, or the contents of Jira work items beyond the key and summary shown above.
The app reads data from Jira only to display it to you, using your own Jira permissions. It never shows a user anything they could not already see in Jira. This includes:
Only the final estimate agreed by your team, and only when the facilitator clicks Save estimate. It is written to the field selected when the game was created: the story point estimate field, or the original time estimate.
The Suggest estimate feature runs a JQL search over your own Jira site for work items whose summary shares keywords with the current one and that already carry a story point value. It takes the median of those values and snaps it to the nearest card in the game’s deck. The search runs against your Jira instance with your own permissions. No AI or machine-learning service is involved, and no data leaves Atlassian.
When a Jira request fails, the app writes a diagnostic message to the Forge logs operated by Atlassian: the work item key, the HTTP status and the error text returned by Jira. Votes, notes and participant names are never logged. Logs are retained by Atlassian under the Forge platform’s own policy.
The app contains no analytics, no tracking pixels, no advertising and no third-party scripts. It does not profile users and makes no automated decisions about them.
In Forge’s key-value storage, operated by Atlassian. Hosting location and residency follow the Atlassian Cloud infrastructure your site runs on. See the Atlassian Trust Center for details.
Since the app processes data on behalf of your organisation, requests for access, correction or deletion of personal data are best raised with your own Jira administrator, who controls the site. We will assist administrators with any such request — contact info@vudia.site.
Material changes will be published on this page with an updated date above.
Vudia — info@vudia.site