Card-sorting
When is this method appropriate?
Card-sorting helps to detect the users’ perception in relation to certain terms in context to a website or an intranet. This test helps to understand how users group information. The goal is to develop information structures that have a high user acceptance and to maximise the probability to find information or products.
The results of the sorting are the basis for the mental model and provide a valuable data base to develop a user-friendly information architecture.
Possible questions
- How can new and existing contents be structured?
- How many main categories are necessary?
- How should the main and sub categories be named?
- What contents or sub categories should be arranged to the main categories?
And many more…
In what project phase applicable?
- Development/conception
Card-sorting tests should be implemented in the beginning of the conceptional phase of a web/intranet project or at the end of the analysis phase. At this point it is generally clear which contents and services should be provided or the requirements and expectations that the desired target-groups presume. One has to consider to structure the contents conform to expectations, user-friendly and to find the correct, expressive terms for the main categories.
What is the procedure of this method?
In a card-sorting test the users write down terms (= terms for sub categories) on cards and arrange them in groups. Generally one differentiate 2 methods: In the closed card-sorting the terms are grouped to given categories. In the open card-sorting the user creates own categories which (should) represent the main categories.
The grouping takes place while considering the contentual and functional thoughts. Terms which are hard to group are separately considered..

Test subject collating terms and contents
In a multi-level card-sorting the groups that were created during the first session are continuing summarising. Before the sorting the comprehensibility of the terms can be additionally tested (wording analysis).

By a user structured sitemap of a website (clip)
Strengths of this method
- The method enables structuring large information from the user’s point of view.
- Menu structures, categories and terms can be tested with regard to their conformity expectations.
- In addition, it is possible to test alternative terms.
- This method does not require an operative prototype.
- The method can be implemented offline and online.
What about the sample size?
One needs at least 30 test subjects for the sample so that clear information structures or terms can be estimated. Depending of the complexity (for instance structuring of intranets) multiple hundred test subjects are necessary.
How much time takes the implementation?
If the terms that should be sorted are determined, a card-sorting test takes usually three weeks.
What input do you have to provide?
We just need the terms of the lowest level that means of the final contents for example in terms of a sitemap/content overview. On this base we create a test concept. In addition, one has to determine if one uses an open or closed card-sorting that means if the terms for the upper categories are supposed to be given or if the test subjects have to create them on their own. If you want the closed test, we additionally need the wanted main terms of the main categories.
We would like to support you with the identification of appropriate terms for the existing contents, functions and services.
What do you receive from us?
Interpretive chart report with the following content:
- Management summary
- Detailed illustration of the results based on survey and observation data (grouping of terms in textual and graphical form).
- Data interpretation: identification of the reasons and causes for the ascertained data.
- Concrete guidance and advices for optimisation of identified structuring and wording problems:
- Recommendations related to the information structure incl. the number of levels and categories as well as
- Recommendations how to name categories and terms.
On demand we present the results and findings of the test within a moderated workshop.
Do you have any further questions? Please do not hesitate to contact us! We would like to consult you.
- Katja Brand-Sassen
- +49 (0)69-13396541

]
]