The layout was approved with an elegant serif font for the ingredient list on a label. On screen everything read perfectly, the thin serifs looked expensive. A week later the print run arrived, and the small text had partly lost its sharpness.
That's not a printing defect. It's a design mistake that didn't account for the physics of the material and the ink.
For anyone outside design, quickly:
First, briefly, what choosing between them means for perception, because that's part of the decision too, just not the main part.
Shaikh, Chaparro and Fox (2006) showed that serif fonts are consistently perceived as more solid, mature and formal, while sans-serif fonts read as neutral, without a strong character. Juni and Gross (2008) confirmed this with live text. The same satirical article was printed twice, once in Times New Roman, once in Arial. The Times New Roman version struck readers as angrier and more biting, even though the words were identical.
This sense of solidity isn't accidental, and it's far older than printing itself. The first letters with serifs appeared in Ancient Rome, carved into monuments like Trajan's Column, erected in 113 AD as a memorial to the emperor. Calligrapher and researcher Edward Catich devoted an entire book to this in 1968 and made a convincing case: before picking up the chisel, Roman craftsmen first painted the outline of each letter onto the stone with a flat brush.
The brush naturally flared at the ends of each stroke under pressure, and that flared shape became the serif once the carver simply followed it in stone. Serifs weren't originally decorative at all, they were a direct imprint of a writing tool, and specifically the tool used for state, imperial inscriptions. That's where the sense of officialdom and weight has carried through for two thousand years.
Modern research shows this effect still holds today, just with a different tool than a chisel. Henderson, Giese and Cote (2004) confirmed on a sample of 210 typefaces that letterform itself works as an independent signal of impression, regardless of what the text actually says.
If the question is "which font looks more solid," the science answers directly: serif.
But for packaging and labels there's a question that matters more than perception. Will this font survive real printing.
Simple physics is at work here: ink has viscosity, material has surface structure, and both parameters determine whether a fine letterform detail stays legible coming off the press, regardless of whether it's offset or flexography. This is about dot gain, the effect where printed ink covers more area than the layout intended.
Stojmirović, Pleština and colleagues (2021) in Applied Sciences tested how the material affects this, following the ISO 12647-6 standard. They printed text and fine lines on three materials: coated paper, uncoated paper, and OPP film, then used a densitometer and a profilometer to measure how much the letters deformed. On uncoated paper the distortion was noticeably worse, because ink soaks unevenly into the material's pores. On coated paper and film the shape held far more precisely, because the ink stays on the surface.
Mahajan and Arulmozhi (2025) in the Multidisciplinary Science Journal tested the second variable, ink viscosity, in the printing of labels and cardboard boxes. They prepared inks at three different viscosities, printed at speeds from 5,000 to 9,000 sheets per hour, and used a microscope to measure how precisely the dot was reproduced on the print. The ink at 35 Pa·s viscosity gave the sharpest reproduction; thinner inks spread noticeably more.
Material and ink viscosity both affect the outcome in any printing technology. The first study shows the effect of one variable, the second shows the other, but printing processes don't split into separate worlds of physics because of it.
Neither study addresses serif versus sans-serif directly. But together they explain the mechanism behind something anyone who has sent artwork to print already knows: the more fine detail and small protrusions a letter has, the higher the risk that printing won't preserve them. A serif font simply has more of those details, and the serifs themselves are exactly the fine elements that suffer first, whether from dot gain on an unstable substrate or from ink with the wrong viscosity.
Which leads to a practical conclusion. For a large headline on a label or box, the element read from a shelf several metres away, serif can work beautifully, the serifs there are large enough to survive printing. For mandatory information — ingredients, batch, expiry date — the risk is entirely different. EU Regulation 1169/2011 on food information requires a minimum x-height of 1.2mm on packaging, dropping to 0.9mm for small packaging under 80 cm². At those sizes, serifs chosen for the sake of brand gravitas are usually the first thing to lose sharpness on an unstable material.
Brand perception and print physics pull the decision in different directions. A good layout accounts for both at once, rather than picking a font once for the whole design and hoping for the best.