Jeff Hendy      jchendy@cs.ubc.ca

Graphically enhanced keyboard accelerators
Masters thesis

Graduate student recruiting web page
Web design contract at UBC

CPDF management tool
Internship at Pitney Bowes

ATMReserve
Course project

Single display privacyware
Course project

ATMReserve

Course project, Human-computer Interaction, 2007

Additional team members: Petcharat Viriyakattiyaporn, Michael Welsman, Kai Xu, Susana Zoghbi

Overview

This was a team project in the graduate HCI course at UBC. It involved research, design, and evaluation of an interface for improved interaction with Automated Teller Machines (ATMs).

Key skills used

Problem and design approach (from project report)

There have been dramatic changes in the banking industry in recent decades. Automated Teller Machines (ATMs) have superseded human tellers for many banking tasks. Many users prefer ATMs to human tellers, but ATMs can sometimes be slow since they enforce a simple, linear use flow. For users who want to complete many transactions, this means spending several minutes at an ATM. Whether they have a long line behind them, in front of them, or are simply alone late at night, time spent at an ATM is something most users would rather avoid. Online banking (OLB) has made it possible to do many kinds of transactions at home, avoiding this problem, but users must still go to an ATM if they wish to withdraw physical cash or deposit cash or cheques.

Our system proposes a new macro transaction feature that allows users to shift more of their banking time from ATMs to online banking. Users can pre-define sets of multiple transactions from a flexible web-based interface. These become available at the ATM and require only authentication and confirmation rather than a long string of sequential operations. This cuts down on undesirable ATM interaction time considerably, improving user experience. Furthermore, our system will increase ATM throughput, reducing wait times in ATM queues and perhaps reducing the number of ATM machines that banks must deploy.

As we are designing a banking system that can be used to perform a number of standard transactions, we must take a wide variety of users into account. Most of these users want their banking systems to be simple and intuitive. It is safest to design for a novice rather than an expert user, although advanced functions can be beneficial if they do not overly complicate the system. We should also keep in mind the fact that most potential users have an existing mental model of how various banking operations work based on current ATM interfaces.

The capture the diversity of our user population, we created three imaginary users with different goals, levels of sophistication, and disabilities. The first is Edmond, an elderly man who prefers to do banking in person and deals with large numbers of cheques. Edmond requires a system that is simple and easy to use. Next is Jen, a young manager at an IT company, who has a higher expertise and would like her banking interfaces to be fast and powerful. Finally, Mary is an intermediate user with standard needs. We have developed tasks for these three users.

For the purposes of this project we are assuming that we would be unencumbered by most technological concerns. Our system is designed to be deployed on a new generation of ATM which will be standardized, will have flexible and fast connections to back-end databases, and will have new features such as high-resolution touch screen displays and biometric identification. For the OLB piece of our project we are assuming that our system will run on a standard home computer with a standard web browser and internet connection. Identical or similar web applications could be accessed through web-enabled devices such as cellular phones, but dealing with small screen displays and other compatibility issues is out of scope for this project.

Images

Initial Sketches

Interactive Lo-Fidelity Prototype

Final Medium-Fidelty Prototype

Documents

Project Proposal (PDF, 6 pages)

Requirements and Initial Design (PDF, 27 pages)

Redesign and Final Evaluation (PDF, 16 pages)