Arjan Huisman of Elsevier’s Customer Service Focus Project Reports from Amsterdam

Arjan Huisman
Q: Does Elsevier offer support in different languages?
As Elsevier is a global company, customer service improvements are sometimes small but potentially deliver big impacts. In this column I want to tell you about our initiatives to provide support in various languages.
As you can see on ScienceDirect, Scopus and Elsevier.com, the primary language used in our products and on our websites is English. Also, our printed support materials are mostly provided in English. We do however provide support in other languages as well.
Whether inquiries concern print or electronic products, our offices worldwide provide support to librarians located in nations ranging from Albania to Zimbabwe. Our support staff members speak languages including Dutch, English, French, German, Italian, Japanese, Korean, Malay, Mandarin, Portuguese, Russian and Spanish. (I’ve probably forgotten at least a few other languages you can find when contacting our support staff.)
We try to speak or correspond in customers’ own languages when responding to queries, but unfortunately are not always able to do so. We do however guarantee that, based on the region you're located in, our support staff is available to speak to you in English, Japanese or Mandarin.
How else is Elsevier providing local-language support?
- As often as possible, our training — delivered live online or in person — is presented in the language of the nation where participating librarians or end users are located. We think you can benefit most from training when it’s given in your own language.
- We’ve translated Library Connect, ScienceDirect and Scopus brochures into local languages. For example, our Library Connect pamphlet “Ways to Use Journal Articles Published by Elsevier” is available in English, French and Japanese.
- ScienceDirect and Scopus invoices now are accompanied by sheets offering additional information on how to interpret the invoices. Those sheets are provided in English, German, Italian, Japanese and Spanish.
We're constantly seeking to expand our support to customers located in specific countries and speaking specific languages. To determine new ways to provide local-language support, we’re listening to you, our customers! So please keep your input coming. If you have customer service suggestions, please send them to A.Huisman@elsevier.com.![]()
- ScienceDirect interactive tutorials in multiple languages
www.info.sciencedirect.com/using
www.sciencedirect.com (under Help)
- Library Connect practical assistance pamphlets in multiple languages
www.elsevier.com/libraryconnect
- Contacts Finder for Librarians
http://contacts.elsevier.com
Tom Noonan of Elsevier’s User Centered Design Group Answers Your Usability Questions

Tom Noonan
Q: How many users are enough when I test the usability of my library website?
A: The number of users required for usability testing is a controversial subject. To answer this question, it’s important to keep in mind your purpose for conducting the test, your target audience and tasks that are most critical for your users to accomplish. Depending on these considerations, the number of test participants you need might differ.
For example, do you need to test a small piece of functionality with a well-defined set of users or a broad portion of your site with a diverse user population?
If your purpose for testing your website is to discover usability problems, you can learn a lot with surprisingly few users; five is a common number of users for such a test. Each user you test will expose problems. As you test more and more users, you will begin to see the same problems over and over again, along with some new ones. At some point, the law of diminishing returns takes over.
What’s important is that you address the problems identified, and then test again to see if design changes have fixed the problems. Using a small number of users per test makes it easier for you to test, modify your design and test again.
Be aware though that some researchers have suggested that even when testing a large number of users, new problems — even major problems — continue to be found. If you think of a problem in terms of its probability of occurrence, then a low-probability problem likely won’t be observed unless you test a large number of users. With that in mind, you must decide upon the level of risk you can accept when it comes to identifying low-probability problems.
![]()
- Barnum, C. M. (2003, January). What’s in a number? STC Usability SIG Newsletter Usability Interface, 9(3).
www.stcsig.org/usability/newsletter/0301-number.html
- Nielsen, J. (2000, March 19). Why you only need to test with 5 users. Jakob Nielsen's Alertbox.
www.useit.com/alertbox/20000319.html
- Perfetti, C. (n.d.). Usability testing best practices: An interview with Rolf Molich.
www.uie.com/events/uiconf/2006/articles/molich_interview/

