Why I Stopped Using Anki (and What I Built Instead)

Why I loved Anki at the start

I discovered Anki the way most people do: someone in a language forum swore it was the reason they passed JLPT N2, and I spent an afternoon reading about the forgetting curve and felt like I had found a cheat code. Anki is genuinely powerful. The spaced repetition algorithm is battle-tested, the data format is open, and the desktop app is free. If you use it consistently, it works. That is not in dispute.

In the first three months I was disciplined. I reviewed every morning before work. My retention hovered around 85 percent, which felt good. I was learning words I would not have otherwise touched, and when I came across them in reading they felt like old friends. That feedback loop is real and I do not want to pretend it is not. The first year with Anki was the most structured vocabulary learning I had ever done.

I also made the mistake most enthusiastic Anki users make: I spent too much time optimizing the tool instead of using it. New card templates. Custom fonts. A plugin for pitch accent, a plugin for dictionary lookups, a plugin that showed example sentences from a corpus. Every weekend I was tweaking something instead of reviewing. Anki rewards tinkerers. That is part of its charm and part of its problem.

The slow drift away

By year two, the card count was somewhere above 3,000. My daily review queue had grown to 150-200 cards on days I kept up, and if I missed two or three days it became a wall. Anki does not forgive absence. The reviews pile up mathematically, and opening the app to find 400 due cards feels more like debt than learning.

The subtler problem was that creating cards had become the bottleneck. To add a new word properly, I needed to: look up the word in a dictionary, copy the reading, find an example sentence, optionally add an audio clip, check if the card would conflict with something already in my deck. A thorough card took five minutes. I was spending more time building the tool than using it. The irony was not lost on me, but I kept telling myself this upfront investment would pay off.

Around month eighteen, I noticed I was opening Anki less. Not dramatically, not in a single moment of decision. Just: some mornings I chose not to. Then some weeks. The app sat on my phone, the icon unchanged, a small number badge climbing. At its peak, I had over 600 cards due. I did not feel guilty in the old way. I felt finished with the whole project.

The tipping point

The moment I remember clearly was a Sunday morning. I had about forty minutes before a commitment. I sat down to do Japanese. I opened Anki. There were 312 reviews due. I looked at that number, closed the app, and opened YouTube instead. Not because I did not want to learn Japanese. Because 312 felt like punishment for having a life that week.

That day I started thinking about what I actually wanted from a flashcard app. I wanted to open it and immediately be in a session, not managing a backlog. I wanted cards that already had audio, so I was not guessing at pronunciation. I wanted the algorithm to handle itself, not surface options I had to understand before I could use them. I wanted the aesthetic to feel like something I chose rather than a tool built for power users.

Anki was built for people who want maximum control and are willing to pay for it with their time. That used to be me. It stopped being me somewhere in year two. The app had not changed. I had.

What Inku does differently

I built Inku around one constraint: it has to be something I actually open. That sounds trivial. It is not. The hardest problem in language-learning software is not the algorithm, it is getting the learner to show up. Every decision in Inku flows from that.

The decks are pre-built. Hiragana, katakana, 200 core N5 words, 65 N4 words, four phrase packs. You do not build anything. You open the app and the cards are there. The audio is pre-recorded by a native voice actor, not synthesized on the fly, not a text-to-speech robot. Every card you flip, you hear a real human voice say the word the way a real human would say it.

The session model is different too. Rather than surfacing 150 due cards at once, Inku shows you a manageable chunk and ends the session cleanly. There is no accumulating debt. If you miss a day, the algorithm adjusts. You do not come back to a wall. You come back to a session.

The aesthetic matters more than I expected. Sumi ink on washi paper. Calm type. No streaks, no badges, no points. Studying Japanese should feel like a practice, not a game you are losing. The difference in how I feel opening Inku versus how I felt opening Anki at the end is not small. One felt like a chore I had let slide. The other feels like something I am choosing.

Can you use both?

Yes. They serve different learners and different stages. If you have a well-maintained Anki deck with 2,000 mature cards and you are reviewing consistently, do not stop. Mature cards are valuable and migration costs are real. Anki's power-user features, custom note types, filtered decks, add-ons, are genuinely useful if you want to go deep into an unconventional study method.

Where Inku fits better: you are starting Japanese from scratch and do not want to spend two weeks configuring before you learn a single word. You tried Anki, burned out, and need a reset. You want native-quality audio without hunting for audio packs. You want an app that handles the curriculum so you can focus on the language.

Anki is a power tool. Some learners need power tools. I stopped being one of them. What I needed was a tool that got out of the way and let me spend my forty minutes actually learning Japanese instead of managing a vocabulary database.

Common questions

Related reading

• Inku vs Anki: full comparison
• Best Anki alternatives for Japanese
• How spaced repetition works

Try Inku, free for 7 days, or read the full roadmap.