As a product manager I've always been interested and involved with UI and Usability. However over the last few years it was a core area of focus, as my team was responsible for the current and next gen UI of the Siebel Application. In managing UI for an enterprise application that was deployed to 3M+ users I realized that usability is a lot like making movies.
- Everyone relates to them
- Everyone feels they have a killer idea for one
- Everyone feels they can make one happen easily
- Dont make users think (Thanks Mr Krug :-). Aim for simplicity and strive to design a transparent experience where the user can focus on completing the task more than anything else. Trust me, no user wants to be confronted with complexity. Its often (mis-guided) vendors who feel that exposing the complexity of an app to users allows them to demonstrate how rich the app is.
- Leverage standards and patterns and be open to leveraging usability from similar products. Patterns (atleast proven ones) are structured responses to usability problems out there, so dont attempt to reinvent the wheel all the time. But do remember that patterns are loosely structured solutions and so adapt them to your scenario. Also if you're attempting to create a new pattern, remember that its a lot of effort to get a good validated pattern and the probability of success on one of these newer patterns is very slim.
- Know your users and watch them use the product. DO NOT design from a white ivory tower. Also remember this golden rule, internal users especially ones on product teams are not ideal users.
- Put designs back in front of users and iterate on the design.This is one area where I often see application usability initiatives fail. There is often this cover of stealth and secrecy and often some hesitancy to put designs back in front of customers. Well trust me, unless you're designing the next iPhone, stealth and secrecy is not going to work for your usability. As I've learnt in my days managing Siebel Security, stealth and secrecy do not make things more secure and they sure dont make things more usable either.
- When the heat is on stick to your guns.Often I've seen UE folks fold up/cave in/give up (use your favorite phrase) when confronted on their designs. To me that's just a sign of a "not very well thought out" design. If you believe in your design then stick to it, be open to criticism, plan to defend with data, learn to isolate good feedback from criticism noise and even if you have to compromise dont give in on the core principles.