How Do You Design a Page for People Who Will Not Read It?

How do you design a page for people who will not read it?

You make the skim carry the meaning. Headings, opening lines, and the few bold phrases have to deliver the whole argument on their own, because for most visitors that is the entire page. Everything else is there for the minority who go deeper.

Designers know this in theory and then design as though it were not true. We do it too. It is far more satisfying to write a beautiful paragraph than to accept that most people will see six words of it.

So this piece is the uncomfortable version. What the research says people actually take in, and what that should change about how a page is built.

How little do people actually read?

Less than almost anyone designing a page assumes. Jakob Nielsen published an analysis on 5 May 2008 finding that users have time to read at most 28 percent of the words during an average visit, and that 20 percent is the more likely figure.

The provenance is worth stating because the number gets thrown around loosely. It came from academic research by Harald Weinreich, Hartmut Obendorf, Eelco Herder, and Matthias Mayer, published in ACM Transactions on the Web in February 2008, using data collected in 2005. Nielsen's analysis covered 45,237 page views from an instrumented browser study.

That is old data, and browsing habits have changed since. We cite it anyway because nothing in the intervening years suggests people got more patient, and because the more recent eyetracking work points the same direction. Treat it as a shape rather than a precise current figure.

Does adding words help the people who do read?

Barely. The same study found that users spend only 4.4 seconds more for each additional 100 words. At a normal reading speed that works out to roughly 18 percent of the added content actually being read.

That ratio is the one that should change your behaviour. Adding a paragraph does not add a paragraph of understanding. It adds about a fifth of one, spread thinly, while pushing everything below it further down the page.

Which means the cost of extra copy is not neutral. Every section you add makes the sections after it less likely to be seen. On a page with a specific job, that is a real trade rather than a stylistic preference.

What do skimmers actually look at?

The top, and then rapidly less. Nielsen Norman Group's eyetracking study of 15 April 2018, analysing over 130,000 eye fixations from 120 participants, found that 57 percent of page viewing time was spent above the fold, with 74 percent within the first two screenfuls, up to 2160 pixels.

Within those screens, attention lands on things that look different from their surroundings. Headings, numbers, short lines, bold phrases, faces, and anything that breaks the visual rhythm. Plain paragraph text is the last thing looked at and the first thing skipped.

This is why hierarchy is not decoration. It is the mechanism by which a skimmer decides what to sample. Our notes on visual hierarchy cover how to build that deliberately rather than by accident.

How should headings carry the page?

A reader should be able to read only your headings, in order, and come away with the argument. That is the test, and most pages fail it badly because their headings are labels rather than statements.

Overview, Features, Benefits, and Why Us tell a skimmer nothing. The same sections titled as what the content actually says would do the whole job. A heading that contains a claim or an answer is doing work. A heading that names a category is just furniture.

Write the headings first and read them alone before writing anything underneath. If the sequence does not make sense on its own, no amount of good paragraph writing will rescue it, because the paragraphs are the part being skipped.

What does bold text do, and what does it undo?

Bold text creates a stopping point, which is powerful and easy to destroy. Two or three bold phrases per screen give a skimmer somewhere to land. Ten bold phrases give them nothing, because when everything is emphasised the emphasis stops carrying information.

Bold the thing you would want someone to know if they read nothing else in that section. Not the keyword, not the product name, not the phrase you happen to like. The consequence.

The same discipline applies to every other attention device on the page. Coloured callouts, icons, oversized numbers, and animation all compete for the same limited attention. Using three of them in one section means none of them wins. Our piece on white space covers the other half of this, which is giving the important things room to be noticed.

Why does reading level beat clever writing?

Because skimming is already a low effort mode, and difficult sentences push the reader out of it entirely. Unbounce's 2024 Conversion Benchmark Report, drawn from over 57 million landing page conversions, found that simple copy converts 514 percent better than overly difficult copy in SaaS.

The breakdown is stark. Pages written at a fifth to seventh grade reading level achieved a 12.9 percent conversion rate in that dataset, against 2.1 percent for professional level copy. Unbounce also found that high converting pages use between 50 and 140 difficult words, meaning words of three syllables or more.

That last figure is the practical one, because it is countable. Count the long words on your page. If there are two hundred, the problem is not your layout. This applies with particular force to B2B, where industry language accumulates without anyone deciding to add it.

Where does visual design help skimming, and where does it hurt?

It helps when it creates difference and hurts when it creates noise. A section that looks visibly unlike the one before it gets sampled. A page of twelve equally weighted cards gets scanned for half a second and abandoned.

Where it hurts most is animation on scroll. Content that fades in as you reach it is designed for someone moving at reading pace. A skimmer moving fast sees a page of empty boxes, and some of them never fill. That is a real failure mode we run into constantly on otherwise well built sites.

Images earn their place when they carry information and cost you when they are decoration. A screenshot of the product tells a skimmer more in a second than three paragraphs. A stock photo of a team in a meeting tells them nothing and takes the space where the answer should be.

Does this mean you should write less?

Not necessarily. It means you should structure more. A long page can work perfectly well for skimmers if each section is self contained, opens with its own answer, and can be understood without the sections around it.

What does not work is a long continuous argument where the payoff sits at the end. That format assumes a reader who starts at the top and finishes at the bottom, and the data says that reader is rare. Put the conclusion first and the reasoning after it, in every section.

The other thing to avoid is hiding the answer inside a paragraph. If the important sentence is the fourth one in a block of text, it does not exist for most of your audience. Move it to the front or give it a heading of its own. Our notes on microcopy cover the small text that does more work than it gets credit for.

What would we change first?

Three things, in order. Read your headings alone and rewrite any that are labels rather than statements. Cut bold text back to two or three phrases per screen and make them the consequences. Then count the words of three syllables or more and halve them.

None of this requires a redesign. It is an editing pass on a page you already have, and it usually takes an afternoon. The reason it works is that it aligns the page with how people were always going to use it, instead of how we wish they would.

If you want a second pair of eyes on a page that is not landing, or help running that editing pass across a site, we are happy to take a look. You can find us at phoenix.studio.