Inspections and Walkthroughs

Inspections and walkthroughs started as software engineering methods for improving the quality of requirements, documentation, and code (Fagan, 1976; Gilb & Graham, 1993). The inspection and walkthrough approaches used in software engineering were partially borrowed by usability practitioners in the late 1980s and early 1990s for use in evaluating software and hardware for usability problems (Cockton, Woolrych, Hornbæk, & Frøkjær, 2012). Walkthroughs and inspections in the user experience (UX) domain were often viewed as informal (or semiformal), relatively fast, practical, and flexible (Hartson & Pyla, 2012)—attributes that make them appropriate for today’s agile design environments.

In the early 1990s, the user interface (UI) inspection gained acceptance as a way for both novices and experts to evaluate products and services. The classic book, Usability Inspection Methods (Nielsen & Mack, 1994), summarized the state of UI inspection methods in the early 1990s and provided both procedures and supporting research. Hundreds, perhaps thousands of articles have appeared on different UX inspection methods and approaches. These numerous articles discuss the strengths and weaknesses of particular inspection methods and often compare them to usability testing (Cockton et al., 2012).

In this book, six inspection-walkthrough methods are covered beginning with heuristic evaluation and ending with the formal usability inspection method.

Chapter 1 focuses on heuristic evaluation, perhaps the best-known inspection method. Heuristic evaluation generally requires a small group of evaluators to review a product against a set of general principles (the heuristics). A set of heuristics might include items like these, among others (Nielsen, 1993):

Heuristic evaluation was originally conceived as a simple tool for usability practitioners and other groups, such as software developers or QA engineers, with relatively little training in usability. Chapter 1 describes the heuristic evaluation method and as well as ideas from researchers, practitioners, and this author, on how to improve the effectiveness of heuristic evaluation.

Chapter 2 focuses on the individual expert review, where a lone UX practitioner is asked to conduct a usability evaluation of a prototype or working product. This is more of a personal chapter, based on this author’s experience as a consultant and individual contributor in commercial environments over the past thirty years. Many UX practitioners are asked for their expert evaluation of a product, often on short notice with no budget for testing or inspections with multiple evaluators. Although an individual expert review is not the optimal method or approach for finding the majority of usability problems, it is a common approach for many in consulting and corporate environments. This chapter focuses on how practitioners can increase their effectiveness when working alone on a usability evaluation.

An individual expert review can incorporate components of think-aloud testing, heuristic evaluation, checklist reviews, perspective-based inspections, and other evaluation methods.

Chapter 3 describes the perspective-based UI inspection, which is based on the principle that taking on different perspectives will increase the likelihood of finding UX problems. For example, you might ask a set of evaluators to each adopt different perspectives for their individual reviews; one person might be a “consistency czar” who looks at visual, interaction, and cross-product consistency. Another person might take on the role of an “efficiency expert” who focuses on shortcuts, automation, and the number of steps required to accomplish a task. Yet another person might take on the perspective of a “readability expert” who looks for text that is too small, low contrast between text and background, and excess verbiage. The persona-based inspection is a type of perspective-based inspection where selected colleagues assume the perspectives of the personas in reviewing the product. Persona-based reviews support a better UX by focusing inspections on the needs, background, tasks, and pain points of the different personas.

Chapter 4 covers the cognitive walkthrough, an inspection method that focuses on ease of learning (often referred to as “learnability”). The original targets for this method were walk-up-and-use systems such as automated banking machines, postal metering systems in malls, or information kiosks in museums. In the twenty-five years that the method has been in use, there have been several attempts to simplify it for use outside academia (Grigoreanu & Mohanna, 2013; Spencer, 2000). This chapter describes the original method and variations meant to reduce the complexity for evaluators.

The first four methods in this book do not generally require actual or potential users of a product. Chapter 5 describes the pluralistic walkthrough (Bias, 1994) where users are invited to provide feedback on a design while members of a product team listen, observe the users, and ask questions when appropriate. This method is a form of participatory evaluation that involves entire product teams who are schooled in how to interact with users (e.g., they are asked to let the users speak first).

The formal usability inspection (FAA, n.d.; Kahn & Prail, 1994), the last of the six methods, is based on the formal software code inspection, which involves strict rules about preparation, evaluation procedures, data collection, and fixes. It is described in Chapter 6 that the formal usability inspection was best described by Kahn and Prail (1994) in the groundbreaking book, Usability Inspection Methods (Nielsen & Mack, 1994). The chapter includes a description of the formal usability inspection method and a discussion of some enhancements from the software engineering and quality literature. Formal inspections might be most appropriate for complex applications or products where there are serious consequences for errors (e.g., air traffic control systems, building design and construction, hospital support systems).