298 Beyond the Usability lab or you want to better understand abandonment in an ecommerce transaction, you will need to consider different vendor solutions. In making this decision, it's helpful to think about what type of Web site issues you want to solve. You can use an out-of-the-box solution, such as Loop11, WebEffective, RelevantView, or UserZoom (see Chapter 7). These solutions will give you tremendous flexibil- ity in how you design the study, and will also give you access to a wide range of data, most notably clickstream data. Alternatively, if you have very simple needs, perhaps you should consider building your own online study using a standard survey tool (see Chapter 8). How much support do you need? If you already have experience in setting up online surveys or if you have a very simple study, you may need less assistance. This might be your first foray into online usability or you simply don't have the time. Regardless, you need to deter- mine your own role relative to the tool you chose. Luckily, the options run the gambit, from having a vendor design and carry out the study with your guidance to doing it all yourself. Obviously, your budget will also play a key role in this decision. Chapter 2 discusses some of the issues you will need to consider when deciding how much support you need. 10.2 thinK oUtSide of the (WeB) Box Most researchers think of online usability testing occurring with Web sites, particularly Web sites that are live or in production. Likewise, most vendor solu- tions are geared toward evaluating Web sites, with an emphasis on collecting clickstream and other behavioral data about the user's experience. However, this does not have to be the case. Although most of our discussion through- out this book has centered on Web sites, we argue that online usability test- ing can and should go beyond the Web. For example, online usability testing could easily be used to evaluate products such as mobile devices, desktop software, video games, consumer electronics, and voice response systems. The basic premise of an online usability test using something other than a Web site involves the participant switching between the study (a Web browser) and the product they are interacting with. For example, a participant is given a task via a Web survey, they perform a series of actions with the product, and, once completed, they go back to the survey, answering a series of questions. Depending on how the survey is implemented, you can capture task success, task time, ease and confidence ratings, satisfaction, and any other self-reported metric. Online usability tests are often far less expensive than testing Web sites in a labo- ratory setting. You can easily use a standard survey tool to capture most data you would be interested in. Of course, you can't collect data about actual use, but with a thoughtful set of questions, you can infer behavior quite well. You can also run a follow-up lab study to gain a better sense of actual behavior with the