{"id":48396,"date":"2023-04-14T19:18:47","date_gmt":"2023-04-14T19:18:47","guid":{"rendered":"https:\/\/www.arimetrics.com\/glosario-digital\/card-sorting"},"modified":"2026-10-02T06:56:32","modified_gmt":"2026-10-02T06:56:32","slug":"card-sorting","status":"publish","type":"encyclopedia","link":"https:\/\/www.arimetrics.com\/en\/digital-glossary\/card-sorting","title":{"rendered":"Card Sorting"},"content":{"rendered":"<p><img decoding=\"async\" class=\"boxpad alignright wp-image-48501 size-full\" src=\"https:\/\/www.arimetrics.com\/wp-content\/uploads\/2023\/04\/card-sorting.jpg\" alt=\"Card sorting\" width=\"300\" height=\"300\" srcset=\"https:\/\/www.arimetrics.com\/wp-content\/uploads\/2023\/04\/card-sorting.jpg 300w, https:\/\/www.arimetrics.com\/wp-content\/uploads\/2023\/04\/card-sorting-150x150.jpg 150w\" sizes=\"(max-width: 300px) 100vw, 300px\" \/> <strong>Definition:<\/strong><\/p>\n<p><strong>Card sorting<\/strong> is a research method in which participants group items represented by cards according to criteria that make sense to them. They may also name the groups or work with categories defined in advance.<\/p>\n<p>It is mainly used to explore or evaluate how information should be organized in a website, application, intranet, catalogue, or another system. Its results provide <strong>evidence about groupings<\/strong> and labels, but they do not automatically produce a final architecture or prove that a structure is easy to navigate.<\/p>\n\n<h2>Types of card sorting<\/h2>\n<p>The type is selected according to the research question. The three common approaches differ in the degree of freedom given to participants:<\/p>\n<ul>\n<li><strong>Open:<\/strong> people create the groups and name them. This can reveal classifications and vocabulary that the team had not anticipated.<\/li>\n<li><strong>Closed:<\/strong> categories are defined beforehand and participants place cards within them. This helps evaluate a proposed structure.<\/li>\n<li><strong>Hybrid:<\/strong> initial categories are provided, but people may create new ones when none is suitable.<\/li>\n<li><strong>Individual:<\/strong> each participant completes the task separately, in person or remotely, making independent results easier to compare.<\/li>\n<li><strong>Collaborative:<\/strong> several people agree on groupings and explain their decisions, adding qualitative discussion to the final arrangement.<\/li>\n<\/ul>\n<p>These variations are not interchangeable. An open study is better suited to exploratory questions, whereas a closed study checks whether existing categories are understandable. The <strong>choice of method<\/strong> should come before card preparation.<\/p>\n<h2>Preparing the study<\/h2>\n<p>Preparation starts by defining the system, audience, and decision the study is intended to support. Not every page or product needs to be represented. An excessive set causes fatigue, while a set that is too small may hide relevant relationships. Cards should be <strong>clear and comparable<\/strong>.<\/p>\n<p>Each card usually contains a <strong>short label<\/strong> representing a content unit or function. Ambiguous names, duplicates, obvious differences in detail, or visual clues that push participants towards a particular group should be avoided. When an item needs an explanation, it should be neutral and consistently applied.<\/p>\n<p>Participant selection depends on the system&#8217;s audience. A sample does not become representative merely because it includes many people: experience, context, language, and familiarity with the subject also matter. Experts and beginners may use <strong>different mental models<\/strong>.<\/p>\n<p>Before starting, the team prepares the <strong>initial instructions<\/strong>, a way to record questions, and a rule for handling difficult cards. A pilot test can detect unclear wording, technical problems, and items that everyone interprets in an unintended way.<\/p>\n<h2>Running the activity<\/h2>\n<p>The session may use physical cards or a digital tool. The moderator explains the activity without suggesting the expected structure. It is useful to observe <strong>questions and reasoning<\/strong>, not only the final location of each card.<\/p>\n<p>The procedure commonly includes these actions:<\/p>\n<ul>\n<li><strong>Introduce the task:<\/strong> explain the practical purpose, rules, data handling, and how participants can request clarification.<\/li>\n<li><strong>Sort the cards:<\/strong> let each person review, move, and regroup items before finishing.<\/li>\n<li><strong>Name the groups:<\/strong> request labels in open or hybrid studies to capture the audience&#8217;s vocabulary.<\/li>\n<li><strong>Record decisions:<\/strong> retain groupings, categories, completion times, discarded cards, and relevant comments.