About Koohii and the Heisig method

Don’t know if this is the right place to ask but I couldn’t think of anywhere else to ask!

Anyway, fueled by a feeling of Wanikani being a bit too slow for the pace I currently want to study at (I am on exchange in Japan and have a ton of free time!) I started looking into alternatives and found the Heisig method. I was referred to Koohii by a reddit thread, and already made an account for the website, and started my studies. Out of curiosity I did some googling and found out the website closed its forums some years ago (even the rest of the website was gone for a while?). Just wondering if it’s safe to use this website and start on the srs system without the fear of just waking up to it all being gone ;).


The Heisig method is basically what WK came from. WK is super slow, and you probably are better off finding an RTK deck (maybe even the revised decks that covers around 1200 of the most common kanji) and then just learning as you go.

Tokini Andy just made a video on this and I think he’s right.

Before the Kana addition I would have maybe said stick it out, but you gonna be stuck here forever at this rate. Still learn Kanji, find lists online and maybe put them in an Anki deck or something.

If you’re looking for something faster there are plenty of free resources out there.


I am really liking Koohii so far - it is super convenient to add the (amazing!) mnenomics made by other people - how would you recommend supplementing it, if I wanted to continue with Koohii?


If you go at full speed you’ll have 300+ reviews per day. Not sure how fast you want to go, but even if you have plenty of free time if I were in you I’d rather spend some of that time to actually go out and see Japan, and take the opportunity to interact with people and be exposed to the language.


(I did RTK via that website many years ago, and was active on the forums until they closed down.)

IIRC Fabrice got fed up with running and moderating the forums, so he closed those down, but the main site itself has kept going for years since then, so I think it is no more likely to vanish than any other free app/resource on the internet. You could also use an Anki deck and just use the website to look at the mnemonics other users suggest if you prefer.

The major difference with RTK compared to WK is that RTK only teaches kanji keywords and makes you review them keyword-to-kanji. This is particularly good if you want to write, maybe overkill for reading. You’ll need to learn vocabulary elsewhere. RTK also makes many fewer concessions to frequency of kanji, and is structured much more to start with simple shapes, introduce kanji with the same radicals in a coherent group, and so on.

My personal experience with RTK was that I found it didn’t really tie in very well with the rest of my Japanese study. I ended up able to write 2000 kanji based on a keyword prompt, but when I saw the kanji I tended to think of the Japanese words, not the English keywords, and so it didn’t really tie in to helping to remember “how do you write 勉強” type difficulties, which is the problem I was really hoping to solve at the time. (I later experimented with Japanese keywords, which didn’t really work out either. But mostly I no longer had a need to handwrite Japanese, so my motivation to try to work on it has gone.)

If you have spare study time, my personal opinion is that a more effective use of it would likely be to study grammar, more vocab, do more reading, etc, rather than study kanji more intensely.

(Also, I noticed in a blog post on the site that in the sixth edition Heisig finally swapped the “town” and “village” keywords everybody always complains about. Better late than never…)


There are two issues with studying Kanji 1) How do I study each Kanji?, and 2) Which order should a Kanji be studied, ordered by radical / Kanji complexity?

I went through a few hundreds of Heisig, all of Kangxi (in Anki), and the whole track of WaniKani. However, I never really did the Koohii web app. I think that 1) Studying each Kanji should have supporting vocabularies, especially early on. This is to avoid too much English, and studying no real Japanese. 2) The order of Kanji is less important later on, but radicals remain helpful.

After a few ten levels, it might be ok to drop WaniKani and do Kanji self-study, as the ordering become less important. However, if you like the SRS part of WaniKani, it might be good to continue, with compromise, as you can’t complain much if WaniKani misses some vocabularies for a Kanji, or has too many vocabularies for a Kanji reading. WaniKani may also miss a Kanji meaning sometimes.

That is, if you want to do Kanji-based vocabulary studies.

