As a designer with former experience in research and education, I enjoy taking on projects involving brands whose missions include educating and encouraging users to improve themselves. Therefore, I was excited to take the opportunity to improve the Duolingo mobile app.


The goal of this project was to improve user retention by adding a feature that helps users better accomplish their goal of learning a new language.

Add a new feature



To avoid reinventing the wheel and to determine opportunities for Duolingo to stand out amongst the numerous language learning platforms, I conducted a brand comparison and feature analysis between Duolingo’s most popular competitors including Babble, Rosetta Stone, Memrise and Mondly. While Duolingo reigned in gamification, with Memrise as a close second, the app faded in comparison to Mondly and Rosetta Stone in creating a space where users can more naturally immerse into a new language. 

With this competitor analysis, I found that Duolingo can better satisfy the goals of its users and compete in the language learning market through the inclusion of a diary feature where users can interact with Duolingo’s courses and interface on a more personal level.


Because this feature would be unique to Duolingo, I also did a competitive analysis between the two most downloaded journaling apps, Reflectly and Day One. In this competitive analysis, I found Duolingo’s pricing strategy, with its free platform, champions over Reflectly. However, Reflectly’s use of quotes and AI for diary prompts and entries makes it understandable that the journaling app has been labeled as “a “personal mental health companion.”



Secondary Research

Before conducting former user research, I started secondary user research where I mined reviews and Duolingo community forum responses to understand the pains and frustrations of recent Duolingo users. Through this research I found a variety of frustrations that were later confirmed with user interviews and user surveys.

Screen Shot 2021-02-27 at 12.36.25 PM.pn

Primary Research

The results of the survey indicate opportunities to help users develop new habits as well as maximize learning by incorporating their ideal learning styles.

Users primarily learn a new language as a personal interest.

Majority of users consider themselves verbal learners where they learn better through written or verbal words.

Majority of users do not learn a new language or stop learning a new language because they are too busy.

Majority of users do not have a habit of journaling their thoughts, daily interactions, emotions, or daily activity.

Majority of users do share their thoughts, daily interactions, emotions, or daily activity in conversations with others.

Screen Shot 2021-03-01 at 7.34.31 PM.png

The goal of the user interviews was to determine the method in which the diary feature would be created. This goal was achieved by asking five users ages 14 to 35 about their free time activities. In addition, I gained more detailed reasonings as to why users abandon learning a new language. 

In these interviews, I learned about the various frustrations that language learners have with current language learning methods. Unsurprisingly and aligned with responses from the survey, the greatest barrier that users face when learning a new language is “lack of time.”
In addition to asking users about their past experience with learning a new language, I had users walk through the Duolingo app via Zoom or in-person. This allowed me to gain better insight about Duolingo’s visual design and current features. Here are some common and surprising responses:

Affinity diagram- interview.png
Affinity diagram-grouped.png

"Part of it was not just diligence on my part as needed in terms of making time for it"


"Why would I want gems? I don't care what my owl is dressed like."


"In my mind, I'm just there to learn the language…Playing a game is not my expectation when learning from a language learning app."


"The little owl is cute."

Shop Feature
"I have never done an in-app purpose and this [Shop feature] doesn't make me want to start."


To better emphasize user pains and gains, I created a Value Proposition Canvas to summarize the user goals and frustrations discovered through user research. In addition, I determined the possible value that this diary feature can bring through How-Might-We’s that include some of the following:

How might we make gamification feel less childlike for adult users?


How might we improve the learning experience for more advanced language learners?


How might we create more incentive to use the Shop feature?

Value Proposition Canvas.png

I found that this Value Proposition Canvas and the How-Might-We’s not only helped me to determine the value of this feature for users but also the value that this feature will bring to stakeholders whose goals may be to improve retention and possibly increase interactions with current Duolingo features.


Once I analyzed my data, I created a user persona and user journey map to better empathize with users. 

Meet Talkative Talia

Talia is in her late twenties with a career that involves creativity and frequent communication with others. She lives in a diverse, metropolitan area where it is not uncommon to approach individuals whose first language is not hers. Therefore, she believes learning a new language will not only improve her career but also her activity in her community. She is introverted enough to enjoy learning new skills alone but extroverted enough to enjoy practicing those skills with others and learning about other people’s cultures through verbal communication.

