Usability Testing on a Shoestring: Test-Driving Your Website
By Danielle Becker
Would you buy a car without driving it first? Probably not. Then why would you consider launching a website without at least taking it out for a test-drive? Usability testing gives you the opportunity to get it out on the open road and work out the kinks. Before you put someone behind the wheel, there are some strategies you should consider when trying to keep costs down.
Advice on usability testing is abundant. Cutting through the forest of suggestions, opinions, and warnings to pinpoint what’s useful in your situation can be daunting, particularly when you’re budget-challenged. First, let’s be clear on the basics of usability testing:
Usability testing is a way to observe users interacting with your website. Are they finding what they want when they want it?
Usability testing is a tool to discover if users are using the site as it was intended. Are they interacting with the site in ways that the design team didn’t anticipate?
Usability testing is best when done repeatedly and the information gathered is applied to the following stages of the website redesign. If the users don’t understand a label or a search box isn’t functioning as intended, you can fix it and test it again until it does.
Usability testing is not to be used as a tool to validate your design decisions.
Usability testing is not useful when done after the site is finalized and few changes can be implemented.
Usability testing is not identical to a focus group.
Usability testing myths debunked
Hiding in that forest of usability advice are some usability myths. Here are a few that need to be debunked:
Hiring professionals who specialize in usability testing is the only way to effectively test a website. You can design and facilitate a test that gives you invaluable information. There are many good reasons to hire outside help, but it isn’t mandatory. With a little effort, your tests can be just as effective.
A lot of research in usability testing theories and best practices is required to properly write and conduct a usability test. You can research as much as you want or not at all. You want to know how well your site works and what the pain points are. This article will give you examples of how to get started.
Usability testing requires a large sample size of test subjects to get the most accurate results. I’ve done it both ways (a larger sample size conducting one test, smaller sample sizes doing multiple tests) and discovered that fewer participants and more frequent tests gave us the most valuable results. Popular opinion these days suggests that testing five users per session is plenty.
Usability labs are the most suitable places for usability testing. If you have a table, chairs, and a computer terminal, then you have all the equipment that is necessary to start testing. If you have video cameras, eye-trackers, and screen capture software, that’s great and makes testing a little easier. However, it falls under the “nice to have” category, not the “need to have” category.
You need a lot of money to offer meaningful incentives to your test subjects. There are many creative ways to entice your target users to participate in your study that can fit any budget.
Checklist to get your testing started
Having reviewed what usability testing is and is not, I’d like to offer some suggestions about how you can do your own usability testing without putting undue strain on your budget. I use the following checklist:
Write a list of objectives of what you hope to learn by usability testing.
Determine who should do the testing. Do you want a monitor (person who gives the test) and an observer (person who takes notes)?
Write the budget and decide on testing incentives.
Decide how many users you want to participate and how many tests you want to conduct. Then decide who you want to test and why.
Select a time frame for conducting and completing the tests.
Secure the equipment and room where the testing will take place.
Write the questions and establish the protocol for administering the tests.
Finalize the costs and incentives that will be offered to test subjects.
Conduct a series of mock tests on anyone not involved in the project.
Rewrite the questions based on results of mock tests.
Decide how to recruit test subjects.
Figure out what you want to do with the results.
Recruiting test subjects on a budget
Academic libraries can partner with course instructors to offer extra credit incentives for completing a usability test. It’s best to approach instructors before the semester begins, while they are still designing their courses. It doesn’t matter which courses you choose to test, but I’ve had the greatest success with undergraduate courses with many sections such as English or speech/communications. Those courses enroll the greatest number of students, and many instructors are more flexible with their course design. The key is perseverance.
Another rich resource for all libraries recruiting users is your local library school. These students are bright and enthusiastic, and they catch the details that many students without an eye trained toward the information profession may not catch. Many of these students are grateful for the opportunity to meet other librarians, network, and observe your library in exchange for taking the test. Some of the best feedback I recorded in the beginning of our testing was from library school students. This is also a great way to meet your future colleagues, interns, or volunteers.
