An intro to Linear for product managers
How projects, cycles, issues and roadmaps fit together in Linear, and how to turn a PRD into work your team can ship.

Zander Whitehurst
8 min read
Linear has become one of the most popular tools for planning and tracking product work. It is fast, opinionated and designed around the way modern software teams actually build. For product managers, it can be the place where strategy turns into shipped work, but only if you understand how its pieces fit together.
I resisted switching to Linear for a long time. I had years of habits built around other tools, and I assumed it was just a prettier issue tracker. Then I ran a cohort project with a team who lived in it, and I noticed something. Nobody was asking where things were or what was happening next. The structure answered those questions for them. That is when I started teaching it to the PMs in our courses.
The core building blocks
Linear has a small number of concepts, and once you understand them the rest of the tool makes sense.
- Workspace. The top level container for your whole organisation. Everyone in your company usually shares one workspace.
- Teams. Groups of people who work together, such as a mobile team or a payments team. Each team has its own issues, workflow and settings.
- Issues. The individual units of work. Each one has a title, description, status, assignee and identifier, such as PAY-142.
- Projects. Collections of issues that deliver a specific outcome, like a new onboarding flow. Projects have a lead, target date, milestones and status updates, and they can span multiple teams.
- Cycles. Time boxed periods of work, usually one or two weeks, similar to sprints. Unfinished issues can roll over automatically into the next cycle.
- Initiatives. Higher level goals that group several projects together, such as improving activation this half.
- Roadmaps. The timeline view of your projects and initiatives, showing what is planned, in progress and complete.
Think of it as a hierarchy. Initiatives describe why. Projects describe what. Issues describe how. Cycles describe when the team will do the work.
Labels, triage, estimates and views
A few supporting features make that structure work day to day.
Labels add categories to issues, such as bug, feature or improvement, or the area of the product affected. Keep the list short. Too many labels and nobody applies them consistently.
Triage is an inbox for new issues that arrive from outside the team, such as bug reports, Slack requests or support tickets. Someone reviews each one and decides whether to accept, merge, decline or snooze it. Many teams rotate triage responsibility so it never becomes one person's burden.
Estimates let engineers size issues using points. Linear supports several scales, including linear, exponential, Fibonacci and t-shirt sizes. Estimates help you understand how much a team can realistically take on in a cycle.
Views are saved filters that show exactly the issues you care about. You might create a view of all high priority bugs across teams, or everything assigned to one project that is still in review. PMs often live in a handful of custom views.
Turning a PRD into projects and issues
This is where Linear becomes genuinely useful for product managers. A PRD describes a problem, a solution and a scope. Linear turns that into trackable work.
- Create a project. Give it a clear name, add the problem statement and goals to the project description and link the full PRD. Set a lead and a target date.
- Add milestones. Break the project into meaningful stages, such as internal beta, customer beta and general release.
- Break scope into issues. Work with your engineering lead to turn each part of the scope into issues. Aim for issues small enough to finish within a cycle.
- Connect it to an initiative. If the project supports a bigger goal, link it so leadership can see how the work rolls up.
Do the breakdown with engineers, not for them. They will spot technical tasks you would miss, and they will feel ownership over the plan.
Writing good issues
A good issue saves hours of back and forth. A vague one causes confusion, rework and frustration. Every issue should answer three questions: what needs to happen, why it matters and how we will know it is done.
A simple structure works well:
Context: Why this matters and who it is for.
Task: What needs to change.
Acceptance criteria:
- Specific, testable conditions that must be true
- Written from the user's point of view where possible
Links: Designs, PRD section, related issues.
Acceptance criteria are the most important part. "Improve the invite flow" is not testable. "An admin can invite up to ten colleagues at once from the team page, and each receives an email within a minute" is. Clear criteria mean engineers know when to stop, and reviewers know what to check.
Keep titles short and specific. Someone scanning a list should understand the issue without opening it.
Running cycles
Cycles give teams a steady rhythm. At the start of each cycle, the team agrees which issues they will take on, based on priorities and their recent capacity. At the end, they review what shipped and what rolled over.
As a PM, your role in cycles is to make sure the most important work is ready. That means issues are written, designs are linked and open questions are answered before planning begins. A cycle full of half defined issues will stall within days.
Watch the cycle progress graph. If scope keeps growing mid cycle, something is leaking in from outside the plan. If the same issues roll over again and again, they are probably too big or blocked by something nobody has named.
Keeping the roadmap current
A roadmap is only useful if people trust it. In Linear, that trust comes from keeping projects up to date.
Ask project leads to post regular project updates, marking each project as on track, at risk or off track with a short explanation. Linear can remind leads when an update is due. Those updates then appear on the project and roll up to initiatives, giving leadership a live picture without another status meeting.
Review the roadmap weekly. Move target dates when reality changes, archive projects that have been deprioritised and make sure new work is connected to the right initiative. A roadmap that is honest about delays is far more valuable than one that looks perfect and is wrong.
Linear's AI features
Linear has been adding AI features that help with the admin side of product work. These include suggestions during triage, such as likely labels, assignees and possible duplicate issues, as well as AI generated summaries of long issue threads and project activity. Linear also supports assigning issues to AI agents that can pick up work directly. The exact features available depend on your plan and settings, so check what your workspace has enabled.
Connecting Claude via MCP
One of the most useful setups for PMs is connecting Claude to Linear through the Model Context Protocol, or MCP. Linear provides an official MCP server, and once it is connected Claude can read and create issues, projects and comments in your workspace, within the permissions you grant.
That opens up some powerful workflows. Paste a PRD into Claude and ask it to propose a project structure with milestones and issues, each with acceptance criteria. Review the plan, adjust it, then ask Claude to create the issues in Linear. You can also ask it to summarise everything that happened in a project this week, or to find all open bugs related to a feature.
A PM I worked with recently used this to turn a twelve page PRD into a fully structured project in about twenty minutes. She still spent an hour refining the issues with her engineering lead, but the blank page problem had disappeared.
Always review what the AI creates. It will write confident acceptance criteria that may not match technical reality, and a quick check with your team keeps the backlog trustworthy.
Start small
You do not need to use every feature on day one. Start with teams, issues and projects. Add cycles once your team has a rhythm, then introduce initiatives when you need to show how work connects to bigger goals. Linear works best when its structure reflects how your team actually works, and that takes a few weeks to settle. We cover PRDs, planning and AI workflows like this in our AI Product Management Bootcamp.
Memorisely