Even though ATMs are devices most of us use multiple times a week, if you watch people using them you can see that with each screen refresh the user is putting effort towards parsing the information before them and hesitating before they make their choices. I believe this is not just because the stakes are high (it is money, after all) but because the nature of ATM interactions are at odds with the typical ATM interfaces: ATMs require the user to make a quick succession of choices (“Cash? Transfer? Deposit?” followed by “Savings? Checking?”, etc.) and at each juncture, the screen is redrawn, forcing the user to re-orient themselves and study the choices onscreen. One solution is to use fluid animation to transition from state to state, but here I’m proposing an interface where all the options remain onscreen and each new choice visually extends out of the previous choice. The user always knows where they are, every choice to make is already within context, and any choice is easily reversed or changed. Eventually, the user will be able to visually recognize the pattern for tasks they want to perform, aided by muscle memory.
This interface sketch is based on a Wells Fargo ATM experience, and assumes the user has already logged in. All possible actions remain onscreen, and the user can choose to interrupt any action with another button action. For this demonstration, a few sample paths of action are indicated by a blue arrow as seen here (on the Get Cash button, for example).
Once a transaction is processing, most buttons will become disabled, appearing in faded gray.
Try clicking the following buttons:
- Get Cash
- Deposit Money
- $100 Cash (Shortcut)