This storyboard follow Kevin's design sketches #2 closely. The idea is to present many possible options to the users who are clueless about what they want to get. A user who does not know what tattoo to get, when presented with options, will know which option the user likes best. This process will be repeated several times, and each iteration will present to the user a more specific set of options. Which options are presented will be based on categories (not shown to the user) of each image that the user chose. Moreover, to educate the user, each iteration will be accompanied by a useful fact about tattoos.
In the end, the user will have the following things:
This interface of the image selection process has good learnability, as all the user has to do is choose only one option out of all the presented images. As long as clear instruction is displayed on the screen, the user should be able to navigate through the process without any problem. As the user proceeds, the user will then be able to change his selections by viewing the bottom panel and click on the [x] buttons at the corner of each images. Again, this is a very simple interface.
Currently, the efficiency of this interface depends on how fast the system's confidence on the user's preference converges to a threshold value, at which point the user will be presented with the suggested artists list page. Making the user go through the selection process just to user the artists search page is also not good for efficiency; perhaps a shortcut should be given to the user at the beginning of the process to go to the artists search page.
The only places where the user can do wrong are: