Firstly, few things I need to explain about how I use the app:
I have noticed that people seem to have about 100-200 reviews per day. I am at that point as well, but I wouldn’t mind having twice that much. I love reviews. I am wondering if I should take more lessons per day. I currently do 15 as recommended. I push it up to 20 when there are just few more symbols, for example today I had 39 kanji unlocked (all remaining at my level) and I decided to take half (20) to finish it in two days.
My current success rate during reviews is around 95%, sometimes I get 100%, sometimes 90%, rarely 80%. From what I have read here, that is a bit high.
I know some people have finished WK in around one year, so they must have taken more lessons per day. I am thinking about doing two learning sessions, one in the morning, and one right after the first reviews from morning session.
So, am I being inefficient? Should I take more lessons until the rate drops to 80-90%? What is ideal percentage for efficient learning? Or am I looking at it all wrong and taking more lessons is bad idea no matter the percentage?
You can do as many lessons as you would like - especially in the beginning if you feel they are easy.
For your progression pace, what really matters is the speed at which you do your radicals (because they unlock the second half of the kanjis) and then the speed at which you do that second half.
When you get to higher levels, the number of reviews will go up markedly as master/enlightened items come back for review AND those that leech will be again part of your review pile. So depending on your success rate longer-term, your reviews may be 2-3x what you have now. This is where (I think) people tend to burn out, and my only advice is: always complete your reviews everyday and cut on your lessons if you have too much.
No, I am at 100-120, I said that I have seen others mention numbers between 100 and 200. Also, I am level 5, I guess it is because I got it on phone and not here.
I am currently encountering reviews that move things to master level, so I assume I will later meet those same reviews two more times, increasing my workload by 2x lessons per day. That would be around 150-200.
It’s important to remember that the “efficiency” (or lack thereof) at which kanji (and other study items) are committed to memory will depend on the individual. So there is no correct rate.
It doesn’t hurt to push yourself harder if you feel like you can do more, and it’s equally valid to back off if it’s too much, too soon (though I’d advocate for being more aggressive with cutbacks than ramping up) – especially depending on what else is on your plate outside of WaniKani. Good luck!
Review accuracy being “too high” isn’t something that should concern you or needs fixing.
The only relevant question here is: do you feel like you can do more lessons and reviews every day? Do you have the time and motivation to do so consistently? If so then go for it, otherwise keep doing what you do. To complete one level/week you need between 20 and 25 lessons/day depending on the level (and you’ll have to use the lesson picker to start radicals and kanji with the right timing).
Going as fast as you’re comfortable for the first ~30 levels is well worth it IMO because it’ll give you an extremely solid kanji foundation that will make reading much easier. Doing the full 60 levels in one year seems pointless to me however, the only reason to do it would be for the achievement itself. You shouldn’t be wasting study time doing 300 reviews/day at level 53 when you could be reading instead.
I think I post this chart once a month on these forums but it’s really the key to using WaniKani efficiently for me:
You can see that you get amazing returns in terms of kanji coverage during the early levels and comparatively minuscule returns later on. The first two levels give you around 14% kanji coverage in the wild, the last 30 less than 10%.
No, that’s because the forums and WK don’t sync perfectly. It should be fixed if you log out and then log in again. (You have to do this after every level up, though, if you want the forum to display your actual level)
I get your point. On the other hand, people are motivated by different things. I can’t stand doing 90% of progress on something and leaving it there just because technically it is not very useful to do 100%. Other people couldn’t care less.
Personally I’d invest that time learning the language in other ways ie studying grammar, reading on Satori Reader, doing listening/speaking practice…you’re a low level now but once you reached like lesson 12/13ish onwards you’d quickly burn out and not have any time to spend learning the language in any other ways if you did too many ie regularly over 15 lessons a day and over 200 reviews. Personally as close to 100 reviews a day is enough for me, I have a lot going on in life outside of japanese study ie work/home so I really can’t afford to spend more time on SRS than that as it would mean I wasn’t doing things that are more useful to me like reading. So whilst you are still early on really just do as many lessons as you feel that you are wanting to do but just be warned it will accumilate and bog you down very quickly so I would recommend taking it steady especially from level 10 onwards. Slow and steady wins the race!
Absolutely and I’m not advocating for not doing those levels at all (well, I have qualms with the last 10 or so, but that’s a different discussion). My point is more about the importance of short term goals and not necessarily committing to the same pace for the entire 60 levels. IMO it makes the most sense to go fast early on and then pull back to spend more time reading and studying other aspects of the language as you get into the second half of the course.
Rushing WaniKani to learn niche kanji that you may have forgotten by the time you actually need them is not a good use of anybody’s time.
…at least, not with the WK system. AIUI, in a generic SRS system, one response to “accuracy is very high” is that you can make all the intervals a bit longer, which will probably reduce the accuracy a bit since it’s now giving you a longer time to forget before re-asking, but will definitely reduce the workload. But since WK intervals are fixed by the system and unchangeable, there is no config knob you can turn to make this trade off, and high accuracy is just an uncomplicatedly good thing (since it implies fewer repetitions and fewer leeches).
Yeah that’s certainly an issue with the fixed intervals of WK. Even the bad legacy algo used by Anki has the “ease” factor to limit unnecessary reviews.
But it is what it is, if you can’t fight it, embrace it!
