Using eye tracking, facial expression analysis (FEA), and galvanic skin response (GSR) software, a team of 3 and I ran a study to determine the effectiveness of the design of 3 different existing websites and one video advertisement.
We designed 4 tasks for each website, each with a specific purpose, with the hopes of gaining some insightful information on what works for the websites, as well as what can be improved.
Prior to running this study, we ran tests on J. Crew, Banana Republic, and Cup of Jo to get a feel for how to run a study and analyze the data.
Chewy and Petco
Our 4 participants looked at a page for this Blue Buffalo dog food on both Chewy and Petco, two popular pet supply stores. The participants were asked to complete the following tasks on each site:
|  |
To account for any potential bias, some participants saw the Petco website first, while others saw the Chewy site first.
The first two tasks, regarding the dog's age and weight, were completed by looking in the same area on the page. They yielded similar results in which we found one finding - demonstrated below.
Chewy | Petco |
These tasks required participants to look in this important information section. Both tasks yielded similar results, in which the Petco page took twice as long to complete - regardless of which site the participant saw first.
The Areas of Interest (AOIs) for the two tasks are pictured below, demonstrating the location of the age task on top, the weight task on the bottom, Chewy on the left, and Petco on the right.
Additionally, we received data on the Time to First Fixation (TTFF) for the task regarding the dog's weight. The data confirms how Petco took participants twice as long to find. The age task has evidence that 3 of our 4 participants took less than 15 seconds to find the AOI on Chewy, while 3 of our 4 participants took more than 35 seconds to find the comparable AOI on Petco.
All that being said, we determined that Petco could create a more user-friendly site by moving the important information section up closer to the main information you get when you first open the page.
Serious Eats
We ran the same 4 participants through tasks on Serious Eats, a popular recipe site, looking into the purpose and organization of the home page. Participants were asked to:
Find a "World Cuisines" article.
Find a "Fresh Recipe" to try.
Find the "Personal Essays" section and click on an article.
Find the "About Us" section.
Overall, we found that Serious Eats has a problem with visual hierarchy and clear messages which consistently created false confidence within the participants.
All 4 participants confidently clicked on an article within 35 seconds, however only one had correctly identified the section we were looking for, the "World Cuisines" section, pictured below.
Instead, 3 of the four participants found an article or recipe like the one pictured to the right, which was a recipe from another country. While they did technically find a "world cuisine", they did not find the section we were looking for. Serious Eats could improve its user experience by organizing the site more. The current website causes unnecessary confusion, and none of the participants were quite sure what the site was about prior to completing tasks. |  |
 | While practicing running studies and interpreting data, we used Cup of Jo which had the organization pictured to the left. Participants in that practice found the line that says "STYLE" to be useful in navigating the site. Using the comparable line on the Serious Eats screenshot above, it could instead say something along the lines of "THAI | WORLD CUISINES" to be more useful and instead of creating false confidence, make it more evident what the user is looking at and increase the site's organization. |
Advertisement
Finally, we connected our participants to a GSR software and ran a FEA program to determine any emotional reactions our participants may have felt. We chose to evaluate the effectiveness of the following Toys 'R' Us Star Wars commercial.
Using the data collected, we determined the ad was effective. Our four participants, despite being outside of the ad's target audience, had emotional reactions at the times the audience should have a reaction. This can be seen in our GSR and FEA aggregate data, pictured below.
The participant's GSR peaks occur right when the daughter leaves at 0:27, and when we see the dad excited to see his daughter playing with the lightsaber at 0:43. Additionally, participants smiled when the daughter smiled after the "I am your father" line at 0:50.
