Skip to main content

Buyer guide

Linear Team Setup Playbook: Triage, Cycles, and Pull Request Automation

Seven setup steps for a company bringing Linear to its first teams, or one whose leaders still ask engineers for status the tracker should already show.

Reading time
3 min read
Sources
13 cited
In this piece
Linear+41 Positive

Give Triage an owner, let pull requests move issue status, and review cycle carryover, so Linear's board matches what is shipping without anyone chasing updates. Decide those settings in the first week, then check the board every week.

When you need this

  • You are setting up Linear for the first teams, or it is in use and the board drifts from what is shipping.
  • Requests arrive from Slack, support, and other teams with no single place to sort them.
  • Leaders ask engineers for status that the tracker should already show.

Evaluation criteria

  1. Shape teams around how work is owned

    Which groups own a backlog and a rhythm of their own?

    Look forCreate one team per group that plans together. On Business and Enterprise, use sub-teams when a group needs its own team inside a parent with inherited cycles and labels. Linear issue identifiers belong to the team, so an agent assigned ENG issues sees every ENG project. Scope instructions by project or label as well.

  2. Give Triage an owner

    Who decides what happens to work that arrives from outside the team?

    Look forTurn on Triage for each team so issues from Slack, Sentry, and people outside the team enter an inbox. The owner accepts, declines, merges duplicates, or snoozes issues. Rotate the owner with Linear's Triage responsibility setting. Sourcebot routes every GitHub bug report into Linear with a Bug label and sets 2-day and 7-day SLAs by priority.

  3. Let pull requests move status

    Does anyone still drag cards by hand?

    Look forGitHub or GitLab connected, with branch names copied from Linear issues. Team status automation set for pull request opening, review requests, and merge, and rechecked after every workflow change. One team's merge-to-dev automation stopped moving issues to In Staging for several weeks while merge-to-main kept working.

  4. Use cycles as a rhythm

    How long is a cycle, and what happens to unfinished work?

    Look forCycles are optional per team, last 1 to 8 weeks, and roll open issues into the next cycle. A capacity dial estimates from the last three cycles' velocity. A Hacker News commenter who planned in another agile tracker for years found rollovers pushed cycles past capacity with no rebalancing, so review carryover at the start of each cycle. Turn off the optional automation that adds active issues to the current cycle if part-time members' work should stay out of it.

  5. Put projects and initiatives above issues

    Where does a leader see progress without asking?

    Look forProjects for multi-cycle work and initiatives for company objectives, which every member except guests can see. Cross-team metrics come from Insights on Business or Dashboards on Enterprise.

  6. Bring customer evidence in

    Which requests came from which customers?

    Look forTurn on Customer Requests. Intercom, Zendesk, Front, or Asks then add a request that links back to the source, so product teams can filter issues by customer count, tier, or revenue.

  7. Check the board weekly

    Is the board still true?

    Look forA weekly check of Triage age, cycle carryover, and In Progress issues without linked pull requests. Xiaofei Du's ten-person team stopped trusting its board after the backlog went stale and sprints exceeded capacity.

Questions to ask vendors

  1. Who owns Triage for each team this week, and how fast must an issue leave it?
  2. Which pull request events move which statuses in each team?
  3. Do we use cycles, and who reviews carryover at the start of each one?
  4. Which projects roll up to which initiatives, and who updates them?
  5. Which customer-facing tools file into Triage, and which team receives them?

Red flags

  • Issues In Progress with no linked branch or pull request.
  • A Triage inbox nobody owns.
  • Cycles that start above capacity because carryover is never reviewed.
  • A backlog of issues nobody has touched in months.
  • Status reports written by hand from data the tracker already holds.

Products in this piece

Sources

Official documentation
9
Public source
4
  1. Linear docs: TriageOfficial documentationlinear.app
  2. Linear docs: GitHubOfficial documentationlinear.app
  3. Linear docs: GitLabOfficial documentationlinear.app
  4. Linear docs: CyclesOfficial documentationlinear.app
  5. Linear docs: Sub-teamsOfficial documentationlinear.app
  6. Linear docs: InitiativesOfficial documentationlinear.app
  7. Linear docs: InsightsOfficial documentationlinear.app
  8. Linear docs: DashboardsOfficial documentationlinear.app
  9. Linear docs: Customer RequestsOfficial documentationlinear.app
  10. Sourcebot: How we reached bug zero using Linear (Feb 19, 2026)Public sourcesourcebot.dev
  11. GitHub issue on linear/linear: pull request merge automations stopped moving issue status (Sep 9, 2026)Public sourcegithub.com
  12. Hacker News: Linear cycles roll unfinished work forward without rebalancing (Aug 8, 2025)Public sourcenews.ycombinator.com
  13. Xiaofei Du: We stopped paying for Linear (Aug 17, 2026)Public sourcexiaofeidu.com