Although Talia likes the idea of learning a new language, she struggles finding the time to prioritize this skill amongst her other interests and obligations including work and downtime. In addition, her past experience led to frustrations that have deterred her from learning a new language altogether. Although she was excited to start the journey of learning a new language when she first downloaded Duolingo, she found the app to be childish and limiting. In addition, she found the lessons and gamification to be demotivating at times. Eventually theses frustrations would lead to Talia deleting the app.

PERSONA - Light.png
Journey map.png

MOSCOW Feature Prioritization

With the user persona and user journey map, I ideated numerous possibilities in which the diary feature can address user pain points. I then prioritized them using the MOSCOW feature prioritization method.

MOSCOW Feature Prioritization.png

Sitemap and User Flows

Once I determined the value of the diary feature and its primary features, I needed to determine exactly how the diary feature would fit in the flow of Duolingo mobile app. First, I created a sitemap based on the current pages of the Duolingo mobile app. Then, I created an updated sitemap to determine under which pages would it make the most sense to include links related to the diary feature. In the updated sitemap, I integrated the diary feature into the Duolingo daily unit lessons. In addition, users also have access to the diary feature through their profile page.

Sitemap As Is.png

Sitemap As Is

Sitemap To Be.png

Sitemap To Be

Once I determined the locations of this feature, I created user flows for the following user stories:

Beginner language learners must be able to input diary entries after completing their daily lesson.

Intermediate and advanced language learners must be able to input diary entries after completing their daily lesson.

Language learners must be able to input diary entries at their own time.

Language learners must be able to use diary entries as a screen block for their other apps.

Screen Block User Flow.png

Low Fidelity

After ideating, I created a low-fidelity prototype of the diary passcode feature and the post-lesson diary feature. In an informal usability test with two users, I found that the diary feature was desirable, but the diary passcode feature had a number of drawbacks. The first drawback was the likelihood of a user wanting to enter a diary entry early in the morning.

Low Fidelity Testing

"I can see [using diary syncing feature] if I set a goal…but I think I would use it [diary syncing feature] tied to the app not the diary. It depends on how the diary is. Super realistically…I would not use it because I use Reddit 8 o'clock in the morning but I would not want to do a diary entry 8 o'clock in the morning."

Diary Permission.png
Unit End Translate.png

Mid Fidelity Prototype

After excluding the diary passcode feature, the mid-fidelity wireframe was significantly simplified. In the usability test for the mid-fidelity wireframes, 15 users were given set tasks including: adding a diary entry after completing a unit lesson, viewing past diary entries to check progress, and adding an audio diary entry without completing a unit lesson. Using Maze to track misclicks, I analyzed areas in which the design can be improved.

Audio Diary Page



maze_screenshot_interaction (1) copy 2.p

The most difficult part of creating the high fidelity prototype was recreating parts of the Duolingo design system, including its icons and images. However, the end prototype is seamlessly integrated into the Duolingo app.

Add diary entry-audio-save.png
Diary entry-unit 1.png
Screen Shot 2021-05-14 at 11.18.21 AM.pn
Screen Shot 2021-05-14 at 11.18.30 AM.pn

There were few but noticeable challenges that I encountered in this project. The first challenges involved the implementation of this diary feature. Users with beginner proficiencies in a new language may struggle to use this feature because they do not know how to formulate complete sentences. Users with intermediate and advanced proficiency may get bored with answering repetitive journal questions. In addition, the journal feature alone will likely not lead users to develop a habit of journaling, especially when other activities and responsibilities often take priority over learning a new language. Therefore, for future iterations, testing this feature for longer periods of time may be needed.

The latter challenges came from recreating parts of Duolingo’s mobile app design system. Recreating the Duolingo icons uncovered some frustrations and limitations with Figma. However, despite the time-consuming task, the design system guided the design for the diary feature.

Regardless if Duolingo creates a diary feature in the future, the language learning app has opportunities to keep users active within its platform. For instance, incorporating AI to motivate users to learn to formulate their own sentences may create more incentive for users to keep the app. Another opportunity would be to reinstate its chatbot. If the diary feature were to go into effect, it is important that future designs include various engaging and relevant questions. I look forward to seeing future changes to the Duolingo platform.