ACM Interactions

Authors:
Zhengjie Liu, Zhiwei Guo, Kai Qian, Huiling Wei, Ning Zhang

---

 

Along with the economic development and the market globalization process, usability and UCD are rapidly emerging as professions in China. More and more leading Chinese enterprises have started practices in this area. To better understand the current situation, in January 2007 we conducted a study on UCD practice in China as a component of the SESUN (Sino European Systems Usability Network) project (www.sesun-usability.org) funded by EU Asia-IT&C Program. Thirteen IT enterprises (multinational companies were excluded) with experience in UCD were selected for the study, including hardware vendors, software vendors, solution providers, and service providers from different regions in China. They represented the leading players in UCD amongst Chinese IT enterprises. For each enterprise we conducted an in-depth interview with a usability practitioner who has at least one year of experience in UCD in that enterprise. The interviews focused on the key areas of UCD according to the UMM model [1].

  UCD Groups

Usability groups in these 13 enterprises were set up in recent years with the earliest dating back to late 2001. Nine of them are under the R&D divisions of product lines, and two are directly subordinate to the top management of the enterprises. Those not part of the product line were used to support platform or market divisions. The team size in most cases ranged from five to 20 people, with the largest having some 70 members. Most of the team members were transferred from other professional positions like interface designer, developer, or tester. As for the origin of UCD teams, 10 were transformed or evolved from the design departments, with only three cases in which top management set up the team directly.

As for the daily work of the UCD practitioners, it is usually a mixture of interaction design and user research. The projects for UCD teams usually come from product lines. Only in a very few cases were they able to initiate their own projects for some prospective research. A typical UCD project (involving interviews or usability testing study), from planning to reporting, usually lasts one to two weeks. The longer ones could go on for up to a maximum of four months, depending on the specific project circumstances.

  UCD Training for Employees

Enterprises usually have a routine staff training program. However, most of the interviewed enterprises did not cover UCD in their training; only members of UCD groups are eligible for UCD training. UCD teams occasionally offer UCD training to the project teams before the project starts. Only a few enterprises offer such training components to the whole company—usually, the trainees are limited to those from managerial level or some staff members interested in usability. The purpose of the training is mainly for increasing the awareness of UCD concepts, with around half of the cases involving training on specific UCD methods.

  User Participation

Most current enterprise processes require user feedback at some stage in the development, but few require direct user involvement in the early stage of development—this is often due to UCD Team inexperience in how to involve users earlier. Seven of the enterprises surveyed use some form of usability testing method. However, only two enterprises use it in the early stage of development. Most of them conduct it only late in the development or just before delivery.

In all the interviewed enterprises, only one ever used summative usability testing for benchmarking. There are two main reasons for the enterprises not to adopt benchmarking. One is the limitation in methods to get appropriate quantitative measures. Another is that many Internet companies prefer using data like click-throughs or page views as measures for user experience.

Iterative design is a basic principle of UCD. However, only two enterprises adopt this approach at an early stage in the development, relying instead on evaluation by domain experts. The Internet enterprises do not even use domain experts; they do not make evaluation and improvement on design until products are online. Some traditional software and solution providers are influenced heavily by CMM (Capability Maturity Model) [2] and waterfall development models that make iterative design hard to integrate into the existing processes.

  Improvement on UCD

In order to enhance their UCD capability, five of the interviewed enterprises have set up usability labs. Some invested a lot on the equipment, even buying eye-tracking instruments and special analysis software. Some enterprises set up several usability labs at different sites to better support different product lines. Quite a lot of them created document templates for UCD activities. These efforts on infrastructure show the increasing investment that Chinese enterprises made on usability and UCD work accumulation.

Most of the enterprises strongly emphasize self-improvement of the UCD team. A commonly adopted method is to summarize and share experiences after each project within the team. Some leading enterprises even have begun to innovate on methods, or to cooperate with professional institutions aimed at gaining professional expertise. However, many of them still lack the capability to absorb the latest knowledge from international frontiers in the fields.

As for external cooperation across enterprises, although most claim to support this type of initiative, there is little evidence that they are actively engaged outside their own enterprise. This is mainly due to the DIY culture common in Chinese enterprises, where concerns of commercial confidentiality stifle these initiatives. This is compounded by a lack of knowledge and experience on how to coordinate and manage such UCD activities.

  Integration with Existing Processes

Although the enterprises commonly believe that usability is an important product-quality attribute, most of them did not integrate it into their quality system, but left it for the UCD group.

