WhyUsable is basic. Pleasing and engaging are expected too by users and should be seen as the new baseline for serious product design. Evaluation gives real information for design and marketing decisions rather than just opinions.
What and WhenEvalutation can be made in all development phases. Low-fi or low-tech prototypes can be evaluated early to ensure that further design and development resources are spent on things that matter for the user. It can also be done on a full working system to verify desired properties and also to ensure compliance with regulations and certifications.
Formative evaluation means there is still a chance to adapt the product to findings, while summative evaluation is more like a grade after the fact -- this is what we got.
Where and setupLocation and settings depend on several factors. Some characteristics may require a laboratory to capture or are too dangerous to capture in a real setting. On the other hand, real life observation is less distracting to the people studied than lab observations.
Evaluation without users can be used for predicting how users will act and may get insigths when it is hard to involve real users. The product may be secret or there is no person with relevant experience of the intended product.
Usability testingUsability testing is a more formal evaluation of the ability to use the system to perform a particular task.
Evaluation without users
Heuristic evaluationNielsen et al developed a formal method for heuristic evaluation where the actual user activity is replaced by a set of heuristics, rules adapted to the context, that should be fulfilled to yield good usability and experience. The method is well researched and tested with different numbers of evaluators --> a handful of evaluators are good but a dozen is better.
WalkthroughsSystem developers can formally walk through typical scenarios in a group session to simulate user's needs and behaviours.
Question: How is usability testing relating to the other evaluation methods?