Projects

A project is one piece of product work. It has its own people, links and memory.

A project is one piece of product work, like "UPI at Checkout" or "Refunds v2".

Each project has:

  • its people: Admins, Managers and Users,
  • its links to your tools: docs, tickets, channels, repos and boards,
  • its own memory, built from what happens in those tools.

Create a project

  1. Open Projects.
  2. Click + Add new.
  3. Type a name and click Create.

Anyone in the division can create a project. You become its Admin.

If you run more than one department, pick the department too. A User's new project goes into the division's first department.

Add people

Open the Participants panel on the project, or the gear button → People (the project's Settings; its Admins and Managers see it).

Two ways:

  • From a team: pick the team, then add the people you want, one by one. Each is on the project for that team.
  • By email: type an email and click Check, then Add (name) to this project. Someone in one team is added for that team. Someone in several teams: pick Add as part of which team? first. Someone in no team is just added. Someone new: click Invite by email. At their first sign-in, they land on this project.

Each person shows the team they are on the project for. Their notes and tickets count for that team's folder. If they leave that team later, a warning says so: remove them and add them again to put them in their new team.

Admins and Managers of the project can add and remove Users. Only Admins can make someone an Admin or a Manager, or change a role.

New people are also added to the project's Slack channels (matched by work email). They are invited as viewers to its Miro boards.

Only the project's people can open it

Being an Admin of the division is not enough. Someone on the project must add you. See Roles and permissions.

Click + Add on the project. You can add:

  • a pipeline stage,
  • the project's chat group (a Slack channel, a Google Chat space or a Teams channel),
  • a Confluence doc,
  • a Jira epic,
  • a GitHub repo,
  • a Figma file,
  • a Miro board,
  • a Notion page or database,
  • a Bitbucket repo or a Zoom meeting,
  • any other link.

A chat group for each project

A project with no chat group shows a suggestion to its Admins and Managers. It offers the chat tools your division has switched on in Settings → Tools and connected:

  • Create a group named after the project: #checkout-flow in Slack, or "Checkout flow" in Google Chat.
  • Link a group you already have: a Slack channel, a Google Chat space or a Teams channel.

Everyone on the project is added to the new group. Any linked chat group counts, so the suggestion goes away. Not now hides it for that project.

Every tool your division turned on has a place in + Add; a tool that is off is not offered.

Everyone on the project sees what SeamTrail learns from what you link. See Links.

Stages

Stages are optional. They split the work into steps, like Design, Build and Release.

  • Add, rename, reorder or remove stages.
  • Inside a stage, add team folders and steps. They can sit inside each other.
  • Link items at any level: the project, a stage, a folder or a step.

Folders in the pipeline

What you link to the project itself (not to a stage) is grouped, so the pipeline stays short.

  • Two or more of a kind make a folder: Jira, Docs, Channels, Code, Designs, Meetings or Links. A single item stays a card.
  • Each team on the project gets a folder. A ticket belongs to a team when it is in the team's Jira team (see Where its work lives). A team with no Jira team holds the tickets its people are assigned, matched by their Jira account, so a different name in Jira does not matter. Any other item belongs to the team it was added in, or else to the team of the person who linked it.
  • Every folder has the same + Add as the project page, for anyone on the project: a stage, a doc, an epic, a repo, a design, a board or any link. Added inside a team's folder, it belongs to that team, whoever adds it.
  • Click a folder to open it full screen. Back, or the path at the top, returns to the project.
  • Every folder and stage has its own address, like #/projects/<id>/folders/team-<id>, so you can share it or come back to it.
  • The numbers match everywhere. A folder's card in the pipeline and the folder itself show the same counts, and the project's card in the list shows the same as its Overview. An epic whose tickets are listed is a heading, not a task.
  • Activity & insights inside a folder or a stage show only what is about that folder or stage: its tickets, its items and, for a team, its people.

The Jira folder

  • Cards at the top count To do, In progress, Done and Blocked for the tickets your other filters show. Opened from a team's folder (Tasks), it lists that team's tickets only. Click a card to show only those tickets. Click it again to show them all.
  • Filters opens a row of filters below it: Status, Labels, Assignee, Team and Blocked. Each choice shows how many tickets it will show. Statuses come from your Jira space and labels from your tickets, so they match Jira. Clear all removes every filter.
  • Search by key or words, and group by epic, status or team. An epic shows how many of its tickets are done.
  • At the top, where the code is: one line per team, with each of the team's own environments (Stage and Prod, or Dev, Stage and Prod, as its repositories are set up) and how many of the folder's tickets reached it, with the last commit there.
  • Each row shows the key, the summary, the status and what blocks it, labels, the assignee, the team, and the furthest of the team's environments its code reached. Click a key to open the ticket in Jira.

A team's folder

A team's folder is the project page for one team.

  • Pipeline: Tasks (the team's tickets, from every stage of the project) and a folder for each kind of item added for the team or linked by its people.
  • Overview: the team's progress, tasks and code, from the tickets its people hold, and what is blocked.
  • Activity & insights: drift, blockers, risks, recent changes and history. Only what belongs to the team: what is about its tickets or its items, and what its people added themselves about no other team's work. A note about several teams' work, or one an AI assistant wrote about no team's work, stays on the project. Each memory is tagged with the team as it is learnt.
  • Recent merges and deploys: one closed row per repository of the team. Open one to see what git said lately, and link a pull request that named no ticket.
  • People: the team's people, with how many tickets each one has.

The team's pull requests are not listed yet. Only code items someone linked show up.

What the project page shows

PartWhat it shows
SummaryWhere the project stands, in a few lines.
ProgressTasks done per team, and how far their code went.
DriftWhat a change made out of date. See Drift.
Recent changesNew decisions, blockers, risks and status changes.
SuggestionsTickets your docs and chats mention that are not on the project yet. See Links.
PlanMatch tickets to the spec, or propose new ones. See Links.
HistoryWhat used to be true, and what replaced it. See Memory.
Promotion rulesWhat must wait for what before a deploy. See Promotion rules and deploy gates.

The tabs (Drift, Recent changes, Suggestions, Plan and History) show up only when they have something to show.

Daily digest

A project with a Slack channel can post a short summary there once a day: each team's progress and the open drift count.

A project's Admin or Manager turns on Daily digest to Slack on the project page. It is off until someone turns it on. It posts at 09:00 India time.

On this page