Duolingo: A User Research Case Study

Kübra Çiçek
6 min readAug 18, 2020

Using language learning apps is a convenient and easy way to learn and practice a language. Duolingo is one of the most popular language learning apps and has more than 300 million users worldwide. By applying gamification techniques, Duolingo aims to make language learning fun and easy.

The app tries to motivate users by reminding them their daily goals, streaks, and rewarding them for progress. There are also leaderboards in which users are placed according to their XPs (experience points).

Research Objectives

My research goals for this project were to discover users’ needs, motivations, goals, positive and negative experiences related to Duolingo.

Disclaimer: I am not affiliated with Duolingo. I only did this research to get experience in user research and create a personal portfolio project.

Participants

I recruited 10 participants for this study (8 female, 2 male) through social media (Instagram, Facebook) and Slack. Participants consisted of 6 users that are actively using the app and 4 old users who used the app in the past 12 months but are not using it anymore. The reason I included old users was to learn about their past experiences with the app and why they stopped using it. The mean age of participants was 29. All participants’ native language was Turkish.

Research Methods

First, I created a survey using JotForm to ask users some general questions, such as demographic information, the users’ motivations for using language learning apps, which languages they learn/learned using Duolingo, how often and how long they use/used Duolingo, and their overall experience with Duolingo.

Then, I interviewed the participants for 15–30 minutes using Zoom or Skype. In the interviews, I asked more specific questions such as why they started using Duolingo, in what ways the app was motivating or demotivating to learn the language, what they like and don’t like about the app, their thoughts on the app’s navigation, interface design, specific features (e.g. streaks, virtual money).

Analysis

For analyzing survey results, I used JotForm’s Report Builder feature. Report Builder gave me the results and automatically created graphs according to results.

For analyzing user interviews, I first transcribed recordings of the interviews. Then I created codes for participants’ statements, and marked these codes with colors for each participant in Excel. This method helped me to see the recurring statements among participants.

After coding analysis, I created affinity mapping in Miro. I placed the repeated statements into emerging themes: positive experiences, pain points, behaviors, user wishes.

Results

General findings

  • User needs and motivations: Survey results showed that participants use language learning apps for various reasons. Top three reasons were for self-improvement, out of interest in the language, and to refresh their language skills.
Created by JotForm Report Builder.
  • Duolingo was the most popular language learning app among the participants.
Created by JotForm Report Builder.
  • Language choices: Majority of participants were speaking English as a foreign language. Participants reported that they were using Duolingo to learn English, German, Spanish, French, Russian, Italian, Dutch, Latin.
Created by JotForm Report Builder.
  • Usage frequency: Participants who are actively using the app provided various responses about their usage frequency.
Created by JotForm Report Builder.
  • Overall experience: Larger number of participants rated their overall experience with Duolingo as good and average.
Created by JotForm Report Builder.
  • In general, participants think that Duolingo app is fun, easy to use and understand. They like that the app is free.

Personas

There was no difference between two genders or between active and old users in interview analysis, therefore I did not include them in results separately. However, based on user goals and behaviors, I defined two personas.

Pain Points

  • Decrease in motivation: Users find the app’s encouragements, notifications, their daily goals and streaks were motivating at first, but these features slowly lose their effects and become not as motivating as before. As a result, users start to use the app less frequently.
  • Long and repetitive exercises: Some exercises can be too long and repetitive. Usually, the same sentences and words come up in exercises. Users feel bored and frustrated.

“I get bored while doing the exercises because the example sentences are always the same.”

  • Less emphasis in speaking: Users are satisfied with their improvements in listening, reading, writing skills, but they wish there could be more speaking options.

“I wish I could speak with other users after a certain level. That would motivate me to go on the next levels.”

“I wish there were more pronunciation exercises.”

  • Limited options in the shop: Virtual money (lingot or gems) that is earned by completing levels or challenges can be used to replenish hearts, freeze streak, buy outfits for the mascot Duo, get bonus lessons etc. However, users either do not spend their virtual money at all or they think the options in the shop are not valuable or satisfying to them.

“I don’t think virtual money is being used effectively in the app. I don’t want to buy outfits for Duo.”

Shop options

Insights and Recommendations

  • Users find some exercises long, repetitive, and get bored, because they expect the app to be entertaining. Long exercises can be shortened and there could be more variety in words and sentences.
  • Users want to improve their speaking skills, because they want to practice the language in real life. More pronunciation exercises could be added. Additionally, speaking to other users might be a feature in the app. Users could speak to other users after they reach a certain level in the language they are learning.
  • Users rarely buy items from the shop either because they are not aware of the things that they could buy with virtual money or they think the options in the shop are not valuable to them. The shop options could be promoted to users more frequently, and more motivating items could be offered to users, such as more bonus lessons or word quizzes.
  • When the completed lessons get ‘shattered’ after some time, this indicates that it is time to revisit those lessons. Because the system leaves this up to users, participants either are not practicing old lessons or they are not doing it regularly. Making this mandatory and regular function might be better for users to strengthen their past knowledge.
Eggs (lessons) get shattered some time after completing them.
  • Because people have different needs and goals when they use language apps, lessons and levels could be customized according to different personas, such as traveler, student, business etc.
  • Duolingo offers only three language courses for Turkish speakers: English, Russian, German. In this study, participants who could already speak English preferred to learn another language in the English version. Nonetheless, more language courses could be added for people who would prefer to learn them in Turkish.

Thank you for reading!

--

--