When designing a human-computer interaction, one of the main goals is to match users' mental models of the interface to the conceptual model intended by the designer. This is especially important when deciding how to group a large number of items (e.g., links or menu selections) into meaningful categories (e.g., menus) in order to facilitate user navigation and decision-making.
One powerful method that can be used to aid in such decisions is the card-sort. Card-sorting is a research method where users sort cards representing relevant concepts into categories. This method has a long tradition in the Human Factors field as well as other fields, such as marketing research. Traditional card-sorting presents a number of challenges, including:
- Card-sorting exercises can require a large amount of physical space and time depending on the number of items to be sorted. This can hinder collection of large enough sample sizes to achieve statistical conclusion validity.
- Tracking items and categories during a sort can be frustrating for the participant and prone to errors, especially with a large number of items.
- Analysis and interpretation of results is very tedious and time-consuming.
Online card-sorting tools have addressed these problems, with varying degrees of success. Many of the available card-sorting tools have their own usability problems. The online tool that we have found to be the best is Websort. We have used it extensively in refining terminology and menu structures for software used in the financial sector (online banking, teller systems) and CRM.
