How Do You Design a B2B Page for Someone Who Will Not Read It?

How do B2B buyers actually read a page?

They do not, mostly. They scan for the specific thing they came for, and they read properly only after something convinces them it is worth the effort. Designing as though the second behaviour comes first is the most common structural mistake I see on otherwise well built B2B sites.

This is not a complaint about attention spans. Somebody evaluating four vendors in an afternoon is behaving sensibly by scanning, because reading everything would be a poor use of their time. The page that wins is the one that lets them complete their scan fastest and then rewards the reading they do afterwards.

What follows is partly accessibility guidance, which turns out to be some of the most concrete published advice on readable text, and partly what I have learned about what a skim actually has to carry.

What does accessibility guidance say about line length?

The W3C is specific. In its understanding document for success criterion 1.4.8 on visual presentation, the stated width for blocks of text is no more than 80 characters or glyphs, with 40 for Chinese, Japanese, and Korean text. That is a concrete number in an area usually governed by taste.

The reasoning given is about losing your place rather than aesthetics. The W3C explains that for people with some reading or vision disabilities, long lines of text can become a significant barrier, because they have trouble keeping their place and following the flow of text, and that a narrower block makes it easier to continue to the next line.

That mechanism applies to everyone doing a fast scan, which is why I treat it as a general design constraint rather than only an accessibility one. A skimming reader is also jumping between lines and losing their place, just for different reasons. If you have ever built a wide desktop layout and found the body text unpleasant at full width, this is why.

Why do spacing and justification matter more than font choice?

Because they affect whether the eye can move, while font choice mostly affects whether a page feels right. The W3C criterion states that text should not be justified, and that line spacing should be at least space and a half within paragraphs, with paragraph spacing at least 1.5 times larger than the line spacing.

That last ratio is the underrated one. If your paragraph gaps are not clearly bigger than your line gaps, paragraphs stop reading as units, and a skimmer loses the structure that tells them where one idea ends. Many sites set generous line height, feel satisfied, and leave paragraph spacing at a value that erases the distinction.

Justification is worth naming because it keeps reappearing in designs that want to look editorial. Justified text creates uneven word spacing and rivers of whitespace that the eye follows downward instead of across. The W3C names it directly in the criterion, and I would treat it as settled for body text on the web.

Are these rules or preferences?

Read the criterion carefully, because the W3C is precise about this in a way summaries usually lose. Its note states that content is not required to use these values, and that the requirement is that a mechanism is available for users to change these presentation aspects, adding that the mechanism can be provided by the browser.

So the formal obligation is about user control rather than about your specific measurements. That distinction matters if someone is citing this criterion at you as a hard rule about your layout, because that is not quite what it says, and the level is AAA rather than the level most organisations target.

I still use the numbers as defaults, and I would encourage you to, because they encode real research about how people read under difficulty. The honest framing is that these are well grounded defaults rather than compliance requirements, and being accurate about that makes it easier to have a sensible conversation when a designer wants to deviate for a reason.

What should a skim reader get from headings alone?

The whole argument. If somebody reads only your headings in order, they should come away knowing what you do, who it is for, and what to do next. Headings that are labels rather than statements fail this test, and most B2B page headings are labels.

The difference is small to write and large in effect. A heading that says Integrations tells a skimmer nothing. A heading that says Connects to the CRM you already use tells them whether to slow down. The label version describes the section's topic, and the statement version delivers the section's point, which is the thing a skimmer is trying to collect.

A useful exercise is to strip everything except headings and read what remains as a paragraph. If it reads as a coherent summary, your page survives a skim. If it reads as a table of contents, everything you want understood lives in body text that a large proportion of visitors will never look at.

Where does the answer go on a page?

First, in the section where the question is raised, before any context. The instinct is to build up to the answer, because that is how we are taught to construct an argument. On a page being scanned, the build up is read and the conclusion is not, because the reader left during the build up.

This also happens to be how content gets used by AI answer engines, which look for a direct response rather than a narrative arc. The two audiences want the same structure for different reasons, which is a convenient alignment. I went into the writing mechanics in writing answer blocks that get cited by AI.

The design implication is that the first paragraph after each heading has a different job from the ones that follow it. It should be self contained, short, and complete. Give it visual distinction if the design supports it, and never start it with a clause that only makes sense once the rest of the section has been read.

What breaks skimming without anyone noticing?

Sections that all look identical. When every block on a page has the same rhythm, the same heading weight, and the same three paragraphs, a skimmer cannot tell which parts matter, so they either read everything or nothing. Usually nothing.

The second silent killer is the important sentence buried mid paragraph. Skimmers read the first line of a paragraph and the last, and skip the middle. Anything critical placed in the third sentence of a five sentence paragraph is, functionally, hidden. Moving it to the front of its own paragraph costs nothing and changes whether it is seen.

The third is decorative repetition of things the reader is trying to find. A grid of logos is the classic case, where the visitor is scanning for one name and the design has optimised for looking substantial rather than for being searched. I wrote about that specific pattern in showing integrations without building a logo wall.

Does designing for skimming hurt people who read properly?

No, if you do it by structuring rather than by shortening. The mistake is to assume skimmable means brief, so everything gets cut to bullet depth and the page becomes useless to the serious evaluator who arrived ready to be convinced.

What serves both is layering. A clear statement first, then the substance underneath it for whoever wants it. The skimmer takes the statement and moves on. The evaluator reads the substance and finds it worth their time. Nobody has to be served badly for the other to be served well.

It is worth remembering who your serious readers usually are. They are often the person who will champion you internally, and they need detail to build their case with. Stripping the page for skimmers removes exactly the material that person needed, and they are the one who decides.

How do you test whether your page is skimmable?

Give someone five seconds and then ask them what you do. That is the entire test and it is brutally effective. Show the page, take it away, and ask. If they cannot answer, no amount of further reading was going to happen anyway.

The second test is the headings-only read I described above, which you can do alone in two minutes. The third is to check your body text width against the 80 character guidance at your widest breakpoint, which usually reveals that desktop is worse than mobile, because mobile forced a sensible measure and desktop did not.

Run these before you commission a redesign. In my experience most pages that feel like they need a redesign need a heading rewrite and a spacing fix, which is a fraction of the cost. If the typography system underneath is the actual problem, that is a different and more structural job, which I covered in setting a typography scale and font sizes in Webflow.

What should you do next?

Copy your most important page's headings into a document, in order, and read them as a paragraph. That takes two minutes and it will tell you immediately whether your page survives a skim. Most people are surprised by how little their headings say.

Then measure one line of your body text at your widest breakpoint and count the characters. If it is well past 80, narrowing the content column is a single change with an outsized effect on how the page feels, and it needs no new design work.

After that, look at the ratio between your line spacing and your paragraph spacing, since that one governs whether paragraphs read as units. Three small changes, none of them a redesign. If you want a second pair of eyes on whether a page is doing its job for the people who never scroll past the first screen, reach out and let's chat.