Well it all of course depends on what type of project you are doing. There are 4 main types of projects as far as I am concerned.
1. Redesigning an existing platform 2. Designing a new platform but something that there already exist best practice and an audience for (i.e. a competitor to Flickr) 3. Designing something with no established best practice (i.e. something new and unique, such as a keyboard controlled by your pinky or a platform or an interface for a car that drives on electricity ala Better Place. 4. Designing something with established practice but a new type of audience (i.e. a platform for connecting refugees) Normally your project falls into one of these 4 categories with category 3. Being the very rare occasion. My proposal normally is to get the user involved in the beginning to figure out "what kind of tasks are the user trying to solve" Not what do they want, how do the user want it to look like, what kind of ideas do the user have. Of course there are sometimes the possibility to find some gems from users inputs, but that is hardly gems that will make the investment worth while. If their gem is so important it's a showstopper to the success of whatever you are doing I would say that you got a whole different type of problem. It's like digging for gold but only finding plastic pearls. No what you want is to get an understand about some of the problems the user have on a more holistic level. Cause that will help you inform your solution and not the actual design decisions. This in return makes sure that you have taken users into account i.e. you are actually looking at what problems they have and NOT whether they think you solved their problems, which is the most used process for UCD from what I have gathered throughout the years. Only the third type really warrants continuous user involvement IMHO, but because you are really testing something different which is the learning curve and not the actual solution. I say this because I believe that in most cases, 99% of the time you can't introduce something new without some sort of learning curve. The real trick is to figure out whether this learning curve is worth it or not. I.e. are you helping the users solve something they couldn't solve any other way and is the time it will take for them to solve it worth it. But this will relate back to whether you are helping the user solves tasks they are trying to accomplish. I then propose that you do some in/house testing for stability of your solution and to see whether your solution do as you intend it to do. Again this goes back to my "how" not "what" principle. Even slight changes in feedback from rollovers, transitions placement, colors etc. can have huge impact. This impact is big because that is where the solution comes alive really. That is where people relate to it, that is what they might (if you are lucky) create an emotional bond with. They like to use your interface, not because how it looks, not because of your design patterns, not because it\s coded in ruby, not because it has a carrousell but because all these things play together to create the experience. How can anyone be so bold to claim that they test the experience by making usability tests or focus groups is beyond my imagination. I never understood it and probably never will. To me the advocate for the users are those who actually look at the problems and tasks they want to solve and design solutions for it, not those who claim to be advocates beacause they value user input over a developer or mangement. By listening to what customers really want the middle section is not necessary and you can involve the users where it really matters which is in the launch of the product. Where all the excuses are gone, where users don't have to imagine the real data, but where the data is real, where everything they see is a companies attempt to help them solve their problems. Not solve the problems that testing in the middle of the process creates for proper feedback. That's just my five cent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Posted from the new ixda.org http://www.ixda.org/discuss?post=45640 ________________________________________________________________ Welcome to the Interaction Design Association (IxDA)! To post to this list ....... [email protected] Unsubscribe ................ http://www.ixda.org/unsubscribe List Guidelines ............ http://www.ixda.org/guidelines List Help .................. http://www.ixda.org/help