Library school students may participate in usability testing for the fun of it or because they’re naturally inclined to be helpful. This doesn’t work for other potential test subjects. They need an incentive to participate. When you’re on a limited budget, this could be a problem. However, it’s a solvable problem.
Approach departments or local businesses to donate cash, snack foods, drinks, gift cards, or prizes to users completing tests. In return, these businesses get free advertising. Their names can be splashed on posters throughout the library that are advertising the opportunity to users, in your local or campus newspaper, or on flyers posted around campus or on local business bulletin boards. A great way to get the most mileage out of these testing incentives is to give a smaller prize for each completed test and then enter the test subjects’ names in a drawing for the larger prizes when all the testing is completed. This drawing can be done in conjunction with a modest website launch party in the library, inviting all tested users and stakeholders. Offer refreshments (if your campus has a dairy unit, ask for free ice cream and have an ice cream social) or takeaways such as pens, stationery, or bookmarks, and decorate the lobby of your library so everyone coming into the library knows about the launch of your new website.
You could host a pizza party. This works best for usability tests not using the talk- aloud protocol. Administer the usability test in paper or online in a computer format to students during heavily advertised and scheduled sessions. Schedule the sessions around the lunch or dinner hour so when the users have completed testing, they can be ushered into a conference room where pizza is being served. Tell your local pizza parlors what you are doing and ask for price quotes based on the number of pizzas you plan to order (usually two to three slices of pizza per person). They’re normally ready to offer free pizzas or discounts.
If you have the luxury of a small budget, you could offer cold cash. The premise of this incentive is that a small amount of free money is better than none at all. Calculate how long each test takes, post sign-up sheets at the contact points in your library (circulation desk, reference desk, or table in the lobby staffed by a recruiter), and offer $5 for completion of a test. If the users entering your library have the time to make a stop at your library, usually they have the time it takes to complete a test. The sign-up sheets are useful in getting a commitment from the users and keeping the tests flowing.
Awarding bookstore gift certificates uses the same rationale as the “cash in hand” incentive, except it works best for libraries that cannot (for whatever reason) give cash. We discovered that a $5 gift card is too little to buy much at our campus bookstore, but $10 was enough to draw users. It also gave students who buy their books online and never go to the campus bookstore a chance to see what the bookstore has to offer.
A usability test mock-up
Once you’ve lined up your usability testing participants, you need to construct the test itself, including the venue. Here are some suggestions:
1. Conduct your tests in a private space without distractions.
2. You can complete the tests with both a moderator and an observer (two people). Alternatively, one person can observe and moderate.
3. Introduce yourself and your role: “My name is Danielle, and I am the web librarian. We are in the process of redesigning our library’s website and are asking people to answer questions and complete a series of tasks while using our website. If you could answer the question or complete the task while talking aloud, that would be great. There are no right or wrong answers. The point of this test is to try to discover areas of improvement for our website draft before we launch the site. Please take a few minutes to look over a copy of the questions and then we’ll begin.” Hand the user a printout of the questions you’ll be using.
4. Begin by asking basic demographic information that might inform you as to who your user is. In our case, we asked what year the students were in school, what their departments or major programs were, if they were transfer students, and if they’d attended a library instruction session.
5. Ask the first question, again reminding them that it is helpful if they talk aloud while they are navigating the site. You will want to sit behind them or at an angle that easily allows you to watch their screen as they answer the questions. This allows you to make observations and take detailed notes (this is why having a moderator and an observer taking notes is helpful).
6. Throughout the test, when users ask if they got the correct answer, remind them that there are no right and wrong answers. If they are stumped on a task and ask for help, you can rephrase the question, but don’t help the user find the intended target. Instead, you can show them where you’d hoped they would navigate to find the information and ask them if they have any suggestions on how to make that information more accessible to them. For example, we changed “Circulation Policies” to “Checkout Policies.”
