Usability — often used, incorrectly, as a synonym for user‐friendliness — is a quality that describes how easy a tool is to use. In this context, “easy” means: “without (or with minimal) aggravation, and/or with near maximum (or optimal) efficiency”; and “use” means: “utilise to accomplish a task”.
There are sometimes other attributes, such as memorability (“how well people remember and recall their learned knowledge in later and/or other contexts”) or robustness (“how well the tool is designed to operate under abusive operation scenarios”), that are defined by different experts as aspects of usability. But primarily, Usability is about how easily a job can be done with the given tool.
Though many engineering disciplines have had long traditions of investigating and implementing techniques to improve usability, electronic devices — including almost all computer user interfaces — still have very immature usability.
There are several reasons for this. One of them is that, of course, electronics (and computers) are relatively new in the scope of total human experience. Another is that one of the most significant intuitive concepts — feedback, or the degree to which a given cause has a perceived affect — is extremely inconsistent, because most electronics operate on a scale beyond human perception, and software is inherently abstract.
But the biggest problem — and especially as far as websites are concerned — is simply that the designers of the interface don’t know much about designing for usability. When you ask an engineer to add a new feature to a TV, they add a corresponding button to the remote control, and so now, remote controls have fifty buttons where they used to only have fifteen. When a website developer is tasked with making the website more “user‐friendly”, they look at ways that they can reduce the number of links — which makes it easier for them, who use the website every day and know how it’s organised — but which may do nothing to improve the usability for a first‐time visitor, and may even make things worse.
The two immediate ideas that will occur to engineers and website designers alike — “increase the number of options so that users can achieve any task with a single step”, and “reduce the interface to a minimal state so that users are not confused by clutter” — both have their merits, and are not binary choices; both can be used to great effect. But both fail to account for the fact that they did not first ask the question: “How would the user actually use the interface?”.
This is the exact same problem that is asked — and in many cases, answered — by accessibility. As a designer, you are just as unqualified to answer: “How would a person with impaired motor‐control use this interface?” as you are at answering: “How would a person who has never seen a website before use this interface?”. Unless you’ve run a study to find out, or you have researched existing studies.
Achieving accessibility is very much like achieving usability, if you start with the question: “What if it was used this way?”, and if you then go and get answers to those questions before you do anything else. And you won’t achieve either accessibility or usability — except maybe by blind luck — otherwise.