Every visitor on your website is hunting. They arrived with a target, a limited budget of attention, and a forest of pages between them and it, and every click is a decision about whether a trail is worth following.

This is not a metaphor I reached for because it was handy. It is the literal research model: in 1999, Peter Pirolli and Stuart Card at Xerox PARC published information foraging theory in Psychological Review, which treats people finding things on the web the way biologists treat animals hunting food.

This piece goes deep on one rule from How Humans Read Websites. The short version there: links need scent.

How a forager decides

Pirolli and Card borrowed their model straight from optimal foraging theory in biology. An animal in the forest estimates what a patch of food offers against the energy it costs to get there, and commits or moves on.

The cues it reads before committing, the smell on the air, the tracks on the trail, are the difference between eating and wandering.

On the web, the patches are pages, the cost is the click and the seconds that follow it, and the cues are words.

Information scent is the strength of the match between what a link says and what the visitor is hunting. Strong scent predicts the payoff. No scent predicts nothing, and a forager facing no information does the rational thing: keeps their energy and moves to the next trail, which is usually somebody else’s page.

Worth noticing what this explains. Visitors leaving after your homepage without clicking anything were not bored and were not confused. They sniffed, found nothing, and foraged elsewhere.

What scent is made of

Scent is made of trigger words, and the trigger words are the visitor’s vocabulary, not the org chart’s.

The heading above the link contributes, the sentence around it contributes, and the link text itself carries the most. A visitor scanning your page is assembling an answer to one question at every underlined phrase: does this smell like the thing I came for?

That question is why scanning behavior matters so much: the left-edge skim in an eye-tracking study is literally a scent hunt, the eye sampling line beginnings for relevant words before it commits to a full read.

The “learn more” problem

Now run the least scented phrase on the web through the model. “Learn more.” It names no destination, names no payoff, and could truthfully lead anywhere, which for a forager is the same as leading nowhere. The rationally correct response to zero information is zero commitment.

What makes this expensive is that “learn more” usually sits under the exact thing a motivated visitor wanted. The case study is behind it. The pricing breakdown is behind it.

The visitor who would have clicked a sentence that said “see the full breakdown of a five-page site build” never finds out, because the link smelled like homework.

The rewrite

Take the phrase, name the destination, use the visitor’s word for it, and cut the rest. “Learn more” becomes “See what a five-page site includes”. “Read more” becomes “Read the 1997 eye-tracking study”. “Our solutions” becomes “Websites for tradies”. Three seconds of honesty, and the trail starts announcing what is at the end of it.

The same math runs on your navigation. A menu label is a link with a bigger decision attached, so vague labels cost more there, not less.

“Solutions”, “Services”, “Discover”: these are the corporate dialect of no scent, and every visitor decoding them pays from the same four-slot working memory budget they need for everything else on the page.

Brands do this on purpose sometimes, chasing personality in menu items. Personality in a nav label is perfume poured over an empty trail. The visitor hunting “pricing” does not feel delighted by “The Journey”. They feel lost, and lost foragers leave.

The scent your visitor hits first

For most visitors the first foraging decision happens before they reach your site at all. A search results page is a menu of trails, and your page title and meta description are its link text.

Same rule, higher stakes: a title that names the thing the searcher is hunting gets the click; a clever title that names nothing gets scrolled past in favor of a blunter competitor.

This is the whole reason I write meta descriptions like a promise instead of a slogan. The snippet is scent. The ranking got them listed; the scent gets them visited.

“But descriptive labels are clunky”

Shorter is not a goal. Clarity is the goal, and it turns out characters spent describing the destination are the cheapest conversion spend on the page. The clunky version that says what it is will beat the elegant version that doesn’t, every single time the two compete for a click from someone in a hurry.

And when a label does have to stay short, the sentence around it can carry the scent. Scent is cumulative across the link, the heading, the paragraph. The web forgives a two-word link in a context that explains it. It does not forgive a two-word link floating alone, meaning everything and therefore nothing.

Scent decides the click; what the visitor can physically see when they arrive decides the rest. Banner blindness covers the invisibility problem, and the complete scannable-writing checklist covers how to build pages that hold up under a skim.

If your analytics show visitors landing and leaving without touching anything, that is a scent report, and it is more actionable than it feels. Send me the page and I’ll tell you which trails smell like nothing.

Sources & further reading

  1. Pirolli, P., & Card, S. K. (1999). Information Foraging. Psychological Review, 106(4).
  2. Information Foraging — Wikipedia (the accessible overview of the theory)
  3. How Users Read on the Web — Nielsen, 1997
  4. Plain Language Is for Everyone, Even Experts — Nielsen Norman Group