We approach the usability of whatever we're testing, whether that is a website, web application, mobile application or anything, to understand who is intended to be using that thing and what are they expected to be able to accomplish.
In one sense we are testing from the point of view of the user, can that user accomplish the actions, tasks or journeys that would be expected and what kind of problems can we find.
But we also think like a tester, to hunt for all the problems that could trip users up or where they could get into a state where they can't proceed or understand how to proceed or exit a journey.
A classic and simple example is filling in a form on a mobile device - the user completes all the fields in a way that they believe is correct, they scroll to the bottom of the form and tap the Submit button. Nothing appears to happen, the user believes the submit button did not work. What has actually happened is that a validation message has been displayed but at the top of the form, out of view of where the user currently is in the form.
There are many frustrations and problems like this across every website and application we test, some are critical problems that affect the abiity of users to complete actions and many are minor annoyances that make using the website or application more difficult than it needs to be.
We specifically go looking for the problems that make tasks and journeys harder for users, or where messages or instructions are difficult to understand or don't make sense, or how inconsistencies across the site can affect the users ability to use it effectively.
The usability testing that we complete is frequently completed as part of a typical project. We also provide user testing services where we provide a group of real users and task those users to carry out a series of scenarios, noting where and how those users get stuck or voice frustrations with the application they are testing.