TaskZ.com
homehomedesign data
 
UCD and the next generation of interactive technology

User-centered design, or UCD, is a comprehensive proprietary software development methodology driven by (1) clearly specified, task-oriented business objectives and (2) recognition of user needs, limitations, and preferences. Information collected using UCD analysis is scientifically applied in the design, testing, and implementation of products and services. When applied correctly, a UCD approach meets both user needs and the business objectives of the sponsoring organization.For decades, UCD has been used successfully in the design of products, services, and user environments in several industries including telecommunications, pharmaceuticals, and the military. It is safe to say that without UCD methods many of the great advances of the last 50 years would not have been possible, including manned space flight, commercial aviation, mass communications, and computer advances (including the graphical user interface). The point is that all of these industries have used professional UCD methods to create design solutions that are a brilliant mix of human capabilities and advanced technology. Based on this experience, the following has been clearly established: The design of screen-based systems must be driven by features and functions that match the cognitive limitations and capabilities of the user.

The end of technology-centered design
Traditionally, software development has been technology-centered and feature-driven. There are several well-known problems with this methodology, including poor usability. Technology-centered design grew out of the work group development methods of the 1960s. This approach was an effective means of managing large and complex programming tasks involving close interaction between hardware limitations and software code production. The focus in these programs was reliability and hardware performance.

Today, when designing intranet systems, e-commerce web sites, or other screen-based systems, the central concern is not hardware performance, but user performance. These systems are meant to be efficient and easy to use in order to improve the user's productivity and overall experience. Because it has a negative impact on the user experience, technology-centered design directly conflicts with these goals. Therefore, many companies are being forced to abandon technology-centered design and adopt UCD methodologies. At first, migrating to UCD methodologies may be problematic for development teams that are accustomed to working from the technology up. However, the need to develop high quality products that respond directly to the needs and limitations of the customer far outweighs the initial difficulties that teams may encounter.

User-centered vs. technology-centered processes
When compared to a traditional technology-centered approach, the overall UCD process is radically different in six important ways:


1. Development of driving functions. The UCD process begins with formalization of the strategic business objectives that are then translated by the management team into functional specifications that can be used by the UCD design team to create and test proposed solutions. By contrast, the technology-centered approach is often driven by a comprehensive list of technical specifications directed at system hardware performance. In a well-structured UCD program, the objectives are focused on defining user needs and meeting real business objectives. Technical execution is always a concern, but never a driving function in and of itself.
2. Basing design concepts on customers' tasks and critical experiences. In the UCD process, design concepts are derived from the execution of a comprehensive task analysis carried out by professional usability engineers with users from all profile types. By contrast, the technology-centered approach bases the conceptual design on loosely defined functional specifications that are often derived from marketing studies. Such studies are generally focused on refining a list of features and ignore the customer's interactions with the system. Without a comprehensive understanding of customers' task and session objectives, a world-class design solution is impossible to attain.
3. Composition of the development teams. In a UCD process, the entire development team stays together during all steps of the process. There are no "hands-off" in the traditional sense. Therefore, the entire team is responsible for the complete design, not just for a discreet package that is then passed on to the next development group, as in the technology-centered approach.
4. Integration of customer feedback in the development process. Under the UCD methodology, customer response to the design of the screen-based experience forms the basis of all design decision-making. The process is heavily front-loaded to ensure that all of the customer's needs and limitations are accounted for before any coding is started. In the technology-centered process, customer feedback is often seen as a means of identifying bugs or minor problems well into the coding cycle. Under this development approach, the design team assumes the role of the customer in the process and makes design decisions based on personal opinion and experience. History has shown such opinions to be biased, leading to poorly designed screen-based experiences.
5. Design freeze time. In the traditional technology-centered process, functional specifications are frozen very early in the development process. By contrast, under UCD design, freeze does not take place until the overall customer experience has been fully created in simulation form and tested for content, style, and user compatibility. The differences are critical. In the technology-centered process the design freeze documentation is a written specification. In the UCD process it is a fully featured simulation that has been tested against customers' expectations and profiles. Therefore, the confidence that a corporation has in the potential success of the design solution is dramatically higher. By contrast, the technology-centered approach places far too much demand on the creativity and skill of the development team's various members by requiring them to interpret the functional specifications, design the system, and represent the response of the customer. This is impossible, and is one of the primary reasons why so many traditional technology-centered software design programs are over-budget and beyond schedule.
6. Use of design iterations in the development process. In the traditional technology-centered approach, design iterations are problematic because they usually involve rewriting complex code. Therefore, every attempt is made to minimize design changes. This is true, even in the earliest stages of coding. As a result, poorly designed features and functions of the software are propagated over an entire software design. By contrast, the UCD approach demands design iterations; so far better solutions emerge in less time. The overall screen-based customer experience grows by iterations of creative problem solving and structured customer feedback. The design, prototype, test, and redesign process is highly iterative and may involve several cycles. It is important to note that hard coding is not undertaken until after a final design has been developed and tested with different user groups.

The benefits of UCD
User-centered design has numerous benefits for business. Firstly, UCD methods result in higher-quality screen-based systems with increased customer satisfaction and confidence. In the competitive online market, a high-quality design can mean the difference between success and failure. It's a fact that customers abandon sites that are cumbersome and do not meet their needs. On the other hand, customers remain loyal to sites they can navigate and feel are useful. Forrester reports that 90% of customers would shop again at sites that met their expectations and 87% would tell friends and family about the site.

Secondly, under UCD methodologies, software development is more efficient. The system will likely go to market faster and cost less. The following table demonstrates that the average time required to fix a problem with software design or specifications is quite long and therefore very expensive. With a UCD approach, the product's design and specifications are frozen late in the development process. The method's constant customer feedback and design iterations weed out potential problems before they require hundreds of hours and thousands of dollars to fix.


Defect Introduction and Cost of Fixing the Defects Introduced
 
Project #1
 
Project #2
 
Phase introduced
% Defects introduced
Average fix time
% Defects introduced
Average fix time
Requirements/
Specifications
--
--
3
575 hours
Design
20
315 hours
15
235 hours
Implementation
75
80 hours
65
118 hours
Test
5
46 hours
17
74 hours
(source: The Science of Debugging, Telles and Hsieh, 2001)

In Software Engineering: A Practitioner's Approach, author Robert Pressman shows that for every dollar spent to resolve a problem during product design, $10 would be spent on the same problem during development, and multiply to $100 or more if the problem had to be solved after release.

Companies can count on long-term savings as well. With a more usable system, customer support costs are reduced. In addition, companies that equip their employees with UCD products will see training time fall and productivity rise. Employees are more satisfied with intuitive, UCD products that make their work easier, not more complicated.

The future is User-Centered
With so many e-business pundits sounding off on the importance of customer satisfaction in the new economy, it is surprising more hasn't been said about UCD. Now that companies are recovering from their initial technological euphoria, many are beginning to think about how they can tame technology and adapt it to better serve their customers. UCD has proven itself to be the best method for relating technology to people. It is finally time to apply this time-tested methodology to the interactive, screen-based systems that are increasingly important to our everyday life.


Email the editor with questions and comments.


    ViewPointz
Columns by:
Charles L. Mauro, Editor

Ken Keller, Esq.
Henry Lichstein
Deborah J. Mayhew, Ph.D.
Elizabeth Rhodes
Jef Raskin
Carol Righi, Ph.D.
Scott Isensee