We live in a search-driven digital world, which is great as long as you know the keywords you need. But what happens when you don’t?

When users want to find an item or a piece of information and they don’t know what it’s called, they face a difficult problem. Sometimes in our research, we see users engage in keyword foraging.

In keyword foraging, a user conducts a preliminary search (usually in a web search engine like Google) to determine the right keywords for her information need.

Think of keyword foraging as the search before the search: one or more preliminary queries to help the user formulate her actual query.

These tend to occur when the user:

  • Knows what an item looks like, but doesn’t know its name (e.g., doesn’t know that pointed-toe formal shoes are called “wingtip” shoes)
  • Needs a solution to a problem, but doesn’t know what tool, technique, or service could provide the solution (e.g., needs to replace a piece of trim around the vent at the front of a car, which is called “grille trim”)
  • Used to know what an item was called, but doesn’t remember (e.g., used to know the term “hangar” but forgot it, and must google “plane shed” instead)
  • Knows what an item or concept is called in one language or dialect, but not another (e.g., a British person who doesn’t know that “spring onions” are called “scallions” in American English)
Imagine you want to buy these shoes, but you don’t know what the style is called. You might have to Google “men’s formal shoes” and then “men’s formal shoes pointed toe” before determining that they’re called “wingtip” shoes.

Example: Duster Cardigan

One study participant knew she wanted to buy a specific type of long sweater (called a “duster cardigan”) that was trendy at the time.

“One of the things that I think is hardest about fashion stuff is that you have to know what the name of something is before you can look for it. Trying to find out — I want something that’s like a cardigan, but it’s long. You have to know all the words, there’s all these terms…”

She explained that she typically starts by searching on Google Shopping for a basic idea of what she wants; for example, “long sweater cardigan.” She looks through the image results until she finds something similar to what she wants, and then checks the language used to describe it.

One participant used Google Shopping to determine the keywords she needed to search for. “Here we go. So I’m actually looking for something like this,” she told us. (This participant was in an eyetracking study — the red dots represent her gaze fixations on the page.)

It’s important to note that this particular participant happened to be a librarian, so she likely had more advanced research skills than the average user. When people have lower research skills, they often struggle with this type of keyword discovery.

A different participant with lower research skills was planning an upcoming trip to Paris. She wanted to find out how to get cash for the trip in advance, but didn’t know what currency was used in France. She tried searching on Google for “shopping in Paris.” Unfortunately, this search only brought up tips for which neighborhoods and shops to visit in the city. Her keyword foraging failed — she never discovered that euros are used in France. 

Search tools often fail when users can’t supply the right query terms.

Help Your Users Find What They Need, Even Without the Correct Term

Your users may know what they want, but not what that item is called. This situation is common particularly when users are unfamiliar with the search domain. If your subject matter is complex or specialized (for example, skin care, car parts, or home improvement), this issue may be particularly important for you. 

Ecommerce sites should also pay attention to this problem (as always, people can’t buy what they can’t find). Faceted systems can sometimes be helpful for formulating queries in unfamiliar domains. For example, a user may be looking for a range with an electric oven and a gas top but have no idea that these are called “dual-fuel ranges.” The filter options displayed on the side can teach him the vocabulary required to perform the search.

Filters (like these on Home Depot’s site) can help users discover and learn the terminology they need.

Some large ecommerce sites may find that image search also helps to address this problem. The user doesn’t need to know the name of the thing she wants — just to be able to recognize it and photograph it. However, most site image-search tools aren’t yet able to deliver the quality results that search engines like Google can. 

For now, focus on anticipating what language people might use to search for your products or content if they didn’t know the correct term. Check your site search logs, and you might find evidence of these searches occurring on your site. Your customers may search the exact term (“duster”) or a rough description (“long sweater cardigan”). The keyword foraging phenomenon is a reminder to use plain language in your content and also to accept imperfect or inaccurate query formulations. 

This problem is also one more reason to avoid inventing new terminology whenever possible, particularly branded terminology. If you invent a new term, your users will definitely not know it. For example, if an oven manufacturer decides to make up the phrase “SmartClean Technology” instead of the well-known term “self-cleaning,” shoppers won’t know to search for it. 

Of course, there are situations that do require a new term. For example, we had to come up with “keyword foraging” to name this phenomenon (and took inspiration from information foraging theory). However, branded terminology is often unnecessary and avoidable. If you do use branded terms, make sure you include the plain language alternative alongside it on the site, to help with SEO as well as educating customers.