1 Like

Thank you for the response! A bit of an unrelated question, but I’ve been wondering why the SRS system on Koohii has such long intervals; it starts with 3 days! As far as I remember, after studying the card, Wanikani has you review it only a couple hours (4-6?) after. With a difference this big, is it even realistic to be able to complete the RTK kanji without changing the intervals in Koohii?

1 Like

On hundred percent! Let WaniKani be slow. Learn 10 words in the morning and go out in Japan and make it your job to use them in conversation and you will never forget them. Don’t look for another study tool. Pull out you genki read a chapter in the morning and go practice. Even if you just find the time to speak it out loud to yourself. So much faster than any study method design for people learning in a place no one speaks Japanese.

No study method is going to make you fluent before you get out there and practice.

Because if you want a normal life, the smaller intervals in WK are largely meaningless. Those intervals specifically are insignificant to the point if you don’t remember it the next day or in this case, 3 days, you’re better off just relearning it. I think 3 days is a little long, but I always found the 2,4 hour intervals redundant.

1 Like

That’s just like how the creator of Anki thinks. Let people rest and do the job once a day.

Personally, I find hours’ interval helpful for things I don’t quite remember, but only once is enough, so no need for 8-hour interval. Also I reduce to 1-hour, in Anki. Maybe I took influence from WaniKani, so then, I made a thread to complain about leech prevention in SRS, Anki, recently.


I think the idea with the Koohii intervals is that the system is for testing your recall, not for making you learn it in the first place. Heisig places a lot of emphasis on coming up with strong images and mnemonics, and avoiding the use of pure repetition to remember the kanji. I think if you didn’t remember the kanji after a three day gap then the proponents of the default scaling intervals would suggest you didn’t spend enough time on studying it to start with. Certainly intervals of only a few hours feel more like part of “learning” than “testing recall”.

1 Like

So I started doing kanji when my Japanese level was much, much lower than what it is today, after 14 years of study and over 10 years in Japan, where my work environment is in Japanese. My wife is monolingual Japanese, too. Naturally, I got better.

I’m not writing this to show off or in an attempt to pose as a “pro” (whatever that even means for language learning). I’m just giving you an overview of my development over the years, and what I believe that means for what I’m going to say.

When I was at a crude, intermediate level, able to converse in daily life, but still always at a loss for words, I was at my height of studying kanji. You can see, my WK account goes way back. At the time, WK seemed to be the thing to do. Learning new vocab while memorising the kanji seemed like a complementary activity worth pursuing. And I believe it was. There are of course drawbacks like vocab that’s obscure or weird, or words nobody writes in kanji, but overall, it was helpful.

I took a break. Forgot some kanji, somehow picked up a few by immersion, and when I tried WK again, none of the vocab items were of use anymore. My Japanese was already at a point where I could tell when something was shoehorned in, and the glacial pace meant I’d have to spend months repeating stuff I already knew, because I had forgotten the odd kanji I’d already burnt. Needless to say, I stopped after a few months, again.

Years after that, I returned with the conviction to do it all, anyway. By then, I had already reached fluency with the rare blackout when a word wouldn’t come to mind, at about the rate it happens in my mother tongue. Now things felt even worse, because I had picked up a LOT of kanji otherwise, and I evaluated WK’s worth for someone who could already get through daily life with almost no problem (there was always the odd kanji I couldn’t read, and it drove me crazy).

Some more years have passed, and what I believe I can now say with some confidence is, the better your Japanese, the better you’re off with RTK. You won’t see anything new, anyway, just stick a character on to existing knowledge.

If you’re really new, study the language itself while doing kanji here. But when it stops working, which will happen at some point, don’t sweat it. Do something else instead, pick up some characters through reading, and if you’re at a level where your main concern is that you mix up kanji in isolation, while being able to read them fine in text – just rush through RTK in a week or two and be done with it. Review from time to time, but don’t bother with SRS. Seeing a freshly learned kanji in a book helps much better with recall.

In hindsight, what WK never got quite right was lookalikes. Not even with plugins. Since RTK makes you write, this will never be a problem.