Did you know information has a scent, and users track it like predators? Information scent in UX

When I hear the word “theory”, I sometimes tense up a little. I think: uh oh, this is going to be hard to wrap my head around. The first time I heard about Information Foraging Theory, during my HCID studies at City University London, and had to use it to evaluate a website, I thought… how am I supposed to understand this?

Today I smile to myself thinking back on it. Because it turns out to be one of the simplest and most on-point theories for understanding how people move around the internet and choose between different information sources to answer a question or satisfy an information need. In a very vivid way, it compares us humans to predators, and information to prey.

If you understand this theory, designing effective information architecture, a menu, say, becomes far simpler. Today I’ll focus on just one part of it: “information scent”, the trail that leads a user onward. I’ll also show how this approach translates into an age where we increasingly look for answers differently, with AI, say.

This article is for you if you want to improve your site’s design so users find the information they need quickly and without frustration. Using two examples, I’ll show you just how big an effect a well-planned (or poorly planned) “information scent” can have on conversion.

Did you know you’re a predator?

In this analogy, an information source can be compared to a product you’re looking for in an online store. You can’t yet see all the details, so you judge based on what’s available “from a distance”: the name, the photo, other people’s ratings, the price, a short description.

That’s the “cue”, the thing that signals whether it’s worth clicking to go deeper.

You’re acting a bit like a wild cat tracking prey. It doesn’t see it right away, but it picks up on signals: movement, scent, tracks. On that basis it decides whether it’s worth approaching. You do exactly the same thing, except instead of scent you have headlines, links and thumbnails.

You also factor in past experience: whether you’ve landed on a similar site before, whether it was helpful or disappointing. And even if something looks promising, once you get there it may turn out not to be what you were looking for.

Four key elements follow from this analogy:

• the actual source, the site or specific page with the information
• the distant cue, a Google link, a headline, a description or a photo
• the actual value, what you actually get once you arrive
• the perceived value, what you think you’ll find, which is what decides whether you click

Information scent is the first impression that tells a user: is it worth clicking here or not? Two things shape that impression:

1. what the user sees, the link title, description and context on the search results page, for instance. That gives a hint about whether “there’s something there”
2. what the user already knows, whether they know the site, brand or author, and whether they trust them

In other words, the user judges within a few seconds whether it’s worth going further. If something doesn’t feel right, they leave, often before you’ve even had a chance to show them your offer.

In the next section I’ll show you two real examples where “a false information scent, or its absence” created friction on the purchase path and lowered conversion.

Straight from the field: a predator loses the trail… when the scent disappears or doesn’t match the target

Example 1: a predator loses the trail because the scent promises something that isn’t there

Our client, Medeocare, came to us with a problem: customers kept calling to ask why they couldn’t choose parcel-locker delivery. From Medeocare’s perspective, everything looked correct – the information about the restriction was on the delivery page, so the issue seemed to be addressed. Yet users were frustrated and kept coming back with the same question.

What did we do, and what did we find?

In an expert review I traced the user’s path through several scenarios. It quickly became clear the user gets a false trail from the very start:

• the homepage – the message “Free parcel-locker delivery from $45” suggests the option applies to every product and gives no way to check the details

• the product page – no information about the restriction, and elements promoting the parcel locker actually reinforce the mistaken belief

• the cart – messaging still suggests parcel-locker delivery is available

• checkout – the parcel-locker option disappears with no explanation at all

The result: the predator got a false trail

The user doesn’t understand what happened. At every earlier stage, everything pointed to that option being available. Only at the end does it turn out it isn’t. And that’s exactly the moment frustration sets in, the phone call to support, the feeling that “something’s off here”.

The information about the restriction did technically exist, but it was placed on the delivery page in a way that was hard to notice and disconnected from context. It wasn’t tied directly to the parcel-locker option, so it was easy to miss. On top of that, the user never lands on that page at the moment of choosing, so the information simply doesn’t do its job.

The solution: show the information where the user makes the decision, instead of on a separate page. Removing the contradictory messages and clearly explaining why an option isn’t available eliminates the false information scent.

