-4.gif)
This time, with my 4-person team, we set the following research goals:
• testing movement and navigation on the website• whether the information architecture is clear to users,
• legibility of labels,
• checking whether the website allows you to complete the tasks we have defined for usability testing,
• overall rating of the website.
In order to prepare for the research as best as possible, we first answered three questions: what we know, what we are not sure of and what we do not know, and then we selected the aspects that we can examine during the research session.
We identified the problems that could potentially threaten the course of the study and answered how we can prepare for/prevent them.
We recruited users for the study thanks to a previously prepared and conducted selection questionnaire. For usability tests, we recruited a sample of 5 people (2 women and 3 men) who have or had a pet and are familiar with using websites with pet products.
• 4/5 people currently have a pet,
• 1 person recently lost,
• 4/5 people shop online several times a month,
• 4/5 people aged 18-26,
• 1 person aged 27-35.
The screener also allowed us to exclude people who are industry-related to UX/research or have recently participated in a similar study, or who do not have a pet, have not had one, or do not plan to have one.
We used task-based usability tests, i.e. testing the interface by the user according to a previously developed task scenario.
Three people took part in each of our studies: a moderator who supervised and conducted the entire study, an observer who took notes and observed users' reactions and behavior on the website, and, of course, the respondent.
Due to the pandemic situation, we conducted research sessions on Zoom and Messenger. We asked them to share their screen, visit the website naszezoo.pl, and to think aloud while performing the tasks (commenting and saying aloud what they were doing).
5 respondents tested the naszezoo.pl website by completing the tasks we prepared. The study scenario included:
1. Introductory questions to break the ice and get to know the respondents and their behavior and preferences related to purchasing pet products
2. Zero task, aimed at initial familiarisation with the website and getting the first impression of the naszezoo.pl website + 3 tasks, each divided into sub-points, placed in the appropriate context:
a. Finding specific pet products
b. Finding specific categories and firstly answering the question of what the user expects from it, and secondly - verifying their assumptions
c. Finding information on the website related to additional services of the naszezoo.pl store
3. Summary questions in which we wanted to learn opinions about the experience of the website and to find out opinions about the use of other similar pet food websites
4. Additionally, when a topic was particularly important to our respondents or when we saw room to expand the topic, we asked questions (e.g. product page, advantages/disadvantages of similar pages with pet products).
We analysed the data in a spreadsheet using a categorisation key (positive situations, negative situations, goals/expectations, similar products [websites, applications] used by respondents, quotes).
We also determined the errors and their severity and proposed recommendations for them.
We also presented other comments regarding the website resulting from the research.
We decided to present the results in response to the goals set at the beginning, going through individual stages of the study (initial questions, tasks
and summary questions) in the form of quotes.
"Why is there so much of this?"
"The 90s-like interface makes me not trust the website as much as any other, modern one"
"It'll probably be downstairs. Oh, but no."
"At least there are pictures of horses, so I know I'm in the right place."
To sum up, although there were a few bad moments, most respondents found that they were able to navigate the website and search for information quite efficiently. Most of the tasks have been completed. The colors of the website are friendly, but some aspects disturb its intuitiveness, friendliness and readability.
We used a three-level scale, including a critical error (preventing the user from reaching a specific place on the page, contributing to frustration. It requires a change as a priority), a major error (not preventing the user from reaching a specific place on the page, but causing frustration. It requires change as a second priority). and a small error (a situation when the user is not fully satisfied with some element on the page. To be changed last.).
Finally, we identified one critical error, three large errors and three small errors.
• It is worth making sure that everything works before we start the test to eliminate all possible technical problems.
• It's good to have additional 'emergency' people willing to participate in a research.
• Being an observer is more demanding than I thought - sometimes the desire to give suggestions is difficult manageable.
• Being careful not to add or confirm that someone did a task well.
• It is not easy to find a golden mean between speaking casually and reading from a piece of paper during the introduction so as not to bore the examinee.
-5.gif)