<\/li>\n<li><strong>Ask afterwards:<\/strong> clarify criteria, doubts, and alternatives without turning the conversation into directed validation.<\/li>\n<\/ul>\n<p>Moderated sessions provide detailed explanations; unmoderated sessions make it easier to include more participants and locations. Remote participation does not guarantee diversity, and in-person work does not guarantee depth. <strong>Recording quality<\/strong> depends on the design and participation conditions.<\/p>\n<h2>Analysis and interpretation<\/h2>\n<p>Analysis begins by reviewing which items were grouped together, how often, and under which names. In studies with several participants, similarity matrices, dendrograms, or other visualizations can identify <strong>shared patterns<\/strong>. These techniques summarize the data but do not replace interpretation.<\/p>\n<p><strong>Observed disagreements<\/strong> also require attention. A card spread across many groups may indicate ambiguity, context dependence, or audiences with different needs. Combining every response into one structure can erase important differences among segments.<\/p>\n<p>Proposed names provide vocabulary, but they should not be transferred automatically to navigation. Their clarity, consistency, and fit with the rest of the system must be reviewed. Card sorting can inform a taxonomy, menu, or content inventory, although the <strong>design decision<\/strong> must combine its findings with objectives, constraints, and other evidence.<\/p>\n<p>The <strong>final report<\/strong> should document participants, type of study, cards, instructions, exclusions, and the analysis method. This makes the origin of each conclusion understandable and prevents a frequent grouping from being presented as a universal rule.<\/p>\n<h2>Differences and limitations<\/h2>\n<p>Card sorting investigates how items are related and categorized. Tree testing examines whether people can find destinations within an existing hierarchy. The methods can therefore complement each other: one helps propose an organization, and the other tests its <strong>findability<\/strong>.<\/p>\n<p>A <a href=\"https:\/\/www.arimetrics.com\/en\/digital-glossary\/usability\"><strong>usability<\/strong><\/a> test observes how people complete tasks with an interface or prototype. Card sorting does not evaluate interaction, visual design, loading time, or the comprehension of complete pages. A <a href=\"https:\/\/www.arimetrics.com\/en\/digital-glossary\/heuristic-analysis\">heuristic analysis<\/a> applies expert criteria and does not replace the perspective of real participants.<\/p>\n<p>The method does not guarantee improvements in SEO, conversion, retention, or costs. A label suggested by participants may describe the content poorly, and an intuitive grouping may conflict with legal, commercial, or technical requirements. <strong>Further triangulation<\/strong> may include analytics, internal search data, interviews, tree testing, and prototype testing.<\/p>\n<p>Its <strong>main limitations<\/strong> include card selection, sample bias, the artificial nature of isolated labels, and the influence of instructions. Findings should be treated as one source of evidence within <a href=\"https:\/\/www.arimetrics.com\/en\/digital-glossary\/design-thinking\">human-centred design<\/a>, not as an automatic decision.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Definition: Card sorting is a research method in which participants group items represented by cards according to criteria that make sense to them. They may also name the groups or work with categories defined in advance. It is mainly used to explore or evaluate how information should be organized in a website, application, intranet, catalogue, [&hellip;]<\/p>\n","protected":false},"author":6,"featured_media":0,"template":"","encyclopedia-tag":[1286,1426],"class_list":["post-48396","encyclopedia","type-encyclopedia","status-publish","hentry","encyclopedia-tag-ux-research","encyclopedia-tag-web-prototyping"],"_links":{"self":[{"href":"https:\/\/www.arimetrics.com\/en\/wp-json\/wp\/v2\/encyclopedia\/48396","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.arimetrics.com\/en\/wp-json\/wp\/v2\/encyclopedia"}],"about":[{"href":"https:\/\/www.arimetrics.com\/en\/wp-json\/wp\/v2\/types\/encyclopedia"}],"author":[{"embeddable":true,"href":"https:\/\/www.arimetrics.com\/en\/wp-json\/wp\/v2\/users\/6"}],"wp:attachment":[{"href":"https:\/\/www.arimetrics.com\/en\/wp-json\/wp\/v2\/media?parent=48396"}],"wp:term":[{"taxonomy":"encyclopedia-tag","embeddable":true,"href":"https:\/\/www.arimetrics.com\/en\/wp-json\/wp\/v2\/encyclopedia-tag?post=48396"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}