Showdown runs planning poker inside Jira. Pull in work items with one click or JQL, vote on your scale with every card hidden until the reveal, talk the outliers out on a timer, and save finals to Story Points as the facilitator — so Jira permissions always apply. Pure Forge, zero external egress: your estimates never leave your Atlassian site.
Same scales, same stats, same save-to-Jira finish — pick the mode that fits how your team actually meets.
A live room. Everyone votes with cards hidden until the facilitator reveals — or auto-reveal fires the moment everyone estimating has voted. Discussion timer at 1:00, 2:00 or 5:00, re-vote rounds for the contested ones, and spectator mode for the curious.
No meeting required. Everyone votes the whole list at their own pace, across time zones. Vote counts are visible so you know who's done — but values stay hidden, even from the facilitator, until voting is closed.
A shared board for sizing a big backlog fast. Drag work items into buckets — or click an item, then a bucket — with everyone's moves syncing in about 2 seconds. Then set finals from buckets and bulk-save the lot.
Everything around the vote is designed to shorten the argument, not the thinking.
A discussion timer (1:00 / 2:00 / 5:00) keeps debates honest, and optional auto-reveal flips the cards the instant everyone estimating has voted. No "are we all in?" limbo.
On reveal: distribution, average, median, agreement %, consensus detection, and a suggested final — the nearest scale value to the median. The ? and ☕ cards never skew the numbers.
Fibonacci (0–100), T-shirt sizes, powers of two, 1–10, or your own custom scale of 2 to 20 values. Pick per session, not per install.
Wondering what a 5 feels like? A reference panel shows your team's existing work items at each scale value, pulled live from the estimate field — calibration built into the room.
Save finals to any number field — Story Points first — written as the facilitator, so Jira permissions apply exactly as they would in the work item itself. In-room context shows description, status, priority, labels and the current estimate.
Pure Forge with zero external egress: no webtriggers, no analytics, no AI, no third-party services. Sessions and votes live on your own Atlassian site — Taskhooker cannot see them.
From install to saved estimates in one sprint-planning slot. Setup is a one-time install; then run sessions in whichever mode fits the day.
A one-time install, done from the Atlassian Marketplace.
Choose the Jira site you want it on. The app installs through Atlassian's Forge platform — there is no separate download and nothing to configure on a server.
The app opens as a full page inside Jira. Anyone on the site can open it and join sessions — no per-user setup.
Interactive, Async or Bucket sizing; Fibonacci, T-shirt, powers of two, 1–10 or a custom scale; and the number field finals should save to (Story Points is first in the list).
Use the project quick-pick — Use current sprint or Open work items — or write JQL with autocomplete, validation and a preview. Up to 50 work items per session.
Teammates open the same app from the Apps menu and join your session — as voters, or as spectators. You created it, so you facilitate: only you can reveal, re-vote, set finals, save, close or delete.
The live meeting — hidden cards, one reveal, quick consensus.
The room shows the description, status, priority, labels and current estimate — plus a reference panel of your team's existing work items at each scale value. Votes stay hidden; you only see who has voted.
Click Reveal — or turn on auto-reveal and the cards flip the moment everyone estimating has voted. Spectators never hold up the room.
Distribution, average, median, agreement % and consensus detection appear instantly. If the room disagrees, start a 1:00, 2:00 or 5:00 discussion timer and let the outliers make their case, then run a re-vote round.
Accept the suggested final (the nearest scale value to the median) or set your own, then save it to the estimate field and move to the next work item.
Estimation across time zones — no meeting on anyone's calendar.
Teammates open it from the Apps menu whenever their day allows.
Each voter works through every work item independently. Vote counts are visible so you can see progress — values are hidden from everyone, including you as facilitator, until you close voting.
When the team is done, close the session. Every work item gets its full reveal: distribution, average, median, agreement % and a suggested final.
Walk the list, confirm or adjust each final, and save them to the estimate field as the facilitator.
Size a big backlog together in minutes on a shared board.
The board shows a bucket per scale value and the unsized work items waiting to be placed.
Drag a work item into a bucket, or click the item and then the bucket. Everyone works on the same board — moves sync to the whole room in about 2 seconds.
One click turns each work item's bucket into its final estimate. Adjust any individual final if the discussion moved it.
Save the whole board to the estimate field in one go, written as the facilitator so Jira permissions apply.
Spotted a bug or want an improvement? Raise it in a minute at the Taskhooker support portal — report a bug or suggest an improvement.
Showdown is sold through the Atlassian Marketplace, with pricing published on the listing and billing handled entirely by Atlassian — the app never collects payment details. Paid Marketplace apps include a 30-day free trial, so you can run real sessions with your team before paying anything.
No. Showdown is pure Forge and Runs on Atlassian: zero external egress, no webtriggers, no analytics, no AI, no third-party services. Sessions, votes and bucket assignments live in Forge storage on your own Atlassian site, and Taskhooker cannot access them.
Anyone on your site can open the app, create a session and facilitate it. The session creator is the facilitator — only they can reveal votes, start re-vote rounds, set finals, save to Jira, close voting and delete the session. Everyone else votes, or joins as a spectator.
Vote values are held server-side and never sent to anyone's browser before the reveal — you only see who has voted, not what. In Async mode it goes further: vote counts are visible, but values stay hidden even from the facilitator until voting is closed.
Jira Cloud rejects unbounded JQL queries. Add a bound such as a project, sprint or assignee — for example: project = ABC AND sprint IN openSprints(). Showdown's JQL input has autocomplete, strict validation and a preview, so you see exactly which work items a query selects before creating the session.
Yes — run the session on any scale. The agreed final is recorded in the session for every scale. Because Jira estimate fields are numeric, only numeric finals (Fibonacci, powers of two, 1–10, or a numeric custom scale) are written to the field; with T-shirt or other text scales the result stays in the session as your team's record. The special ? and ☕ cards never count toward stats on any scale.
Yes — spectator mode. Spectators see everything the room sees but do not vote, and they never block auto-reveal: the room reveals when everyone actually estimating has voted.
Yes — team-managed and company-managed projects both work. Showdown can estimate work items from any project whose work items you can browse, and finals are saved as the facilitator, so Jira permissions always apply.
