Linear makes the queue faster, and the queue was never the problem

The roadmap was fiction, and everyone in the room had helped write it

I watched a product team at a fintech company plan a quarter in about ninety minutes, which is fast, and they could move that quickly because their tooling is excellent. Linear, on every laptop. Cycles armed. The head of engineering pulled up a throughput chart and it went up and to the right, and the team has genuine affection for this software in a way I have almost never seen for Jira. Then someone asked how the new merchant reporting feature could ship at all, given that it needed sign-off from the risk committee, which meets every other week and had three agenda slots left before the freeze. The room went quiet in the specific way rooms go quiet when the constraint has been sitting outside the tool the whole time. The chart said the team was getting faster every cycle. The committee said nothing was going anywhere. Both were true, and only one of them was being measured.

That meeting is the argument. Linear is a superb instrument for the part of software development that happens inside the tracker, and the part that decides whether the work was worth doing mostly happens elsewhere. Speed of transit is not speed of arrival. The tool optimizes one and lets you mistake it for the other, and it does this while being, I want to be clear, the best-made product in its category.

What opinionated software actually installs

Linear's founders, Karri Saarinen, Tuomas Artman, and Jori Lallo, started the company in 2019 on an explicit bet against configurability. Jira offers you a thousand dials and you spend admin hours getting most of them wrong; Linear ships with a method instead, cycles and triage and a small set of statuses, and the dials are gone. This is the honest source of the product's appeal: it is not that Linear is fast, it is that somebody with taste already made the decisions you were going to make badly. Adopting it installs an operating philosophy along with a tracker, and I have watched teams absorb that philosophy with the seriousness of converts. Cycle discipline arrives. Backlog hygiene arrives. So does a new dialect for describing what matters, and after a year the team's sense of what a good week looks like has been quietly defined by a startup in Helsinki.

None of this is a complaint. It is the product. But consider what got installed: a rhythm tuned for small teams shipping software at startup cadence, and the faith that visible velocity is the signal worth optimizing. For an org whose real bottleneck is a decision process that lives in a committee calendar, Linear will make everything around the bottleneck frictionless, and a frictionless queue feeding a blocked pipe does not drain the pipe. It fills it faster.

The pricing says who it is for

The current structure, per Linear's own price list:

Read the gates and you can see the intended growth path. Free is generous about people and stingy about history, two teams and two hundred fifty issues, enough to fall in love and not enough to run a department. Business is where the product starts wanting to be a system of record: private teams, guest access, the intake machinery that turns Slack messages and support tickets into issues, all of it behind the six-dollar jump. And the newest layer is metered separately. Coding sessions and Loops, the agentic features where AI does work inside the tracker, require AI credits, meaning the unit of purchase is no longer only the seat but the action. The company raised thirty-five million dollars from Sequoia in 2022 at a reported valuation near six hundred million and has taken more money since, and venture capital at that scale is not underwriting a beautiful tracker for five-person teams. It is underwriting the moment your support queue, your code, and your roadmap all flow through one metered system. Who benefits if you believe the demo? The demo is true, which is the elegant part. It is just not the whole meter.

The case for the dials being gone

The strongest counterargument is that Jira's flexibility was mostly a tax on the powerless. I have sat in enough enterprises to know the pattern: a bespoke Jira configuration built by a departed admin, defended by a process team, endured by everyone else, and the configuration still felt worse than Linear's defaults on every interaction. Linear is touched thousands of times a day, and shaving a second off each touch is real money at scale, and taste, unlike configuration, does not decay. For the org Linear was built for, a fast-moving team that ships and trusts itself, the opinions are correct. The rhythm is the point. A team that adopts cycles and reviews its velocity honestly will usually ship more, because it has started measuring the thing it can control instead of narrating the thing it cannot.

Both things are true. The best tracker in the category is the one most likely to tell you a satisfying story about the wrong variable.

It depends, then, on whether your constraint is the queue or the decisions feeding it. If it is the queue, buy the taste; there is nothing better at the price. If it is the decisions, Linear will make the queue around them pristine, and pristine is not the same as moving.

The variable worth watching is the credits meter, because it is a second meter, and second meters change incentives. When the system that files, triages, and closes the tickets is itself paid per action, the definition of an issue will drift toward whatever generates one. The committee still meets every other week. Its calendar does not appear in the chart.