Linear's landing-page playbook (and why everyone copies it badly)

Linear’s landing page launched the dark-mode-with-glow aesthetic that has now eaten an entire category of marketing sites. Every developer tool, every AI startup, every “we built our infra and you can use it” pitch from 2023 forward is running some version of Linear’s playbook. Most of them are bad copies. A few (Vercel, Resend, Sentry’s most recent refresh) are good ones.

This piece is about which moves on linear.app are structural (worth copying) and which are surface (the things imitators reach for first and that don’t actually do the work).

Open linear.app in a new tab. Specifically the homepage and the Method page. We’ll wait.

The thing imitators copy: the dark gradient

The visible Linear aesthetic is a near-black background (closer to #08090A than to pure black), a soft purple-blue glow somewhere in the upper third of the page, and small white type set against it. Almost every imitation starts and stops here. The result is a sea of indistinguishable dark-with-glow sites where you can’t remember which one you were just on.

This is the wrong move to copy because the glow is doing approximately none of the work that makes Linear feel premium. The glow is atmosphere. It’s the lighting in a well-shot photograph. What actually makes Linear’s page work is the subject of the photograph, and almost no imitator has done the work to figure out what that subject is.

The subject is the type. The structure. The pacing. The discipline of having very little on the page. Let’s get into each.

Move 1: a type stack that earns the dark background

Linear’s type stack is three faces:

• Tiempos Headline (or a tightened internal cut of it) for display headlines.
• Inter for body and supporting copy.
• JetBrains Mono for accents: version numbers, code samples, the occasional in-line keyword.

Why this stack works on a dark background specifically: Tiempos Headline is a serif. Most developer-tool sites on dark backgrounds run sans-only stacks because they associate “developer” with “monospace and sans.” Linear running a serif headline on #08090A is the single boldest move on the site. It signals editorial intent: this site is going to read like a magazine, not like an SDK doc.

The bad version of the move is to take a free pseudo-serif (Source Serif, Lora, IBM Plex Serif) and try the same trick. It won’t land. Tiempos Headline is doing real work: the contrast in its strokes is high enough to hold up against a near-black background at large sizes without the thin parts disappearing into the dark. Free serifs tend to have lower stroke contrast and read as mushy when scaled up on dark backgrounds. If you can’t afford a quality display serif, run a sans like Sohne or Inter at higher weight, and don’t pretend.

The Inter body copy runs at 16px / 1.55 / weight 400, with one small concession to the dark background: Linear lifts the body color from pure white to roughly #E4E4E7. Pure white on near-black is too high-contrast for comfortable reading; it vibrates. The slight desaturation is invisible if you don’t know to look for it, and is the difference between body copy you can read and body copy you skim past.

Move 2: the grid backdrop is functional, not decorative

Pull up Linear’s homepage and you’ll see a faint dotted grid behind the hero. Most imitators see this, add a background-image: url('dots.png') to their hero, and move on.

The grid is doing two things, and both matter.

First, it’s compensating for the absence of imagery. Linear’s marketing pages have very few product screenshots and almost no illustrations. The grid gives the eye texture to settle on without competing for attention. A pure-dark hero with no texture would feel empty. A grid-backed dark hero feels considered.

Second, the grid scale is calibrated to the type. The dot spacing on linear.app is roughly 32px: exactly twice the body copy size, exactly 1/3 of the hero headline size. The grid creates an implicit measurement system that the type sits inside of. Most imitators just slap on a 16px or 24px dot grid and the result reads as visually arbitrary because nothing else on the page matches its rhythm.

If you copy the grid backdrop, calibrate the dot spacing to your type system. If you don’t have the patience to do that, skip the grid and run flat dark.

Move 3: editorial pacing, not feature dumping

The structural move that separates Linear from its imitators is the most invisible one and the hardest to copy: the page is paced like an essay, not a brochure.

Look at the section flow on the homepage. Each section is a single idea. The headline states it. One or two sentences of body copy support it. There is usually one piece of evidence: a small product clip, a quote, a numerical claim. Then the next section. The whole page reads in about two minutes if you don’t open any of the links.

Compare to a typical SaaS marketing page, which crams feature lists, FAQ sections, integration logos, customer carousels, three different CTAs, a “compare to competitors” matrix, and a footer that’s a sitemap. That page reads in either two seconds (you bounce) or twenty minutes (you’ve decided to evaluate the product). There is no middle.

Linear’s pacing assumes you’re going to scroll through it once on a Sunday with a coffee, get the worldview, and then make a decision about whether to click through to the docs or close the tab. That’s an editorial mindset, not a marketing one. Imitators don’t make this trade-off: they want both the editorial mood and the feature dump, and the result is dark-mode marketing pages that have lost the discipline that made the original work.

Move 4: the product appears late and small

When Linear does show the product, it appears small, embedded in a section, with one specific feature framed at a time. The product is never the centerpiece. The worldview is the centerpiece; the product is the consequence of the worldview.

Imitators tend to do the opposite. They take the dark hero and slap a large product screenshot in the middle of it. The screenshot is the centerpiece, the dark gradient is the frame. The structure is upside down from Linear’s.

There’s a reason Linear’s approach works for them: their product is opinionated and the opinion is the sell. “We have built an issue tracker for people who care about craft” is the pitch. The product is the punchline. If your product’s pitch is “we have a faster CRM than HubSpot,” the worldview-first move is wrong for you. Show the speed. Be the brochure. There’s no shame in it.

But if your product’s pitch is closer to Linear’s (opinionated, taste-driven, “we did this because the existing tools were wrong”) then the editorial structure is the move. Don’t copy the dark gradient. Copy the pacing.

What imitators usually miss

The single biggest miss is treating the Linear playbook as a visual aesthetic when it’s actually a content discipline. The visual elements (dark, glow, serif headlines, monospace accents) are the consequences of three deeper choices:

1. The marketing copy is editorial, not promotional.
2. The page assumes a reader, not a prospect.
3. The product is the consequence of the worldview, not the subject of it.

If you make those three choices, the visual aesthetic falls out naturally, and you can pick any color palette you want. If you don’t make those three choices, no amount of dark mode and purple glow will save the page.

What to steal versus what to skip

Steal:

• The mixed type stack (display serif + sans body + mono accent). This is the most underused structural pattern in current developer-tool marketing.
• The grid backdrop calibrated to your type system. Cheap, considered.
• The editorial pacing: one idea per section, two sentences of support, one piece of evidence, move on.
• The “lift body copy off pure white” trick. #E4E4E7 instead of #FFFFFF on dark backgrounds.

Skip:

• The dark-with-purple-glow gradient. It is the most exhausted move on the web right now. If you must do dark mode, run flat dark with a single accent color. The page will read more confidently.
• The product-screenshot-in-hero pattern that most imitators apply to Linear’s playbook. Linear specifically doesn’t do this. Don’t add it back.

Linear’s landing page is hard to copy because its strength is what it leaves out, not what it puts in. The discipline to put one idea on the screen at a time, in a category that has trained users to expect feature dumps, is the actual move. Everything else (the type, the grid, the glow) is the consequence.