Offshore Usability

by Jakob Nielsen on September 16, 2002

Summary: To save costs, some companies are outsourcing Web projects to countries with cheap labor. Unfortunately, these countries lack strong usability traditions and their developers have limited access -- if any -- to good usability data from the target users.

It's currently too expensive to develop websites and intranets, so project managers are looking to reduce costs any way they can. In the long run, expenses will likely drop for two reasons:

  • Better tools, middleware, and platforms, as well as better Web services and application service providers. Currently, too much software is homemade or built with inadequate tools that were thrown together during the gold rush days.
  • Standardized user interfaces that free designers from reinventing everything and let them focus instead on their specific project's unique content and features.

For now, however, managers are looking to cut costs, and one way do it that's currently gaining favor is to go offshore for all or parts of the project. For example, one of the Intranet Design Annual 2002 winners was headquartered in the United States, but hired a firm in India to design its intranet.

Offshore design and development of websites and intranets can present potential difficulties with respect to usability. One problem is temporary and can be overcome in the long run; the other is more fundamental, but can also be overcome if it's recognized and addressed explicitly.

It is obviously not a solution to separate design and implementation since all experience shows that design, usability, and implementers need to proceed in tight coordination. Even separating teams across different floors in the same building lowers the quality of the resulting product (for example, because there will be less informal discussions about interpretations of the design spec).

Temporary Problem: Lack of Interaction Designers and Usability Professionals

The main offshore countries are India, Russia, and China. All three countries have many incredibly talented programmers and have strong artistic traditions. In many ways, the classic offshore countries are well positioned to take on website and intranet projects from companies based in richer countries.

Unfortunately, India, Russia, and China do not have strong traditions in human-computer interaction, and they have very few skilled interaction designers and usability professionals. Presentations from these countries at international HCI conferences are inevitably much too theoretical and formal to be of any use for practical development projects. Because universities tend to focus on overly abstract and formalistic topics, few students will acquire the skill set needed for user interface projects.

Even though Hong Kong is officially part of China, I don't include it when lamenting China's lack of usability professionals. On the contrary, there is a thriving usability community in Hong Kong. An interesting scenario is that Hong Kong's usability professionals might bootstrap China if they were given sufficient resources and the authority to direct development projects and train the teams.

The Indian National Association of Software and Service Companies predicts that India will employ 4 million people in the offshore IT industry by 2008, generating about US$63 billion in export earnings. Using my standard rule of thumb -- that 10% of project resources must be devoted to usability to ensure a good quality product -- India alone will need to train 400,000 usability professionals in the next six years to meet its own projection.

Since it takes a long time to become a truly skilled usability professional, we won't have the huge number of people we need in offshore countries any time soon. Eventually we will, but it will take time.

Fundamental Problem: Designing Usability at a Distance

If local authorities or company managers decide to invest in usability, professionals can be trained. However, offshore design raises the deeper problem of separating interaction designers and usability professionals from the users. User-centered design requires frequent access to users: the more frequent the better.

One solution is for offshore projects to change the balance between the two main usability methods. That is, do more heuristic evaluation and less user testing. Still, some user testing will always be needed, so we need ways to make this happen. Also, good heuristic evaluation requires that usability specialists have a deep understanding of usability principles, and they can only derive this from watching a wide range of user behaviors. So, if the usability specialists have not observed much user testing, they won't be as good at conducting design reviews.

Similar considerations hold for interaction designers. Professional growth in that discipline comes partly from frequently observing user behavior, which may not happen in offshore locations. We know from many previous studies that it is hard to design for international users, so cultural and language differences make a hard job even harder for offshore interaction designers.

The lack of user testing might be alleviated by running tests with local users, but such surrogate users are unlikely to exhibit the same behavior as the target users, especially for intranets and vertical websites. There are always some issues that can be found by testing any human being; many usability problems are caused by an utter mismatch between humans and computers. But many other usability issues require tests with real, representative users.

The only viable solution here is to emphasize remote usability tests, where offshore usability professionals test target users remotely in the users' local labs. Although I prefer tests where the facilitator sits in the same room with the user, many excellent professionals prefer to sit behind a one-way mirror in the observation room. Once you are in a different room anyway, you might as well be 10,000 miles away and observe through a broadband connection. And, we'll soon have multimedia connections that offer a view comparable to what you'd see through the glass.

Remote testing is already being used in traditional international usability testing. For offshore usability, we might modify it so that each test session would be observed by a large group of usability professionals. The test session might be projected onto a huge screen in an auditorium, for example. Typically, the goal of a usability test is purely to improve the design, and the facilitator's professional growth is merely a side benefit that's not even measured. For offshore projects, however, the ability to observe target users might be rare enough to warrant squeezing every bit of educational benefit out of each study. By gaining an understanding of target users' behavior in this way, offshore usability professionals will become better qualified to review designs.

The potential growth in offshore Web projects might present opportunities for a niche industry in the U.S. and Europe: Companies could establish usability labs that have no observation rooms or test facilitators, but are capable of projecting test sessions to offshore locations.

Share this article: Twitter | LinkedIn | Google+ | Email