Example 2: a predator doesn’t move, because it can’t smell anything

On the MyBed project, during usability testing, we noticed a different kind of problem. Users weren’t being guided through the path in a way that let them understand the offer from the start. Already at the menu and category level, they didn’t understand how the products differed or what to expect. The category names didn’t build any concrete picture.

What did we do, and what did we find?

Testing showed that:

• the bed names in the menu categories don’t suggest what those beds actually are

• users don’t understand the differences between categories (bed types, say)

• they choose paths by trial and error rather than deliberately, and get lost on the listing because of missing key filters and an unnoticed banner about the configuration option

• they don’t know products can be configured, and the configuration option is only discovered on the product page (if they even get that far!), exactly the moment the user has already “arrived”

The result: the predator doesn’t follow the trail

The user doesn’t get enough signals to understand MyBed’s offer and make a decision. They can’t see the difference between bed types, and don’t know that configuration is a key part of the offer. As a result they don’t choose a path deliberately, they move by trial and error, or give up right at the listing stage. Even if they reach the product page, it’s only there that they discover the configuration option, the main value of MyBed’s offer, which should be visible and understood far earlier.

The solution: build the information scent from the very start of the path. Category names should clearly communicate the differences, “ready-made beds” vs “configurable”, say, and a key value like the configuration option has to be visible already at the menu and listing stage. The path should gradually build understanding of the offer, so the user knows what to expect right away. That way they “feel the value” before they click, not only at the very end.

We no longer have to “sniff around” the way we used to. AI has cut the path to information down to almost nothing, and we’ve gone from predators to something closer to house cats.

AI-powered search fundamentally changes how “information scent” works. Instead of following clues through page after page, the user gets a ready-made answer right away. The path from “I don’t know” to “now I know” becomes almost instant. That shortening of the whole process also changes the experience of searching for information itself.

Consumers use AI-powered search for discovering products, comparing features, getting recommendations and more. It’s estimated that around 40-55% of users use AI at some point on the purchase path, with varying intensity at each stage.

A sample AI-assisted e-commerce purchase path:

• Discovery “Which running shoe brands should I consider?”
• Consideration “Is there anything specific I should look for if I tend to roll my ankles?”
• Purchase decision “Which 2 models should I try first?”

source: McKinsey & Company

Optimizing for UX and SEO used to be the foundation, and now GEO, optimization for AI, is entering the game too – and it’s becoming essential if you want to be present wherever users are also making decisions.

In e-commerce, what AI Overview shows in Google largely comes directly from store and manufacturer sites. According to analyses, as much as around80% of sources in this categoryare brand and retailer sites, with only around10%being user-generated content.

source: McKinsey & Company

That means your site is no longer just a place to sell, it’s also a source of answers that reach the user before they even visit it. If your content is unclear, inconsistent, or fails to communicate a product’s value, AI won’t “understand” it and won’t pass it on.

Which is why optimizing a store today isn’t only about UX and SEO, it’s also about how you structure information, name categories, and show the differences between products. A good “information scent” has to be readable not just for the user, but for AI too.

This isn’t a revolution, just a change in approach. To start, it’s worth:

• checking how your brand and products “show up” in AI answers
• organizing your contentso it clearly communicates value and the differences between products
• paying attention to structure (headings, language, specifics) so AI can easily “understand” the content
• looking beyond just your own sitereviews, articles and other sources matter too

Today only a small share of companies actually measure their AI presence, and the gap between SEO and GEO performance can run into the tens of percent.

The idea of “information scent” isn’t going away, only where and when the user picks it up is changing.

The psychology of clicking isn’t disappearing, it’s just changing. Success today requires optimizing both for visibility in AI systems and for the key moments when a user is ready to click. In the AI era, influence means building authority that extends beyond a single interaction.

At Digispot we combine full-spectrum expertise to:

• improve your site’s “information scent” and raise the clarity and comprehensibility of your offer
• prepare content that works for users and AI systems alike

Because today a user’s decision is often made before they even reach your site.

Put as simply as possible: if you want to be visible wherever users are making decisions, you have to start designing content for AI as well as for people.

At Digispot we help brands design experiences that are understandable to users and AI systems alike, from information architecture and UX, to SEO and GEO. If you want to find out where your users are losing the trail and how to improve your offer’s visibility, get in touch.