Goms

GOMS is a family of human–computer interaction models used to analyze and predict how a user will perform a task with a given interface. It aims to describe the sequence of actions a skilled user would undertake to achieve a goal and to estimate the time required for task completion under specified conditions. GOMS was introduced by Card, Moran, and Newell in 1983 and has since influenced interface design and usability evaluation.

The model is built from four components: Goals, Operators, Methods, and Selection rules. Goals are the high-level objectives the user wants to accomplish. Operators are the primitive actions the user can perform, including mental activities (planning and decision making), perceptual checks, and motor actions such as keystrokes and device movements. Methods are predefined procedures that achieve a goal by combining operators in a particular sequence. Selection rules determine which method to apply in a given situation, based on context and goals.

Practically, a task is decomposed into goals and methods, and the duration of a method is the sum of its operators’ times. By comparing different methods and their predicted times, designers can identify efficient interaction paths and predict how design changes may impact performance.

Variants and extensions include the Keystroke-Level Model (KLM), a practical, time-based refinement that predicts execution time by summing fixed times for common operator types (such as keystrokes and pointing). Other forms, such as NGOM and CPM-GOMS, adapt the approach for different cognitive or perceptual assumptions.

Limitations include reliance on fixed operator times, simplified cognitive assumptions, and challenges in modeling complex, long-term planning or multi-user tasks. GOMS remains a foundational framework for structured usability analysis.