It turns out the adorable aggressive owl (chouette π«π·) we all love is a shameless, pay-to-win capitalistic bawd (Schlampe π©πͺ). Duolingo updated their iconic language learning app earlier this year after some poorly executed A/B testing, an order from above to increase revenue no doubt, and almost no creative thinking.
The app changed from a language learning structure of a preset path that offered three topics and lessons to choose from at any given time with a stronger emphasis on study materials (e.g. a few paragraphs to pages of text to read to learn some language patterns that might not be obvious while completing the mini-game lessons) to a preset path that only ever offers one choice, deemphasized reading as a mode of study, and worst of all, introduced an ableist, stressful, speed-review mode where one tries to complete some number of exercises within 1 minute 45 seconds. The task is nearly impossible in my primary course, Japanese (ζ₯ζ¬θͺ π―π΅) and I’ve heard is similarly difficult in other languages. The app team is under the impression that people will pay money for Timer Boosts to receive more time to complete the speed round and receive XP points that promote a user higher in the league they compete with strangers from around the world. No doubt they are milking some grandpa (grand pΓ¨re π«π·) dry somewhere, but they changed the app from a slight pay-to-win model to a blatant pay-to-win model β XP is earned fastest in these speed rounds rather than completing lessons. If Duolingo wanted to increase revenue, they should have simply added more value to the app. Their redesign added no additional value for users. That’s the main takeaway here. Buggy speed runs don’t add value, and they’ve discouraged me as a student (ε¦ηπ―π΅).
Duolingo’s product monetization team also removed the “Podcast” feature, the only way to use the app while driving in the car (γγγΎ π―π΅), as well as the only feature that encouraged back and forth conversation. I was A/B tested right out of the podcast feature, twice.
And finally, they regressed how quickly their audio playback starts when tapping the word-blocks. Presumably, some developer added a network analytics call prior to playing audio, or something similarly pea-brained (strong opinions, weakly held β I’d be happy to learn this is some other more interesting bug).
These changes made me upset (furieux π«π·) enough that I went looking for other language resources. I found two: Memrise and Innovative. Both have their relative strengths and weaknesses. However, once I used each of these apps, I realized regardless of the Duolingo update, my language learning methodology had been missing some components if I plan to be able to use intermediate traveler’s Japanese when I visit (besuchen π©πͺ) the country in a year or two.
Innovative has fantastic (bonne π«π·) learning materials but has some information architecture house keeping to do. The app itself is called Innovative, but it hosts a login to many individual language learning services that follow the naming convention <Language>Pod101. E.g. JapanesePod101, FrenchPod101. They’ve published their materials partially around the web. You can find parts of courses on YouTube and Podcasts. The premium tier offers features like word banks, a way to save and categorize words and phrases, as well as automated assessments and offline learning downloads. Their premium plus tier offers access to conversations with real live language learning teachers and hand-graded assessments. Duolingo is blissfully unaware of these options as they crank up the difficulty on Match Madness β the most ableist of their offerings. I focus on the ableist theme again out of frustration. My dad, proficient in travelers French and Spanish, wanted to use Duolingo to participate in the fun my family was having using it and immediately we realized their hit targets were needlessly small and he couldn’t interact with the phone fast enough to complete their timed challenges. Big accessibility miss from them.
After all this, I’m still not dropping Duolingo. Furthermore, even if they hadn’t made these changes, I still recommend anyone serious about learning a language do more outside of the owl. Innovative and Memrise both emphasize speaking and conversational usage more than Duolingo. The moment I started using the apps, I realized those neural pathways had not been formed by Duolingo, though my vocabulary and spelling were both great. Partially, what keeps me in Duolingo is the fact that I’m so close to a year streak and the pipe dream that they wizen up and fix a lot of what’s wrong with their app. Duolingo also is designed such that you can do a quick lesson on the go, or quick hiragana review. It does help tremendously with vocabulary. The community around the app is extremely valuable. Aside from the vanity, the streak factor does keep me continuously learning the language.
- For an interesting read in how these kinds of decisions happen, see this article: apple.news/AmdU3abla…
- A/B testing is a technique companies use where they try two or more different user experiences at once, try to perform some often flawed statistical analysis on which has more engagement, and then use only the best performing option moving forward. Ever had that conversation with your friend “I don’t see that button in the app on my phone!”