From Goodreads to Better-reads
UX research to understand users' feelings and behaviors towards Goodreads, address the problems, and validate the solutions.
The problem
Hypothesis
Objectives
Book lover users like using Goodreads, but the app displays severe constraints that decrease the app usability and users satisfaction.
To research and understand users' perspective, create a possible solution for a significant problem and validate it.
I hypothesize that by gathering enough data on users' emotions and behaviors and empatizing with them, we can gain valuable insights to create a design that addresses the problems found.
Methods
-
Gathering users opinions: foundational research
-
Assess app problems: heuristic avaluation
-
Create empathy: user persona and validate it with users
-
Emphatizing users and opinions about current products: 1-on-1 interviews
-
Designing prototypes: wireframing and prototyping
-
Testing the prototype: usability studies
-
Gathering users opinions on the prototypes and compare them: Focus group and A/B testing
-
Validating the solution offered: System Usbility Scale
Empathizing and defining
Foundational research
The start point for my research started by checking the reviews of Goodreads in the PlayStore, and these are some representative comments I found:
"It would be great if it were easier to add friends, with a QR code to share on Instagram, for example."
.png){kind=small}
"Meeting place, refuge, source of information, and memory suitcase. Fantastic app :)"
.png){kind=small}
"Fantastic app for reading enthusiasts and for those who want to become one"
"The rating is only for the app, which lacks most of the options that are available on the website... While the website has many features for readers, the app is useful for very little."
.png){kind=small}
"I really like the idea, although they could make it a bit more intuitive. I've been using the app for three days and still can't find how to search for someone by name to start following them."
These reviews show that Goodreads has a loyal community of readers, but on the other side, it highlights some problems that reduce users' satisfaction.
Heuristic Evaluation
I decided to conduct a heuristic evaluation using Jacob Nielsen's 10 usability heuristics to identify and assess design problems in the UI of Goodreads.
This research phase revealed various issues from different heuristic types, as shown in the table above. The details and recommendations for each problem are shown below.
H1: Visibility of system status
H2: Match between system and the real world
H3: User control and freedom
H4: Consistency and standards
H5: Error prevention
H6: Recognition rather than recall
H7: Flexibility and efficiency of use
H8: Aesthetic and minimalist design
H9: Help Users Recognize, Diagnose, and Recover from Errors
H10: Help and Documentation
User persona and its validation
I created a user persona with the aid of artificial intelligence (AI) to improve the understanding and empathy towards users' feelings and behaviors.
.png)
Once I created the user persona, I wanted to validate it with real users. The objective was to assess whether the persona created was a good representation of a real user and to which degree real users identify with it.
After presenting Bella to 2 users, I asked them different questions to understand their thoughts and feelings towards this persona.
The main conclusions obtained were:
Both users agreed on that this persona was a feasible user of Goodreads.
They felt represented by it with a mean of 4.5 out of 5.
The aspect that they both felt less represented by was the desire to use this app to participate in online communities.
User interviews
I conducted interviews to 4 users ranging from 28 to 35 years old to gather information about their feelings when using the Goodreads app.
A Likert scale showed that users feel neutral when using the app. This outcome surprised me because various comments during interviews showed that users preferred using Goodreads among other similar apps. However, I also revealed severe pain points that significantly impair users' satisfaction and app usability.
.png)
The question regarding the features that users liked the least generated multiple answers, so I decided to use Miro's brainwriting tool to visually organize them. Once I introduced the main ideas gathered from each user, I used tags to categorize each idea into a group. Then, I grouped them according to these tags.
Comments for each participant
Comments organized by clusters
This method showed that several comments were addressing pain points related to the apps' adding friends feature. Nonetheless, severe pain points were also found regarding giveaways in Spain, the app's rating system, and the lack of personalized recommendations.
Ideating and prototyping
Digital wireframing and prototyping
After researching user needs and the app constraints, several severe issues were found. I chose one that caused important dissatisfaction with users: finding and adding friends. Thus, I created a potential prototype to solve the problems with this issue and improve its usability.
I created two low-fi prototypes: one prototype of the Goodreads app and one in-house built prototype that potentially solved the issue found regarding adding friends. I decided to avoid using the actual app to elude users' psychological vias when using a curated app. Although the actual app can display some annoying inconveniences, comparing a final product with a lo-fi prototype could decrease the chances of the new prototype succeeding.
Testing
Usability study
I conducted two unmonitored usability studies of the prototypes using Maze to test time on task, difficulty to perform task, and the success rate. These were the results:
Goodreads
prototype
Success rate
100%
Time on task
42s
Level of difficulty
(1-Very difficult,
7-Very easy)
1.7
In-house built prototype
100%
24s
6.5
This quantitative data demonstrates that while both prototypes obtain maximum success rates, time on task and the perceived difficulty level showcase that the new prototype is easier to use and users spend less time performing the task.
Focus group for A/B testing
After conducting the usability studies, users were grouped in a room in which they sat in a chair surrounding a table. I displayed both prototypes step by step. Afterward, I asked them to debate about the differences, features they liked and which one was easier to use.
The general conclusion gathered was:
One prototype is loyal to user goals
Users unanimously agreed that the in-house built prototype was loyal to the goals of the user, while the other one was confusing.
All users arrived at the same conclusion claiming that the old prototype was not intuitive and felt discouragement by only having the option to connect with other users by exiting the app.
Note that users called prototypes A and B, and I did not know which was the one based on the actual app and which was the one I modified.
Surprisingly, the lively debate resulted in other comments and opinions that could further improve the app's usability, such as prioritizing the Friends section in the main menu or being able to find friends according to the books read or nearby users.
System usability scale
Having collected qualitative and quantitative data from both prototypes, I decided to validate these results by running the System Usability Scale (SUS), created by John Booke in 1986. This easy-to-use scale measures the efficacy, efficiency, and satisfaction of a system using 10 statements.
Its regular use has shown that webpages and apps score a mean of 68, and values underneath 68 showcase that some aspects should be improved.
I ran the SUS on my cohort of users after they completed the usability study of both prototypes. The Goodreads prototype obtained a mean of 15 while the in-house built prototype obtained a mean of 87.
These results align with the previous ones, corroborating that simple changes in the prototype have a real impact on usability and improve user satisfaction with this item.
Research outcomes
1
Foundational research, heuristic evaluation, user persona, and user interviews revealed two main conclusions: book lovers enjoy using Goodreads, but the app lacks some fundamental features and displays some complicated interfaces that impair users' satisfaction with it.
2
Research obtained numerous specific constraints. User comments and feelings directed my case study by choosing one problem that significantly affected users: the UI of adding friends.
3
I designed a new interface and tested it to assess whether this new prototype solved the problem and guaranteed usability.
4
The usability study using Maze, the focus group, the A/B testing and the System Usability Scale concluded that the new design improved the UI when adding friends in Goodreads.
Final statement
Research demonstrated that Goodreads is the preferred app for book lovers, but the presence of several constraints plummets users' satisfaction. This case study reveals that small user-centered and guided iterations can have a big impact on usability, further assuring the use of this amazing product.