How Card Sorting Improves User Experience
Discover how card sorting, a simple user experience design exercise that engages real users, can inform a website’s information architecture.
Mightybytes conducted a card sorting exercise with patrons of Niles Public Library to inform the appropriate information architecture for their website. Here are some lessons we learned from the workshop.
What is Card Sorting?
Card sorting is a workshop exercise where you invite project stakeholders to group content in ways they would naturally expect. Given that libraries provide hundreds of resources, both physically and virtually, card sorting provided a meaningful way for us to craft an intuitive and gratifying user experience for library patrons.
For our exercise we had 16 participants ranging from age 11 to 70 years old. This was an intentionally small but diverse sample of individuals. 93% of our sample had previously used the library’s website. All of them are real users of the current library website and patrons of the library’s facility.
Over 100 cards depicting existing and planned site content, such as free events and information on youth services, were presented to patrons via an interactive web interface. The patrons were then responsible for sorting them into logical groups and then giving the groups names.
After the exercise was completed, Mightybytes merged any group names that were highly similar (for example, we merged “for kids” with the label “kids”), and began looking for patterns. Using Optimal Sort’s dendograms (tree diagrams illustrating how cards were clustered within a hierarchy), similarity matrices (a type of spreadsheet that clusters the strongest card groups together), and the raw data, we were able to analyze how these patrons grouped library content in ways that made sense to them.
While this data was incredibly helpful in navigation design, we had to keep in mind that the goal in a card sorting exercise is often to find the biggest points of convergence and the biggest points of disagreement, rather than perfect agreement for all cards between all patrons.
What follows are some observations we made after analyzing the data.
Card Sorting: Disagreements and Challenges
Consensus didn’t always come naturally during these exercises. Here were some challenges we bumped up against.
Of 16 participants, eBooks was put into 15 different categories (94%). One may view this as highlighting the challenge of categorizing eBooks within a library website. While paper books have been around since the Fifth Dynasty, eBooks and the ubiquity of affordable devices to view them, on have only been available as of the last dozen years or so. Additionally many people may not know that eBooks are available as a free library service, and may just associate these with Amazon purchases. Patrons asked themselves, is it media? part of book collections? related to electronics? an online resource? Similarly, the card for “audio books” was categorized 14 different ways.
Beyond traditional collections
The role of a library within a community is complex. In addition to lending out books, music and films, Niles Public Library also provides a wide array of essential services to their patrons. These include distributing free museum passes to families, distributing citizenship information, and even providing notary public service. There was no agreement on how to categorize these services.
It may be that these public service offerings fall under the radar patrons, who may not know that these services are offered, or that the results of our survey may have been colored by our small sample size. The challenge here is if very few people are looking for something, how can you ensure it’s findable in a logical place?
Is a library a brick and mortar location, or hub of services?
Patrons struggled to understand how to categorize both the home delivery of books, and bookmobiles. This may indicate the possible misconception about the library being solely a physical, brick-and-mortar location versus the hub of an array of community services. The concept of a library being “mobile”, or living outside the actual facility confused patrons.
Similarly, patrons created a number of different categories related to technology. These categories featured titles like “online” “technological”, “computers and internet”, and “computers and internet information”, and the cards within those groups seemed to indicate a struggle to differentiate between virtual services available from remote locations (such as a patron’s home or school), and the virtual services available to patrons at the brick-and-mortar library location. If anything, this highlighted the importance of educating patrons on the range of library services that are available to them in their own homes.
The problem with too much information
When stuck patrons would often turn to the concept of “Information” as a category. This lead to “General Information”, “Library Information”, “Information” and “Info on the Library” being created as categories. None of these vague categories reached more than 59% agreement across our sample of patrons.
The main takeaway here is that the library should pursue valuable, understandable page names. Information as a category is too vague and subjective to be used on a website where you want to make things easy to find.
Card Sorting: Similarities and Indicators
Age-specific content was grouped similarly by the majority of participants. Given the specificity of the content (for example “teen advisory board”), it was not hard for patrons to group these resources into natural categories. This proved our hypothesis that age-specific navigational destinations meet library patron’s expectations and are easy to understand.
The majority of patrons also paired and grouped similar items under “my account”; such as when their books are due, how to put an item on hold, and viewing item renewal policies. This clearly indicates that people want an easily accessible, single destination for all of the information regarding their own personal account with the library.
Credible User Experience Design
Gaining these insights from Niles Public LIbrary’s patrons lend credibility to our website design, and will help ensure that the site has an intuitive interface that both helps patrons find what they are looking for quickly and easily, and highlights some of the services that patrons may not be as aware of. We look forward to revealing the results of our work when the website launches, and we hope that other organizations—both libraries and other institutions with a complex array of offerings—can benefit from knowing more about this process.