🔥 Design Webflow sites with AI
A logo gets looked at. A dashboard gets read, fast, often in bad light, sometimes by someone who's also driving or operating a machine. Those are different jobs, and they need different type.
Most branding conversations stop at the logo. The wordmark, the colors, the typeface that says "this is us." But for any company shipping a physical product with a screen, the font that actually carries the brand isn't the one in the logo. It's the one rendering a temperature, a battery percentage, or a heart rate at 2am on a 1.4-inch display.
Companies building for the internet of things run into this fast. The type on the device has to survive conditions no marketing deck ever accounts for: glare, motion, cheap panels, a user glancing for half a second. That's a harder brief than logo design, and it's the one that decides whether the product actually works in someone's hands.
A logo font can be weird. It can have a custom ligature, a modified letterform, a quirk that makes it memorable. It only appears in one place, at sizes the designer controls, and nobody has to read a paragraph of it.
Screen type on a device has the opposite brief. It has to stay clear at nine pixels tall, in a font size the user might not control, on a panel that could be cheap, glary, or low-resolution. Personality is secondary. Getting the number right on the first glance is everything.
There's a well-known cautionary tale here. When carmakers started putting touchscreens and infotainment systems in dashboards, a lot of them reached for Eurostile. It looks the part: squarish, technical, a little sci-fi, the kind of face that's been signaling "the future" in films since the 1970s. The problem, as Monotype has pointed out, is that Eurostile's square letterforms are hard to tell apart when you only get a split-second glance. Which is exactly the situation you're in behind the wheel.
So a font that looks perfect in the brand guidelines turned out to be a poor choice for the one place it mattered most. That gap between "looks right" and "reads right" is the whole game in device typography.
The faces that perform well on devices tend to be humanist sans-serifs. Frutiger is the classic example. Adrian Frutiger designed it for signage at Charles de Gaulle airport, where people read it at speed, at distance, at an angle. Steve Matteson of Monotype called it about the best choice for legibility you can make at small sizes, and Erik Spiekermann once called it the best general typeface ever made.
What makes it work is mechanical, not aesthetic. Open apertures, so the c doesn't close up into an o. A tall x-height, so lowercase letters stay big. Wide, distinct letterforms that don't blur into each other. Those are the same traits that show up in DIN, in Fira Sans, in the fonts people actually trust on instrument panels.
There's a specific failure that small-screen type has to design around: the I/l/1 problem. A capital I, a lowercase L, and the numeral 1 can look identical in a lot of typefaces. On a device showing a serial number, a dosage, or a sensor reading, that ambiguity isn't cosmetic. Fira Sans gets used in interfaces partly because it draws those three characters so you can tell them apart. When you're reading a value off a medical or industrial display, that distinction does real work.
Type foundries noticed this years ago and built whole families for it. Font Bureau made the Reading Edge fonts. Hoefler & Co. made ScreenSmart. Monotype made eText. David Berlow of Font Bureau described the goal plainly: make fonts as readable as Verdana that hold together down to nine pixels per em without losing their character. These are typefaces engineered for the exact environment a connected device lives in.
Part of that engineering is invisible. It's hinting, the instructions inside a font file that tell a screen how to snap letterforms onto a pixel grid so a stem lands cleanly instead of smearing across two pixels. Monotype also worked with MIT's AgeLab to build a faster way to test how legible a typeface actually is, measuring how quickly people recognize letters under glance conditions rather than guessing from how the type looks on a designer's retina display. That research keeps landing on the same answer: humanist shapes with open counters win, square grotesques like Eurostile lose, and the gap widens the worse the viewing conditions get.
Phones were the first squeeze. Then wearables made it worse. A watch face, a fitness band, a connected medical patch: these have a fraction of the real estate, and the data on them is often the whole point of the product.
Designing type for that constraint is its own discipline. You're balancing glanceability against information density, fighting reflections on a tiny panel, and usually working with a user who's moving. Teams building wearable health apps run into this constantly, because a heart rate or a glucose number that's misread is worse than no number at all. The typeface choice stops being a style decision and becomes part of whether the product is safe to use.
And it's increasingly a legal one. Under the European Accessibility Act and updated WCAG guidance, legibility in commercial digital products isn't optional. Typography is one of the most common things that fails an accessibility audit. So the font on the dashboard now carries compliance weight on top of everything the brand team cares about.
If you make connected hardware, you really have two typographic identities, and they don't have to be the same font.
There's the brand face, the one in the logo and the marketing, where you can be expressive. And there's the interface face, the one doing the unglamorous work on the device itself, where legibility beats personality every time. Plenty of strong tech brands run a custom or distinctive logo type alongside a boringly readable humanist sans for the actual UI. That's not a compromise. It's the correct answer to two different problems.
The companies that get this wrong usually do it by forcing the brand font onto the screen because consistency felt important. Then a user squints at a reading they can't quite make out, and the brand takes the hit anyway, just in a worse way.
The font on your packaging is trying to get noticed. The font on your screen is trying to disappear, to hand over a number so cleanly that the user doesn't think about the type at all. A logo wants to be seen. A dashboard wants to be read. Knowing which one you're designing, in which moment, is most of the job.
‍
