We're working with a local educational establishment on an application which helps them receive student submissions. There is a twice-yearly submission window, which means a period of intense activity followed by 5 months of preparation for the next window.
This forces us to follow a neatly defined 'plan-develop-test-release' cycle, which could apply to any other website, but isn't usually top of the agenda.
During the testing part of the cycle the idea is that you test that the system functions properly given a wide variation of parameters, but also that the system is usable, i.e. that perhaps the less savvy users will still be able to understand and navigate the web application.
However, during the release section of the cycle there are inevitably events we can learn from.
- do users fail to understand the terminology and ask the same questions?
- will users benefit from having different concepts separated over multiple screens, so they don't miss anything?
- do website administrators repeatedly ask us to perform system level tasks, where we could create an 'app for that'?
Yes, the thrust of the usability effort has got to be looking over shoulders of new testers during the test cycle, but usability is also understanding patterns which emerge from mass usage.
A six month cycle can really focus the mind and help a website increase any metric - conversions and sales for example. Let us know if we can help you plan your usability analysis and release cycle.