UCD groups work closely with development departments in their daily work. When the UCD groups were first formed, they experienced a lot of trouble communicating with other groups because of the lack of knowledge on technical issues, the limitations on project period, and the low awareness of UCD among developers. As UCD matures in each enterprise, and there is continuous mutual communication between different groups, most groups manage to allay suspicion and find their niche. However, the dominance and authority of UCD groups on user experience issues is yet to be formally acknowledged. Currently, this is up to individual UCD practitioners' professional expertise and interpersonal communication skills.

Finally, the interviewees often complain about the shortage of time available for achieving good quality in their work. The existing processes in the enterprises did not provide enough space for UCD activities. Practitioners have not found a good way to fit their work with short and quick development projects.

  Use of UCD Methods

The enterprises examined have used a wide spectrum of UCD methods. Some of them even tried to develop new methods based on their needs. They indicated methods like competitive product analysis, usability testing (defect finding), prototyping (high fidelity), expert evaluation, and interviews as the most frequently used. However, we also found that many enterprises do not have a systematic and comprehensive knowledge of wider UCD practice—they might have extensive knowledge on some methods they frequently used, but at the same time very poor knowledge on others (including some commonly used in the industry). In general, UCD practitioners' skills and experience on UCD still needs to be improved.

In most of the enterprises, UCD work has not penetrated the early stage of development. Product concepts are usually defined by the marketing or product line departments. Therefore ethnographic methods such as contextual inquiry and the like are rarely used.

  Maturity Assessment

We did an informal usability maturity assessment of the enterprises based on the UMM model—a model developed by the United Nations Centre for Trade Facilitation and Electronic Business to support business development [3]. It seems that the hardware vendors are the best, and the service providers are the poorest, with software vendors and solutions providers in the middle. The fierce competition and the traditional emphasis on design in the hardware business, especially the consumer electronic product sector; the strong influence of software-engineering models for software vendors and customized solutions providers; and the rapid response to market demand for Internet service providers might be an explanation of the results.

  Concluding Remarks

This study gives us an overview of the current UCD practice in the Chinese IT industry. Most leading Chinese enterprises have had two to three years' experience in this field. Some of them achieved great success and have succeeded in making UCD well accepted by the management and product development teams. UCD teams are continuing to grow in size and number and are developing their multidisciplinary expertise. UCD practitioners are generally optimistic about the outlook and are satisfied with the current situation. However, there are still a lot of challenges to face in this area, such as:

• Improving the expertise of UCD teams
• Changing the product lifecycle to use UCD at an earlier stage
• Promoting UCD within a company

We believe the UCD practice to be growing rapidly within China. Given that the data for this survey was collected a year ago, we can already assume progress on the challenges listed above. In fact, we could make some predictions on what might come to pass in next five years in this country based on our recent observations:

• Although some UCD groups might experience failures due to lack of skills and immature UCD environments, there will be a lot more organizations that start to practice UCD, not only industrial enterprises but also government agencies; not only big companies but also SME startups.
• UCD practice will extend from commercial areas to noncommercial public services areas.
• The UCD groups will be gradually upgraded to higher organizational levels and become more integral to the company's strategies.
• User involvement and UCD activities will be extended to the earlier stages of development and correspondingly more fundamental and longer-term projects.
• UCD practice will be very mature and well integrated into the processes of some leading enterprises, catching up with their Western leading counterparts.
• There will be a steady demand for experienced and trained UCD professionals as well as for UCD training; UCD will become a coherent training component for all employees in more and more enterprises. More than the current three universities will provide UCD-related educational programs.

  References

1. Bevan, N. & Earthy, J. "Usability Process Improvement and Maturity Assessment." Proceedings of IHM-HCI'2001, Cepadues-Editions, 2001.

2. Watts, H. Managing the Software Process, Boston: Addison Wesley Professional, 1989. 69

3. Earthy, J. "Usability Maturity Model: Processes." Project Report, Lloyd's Register of Shipping, London, 1999.

  Authors

Zhengjie Liu
Dalian Maritime University
[email protected]

with Zhiwei Guo, Kai Qian, Huiling Wei, and Ning Zhang

About the Authors

Zhengjie Liu is founder and director of the Sino-European Usability Center (SEUC) (www.usabilitychina.com), professor of HCI at Dalian Maritime University, and cofounder and cochair of ACM SIGCHI China (www.hci.org.cn). He has been working in HCI and usability in China since the early 1990s and is a pioneer in the fields there. In 2000 he founded and led SEUC, conducting research and providing consultancy services.

Zhiwei Guo, Kai Qian, Huiling Wei, and Ning Zhang are postgraduates at the Sino-European Usability Center, Dalian Maritime University, working in usability since 2005.

EDITOR

Gary Marsden
[email protected]

  Footnotes

DOI: http://doi.acm.org/10.1145/1353782.1353799

 

©2008 ACM  1072-5220/08/0500  $5.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2008 ACM, Inc.