7. We ended the test by asking what we called “fun preference questions” to learn how often the students are in the library, what they use the library website for, and what websites they normally visit. These questions gave them a chance to give their opinions and state their preferences. This information was useful in telling us which websites might be helpful to link to and how our students use the library’s website, the library, and the internet in general. This section is optional, but if you are interested in knowing more about your users’ habits, then adding this section is something to consider.
8. End the session by asking the students if they have any questions, and encourage them to give additional feedback in the future if they think of any questions by giving them your contact information. Tell them how helpful their time has been, and if there is a monetary or gift card incentive, give it to them and thank them again.
Learning from usability testing your website
Every usability test yields different information, depending on the people participating in the test and the stage of your website project. Here are some of the common things we’ve learned:
Are your color choices and images adding interest or distracting your users? Sometimes you’ll hear users say, “I don’t like the color” or “This is ugly.” Their comments could merely reflect their personal preferences and aesthetics. Although blue is my least favorite color, that doesn’t stop me from shopping at Amazon. It’s a preference, not a deal-breaker. However, if you have an orange background and yellow font, that would be distracting and may make the site less user-centered. Amazon’s site might not be pretty, but it is easy to use, and that’s what matters most.
Are your users getting their information quickly? They don’t want to spend a lot of time navigating a website, so they pull off toward Google once the road gets rough. The more we tested our site, the more we were able to smooth out the ride, eliminate the pain points, and get them to where they needed to go.
Are your users spending too much time reading before they get to the main points? Writing for the web should be concise. Avoid the use of library jargon and use active, descriptive language.
Is your website content fresh? Never stop usability testing. Don’t let your content get stale. Although you’ve earned a much-deserved break, your users aren’t going to wait for you to update your content before they take you off their bookmark list for more dynamic sites. Keep your users’ comments coming by putting a comment link on your homepage soliciting feedback.
Is your content crowded? Users want a site that is clean and visually appealing. White space helps your content breathe. Color, photos, and graphic elements should attract users, not distract them. Despite what I said earlier about not pandering to user preferences, you should still make your site attractive to the eye.
Usability testing can be a fun and eye-opening experience. Watching users kick the tires of your website will at times confound you, even frustrate you, but mostly it will inform you on how your users actually interact with your website. It eliminates the guesswork when making decisions about how to present the most vital information. The old phrase “build it and they will come” doesn’t apply to website design. If your user is not considered in all your decision making, then designing or redesigning a website is a waste of your time and resources.
Cockrell, Barbara J., Elaine Anderson Jayne. “How Do I Find an Article? Insights From a Web Usability Study.” The Journal of Academic Librarianship 28.3 (2002): pp. 122–132.
Ipri, Tom, Michael Yunkin, Jeanne M. Brown. “Usability as a Method for Assessing Discovery.” Information Technology and Libraries 28.4 (2009): pp. 181–183.
Krug, Steve.Don’t Make Me Think: A Common Sense Approach to Web Usability. Indianapolis: New Riders Publishing, 2000. 195 pp.
Manzari, Laura, Jeremiah Trinidad-Christensen. “User-Centered Design of a Website for Library and Information Science Students: Heuristic Evaluation and Usability Testing.” Information Technology and Libraries 25.3 (2006): pp. 164–169.
McMullen, Susan. “Usability Testing in a Library Web Site Redesign Project.” Reference Services Review 29.1 (2001): pp. 7–22.
Nielsen, Jakob.Designing Web Usability. Indianapolis: New Riders Publishing, 2000. 419 pp.
Nielsen, Jakob and Marie Tahir. “Keep Your Users in Mind.” Internet World 6.24 (2000): pp. 43–49.
United States Department of Health and Human Services.Research-Based Web Design and Usability Guidelines. Ed. Ben Shneiderman. Washington: 2006.
Wisniewski, Jeff. “New Rules of Web Design.” ONLINE 32.2 (2008): pp. 55–57.
Wisniewski, Jeff. “Tune It Up, Don’t Tear It Down: Website Redesigns.” ONLINE 32.5 (2008): pp. 55–57.