How many WaniKani lessons per day is right?
Ten a day, and stop taking lessons whenever your Apprentice count crosses 100. That pairing is the best-documented moderate pace on the forum, and it holds a level every two weeks deep into the 40s.
The reasoning behind it is worth having, because the right number for you comes from your own review tolerance rather than from ten being magic.
A lesson is a purchase on credit
A lesson takes about a minute and costs nothing today. What it does is sign you up for eight scheduled reviews of that item on a fixed timetable: 4 hours from now, then 8 hours, a day, 2 days, a week, 2 weeks, a month, and finally 4 months. Roughly six months of reviews per lesson, at perfect accuracy, and every miss re-runs part of the schedule.
So twenty lessons this morning is 160 reviews you just booked, most of them landing within two weeks. Do N lessons a day and you settle at 8 x N reviews a day. That is the floor; nobody reviews at 100% accuracy, so the realistic multiplier is closer to 10-11x.
Reviews per day at a flat cap
Band bottoms are the 8x perfect-accuracy floor. Tops are reported loads where one exists, and the 10-11x with-misses estimate where none does.
About 9,000 items, divided by your cap
WaniKani holds about 9,000 items (485 radicals, 2,074 kanji, 6,501 vocabulary at the December 2021 count, drifting upward since), so a flat cap divides straight into a finish date. Twenty a day is roughly 15 months. Ten a day is a bit under three years, a figure the finishers who ran that pace confirm. Five a day is about five years.
Two caveats before you trust that division. Reaching level 60 and clearing every lesson are different finish lines: levelling is gated on kanji, so prioritising radicals and kanji gets you to 60 faster while vocabulary piles up behind you. And the fast end has a hard floor, because the SRS timetable sets a minimum of roughly 340 days for level 60 no matter how many lessons you take. The level-60 guide works through where that floor comes from.
Why the tripwire is Apprentice, not lessons
A flat daily cap says nothing about how much is currently in flight. Apprentice items are the expensive ones, coming back every 4 to 47 hours, so they dominate the daily queue no matter how disciplined your lesson habit is. Watching that count catches a bad week before it becomes a backlog.
The commonly quoted comfort range is 100 to 150, with 150 described as the point where it starts to hurt. When you cross your line, stop lessons and keep reviewing: reviews drain the queue without adding new debt. Resume once the count is back under 100, or lower if you can see a busy week coming.
Session accuracy is the second gauge. There is no official target and reported figures run anywhere from 60% to 95%, so the useful signal is your own trend rather than the absolute number. Scores dropping into the 50s and 60s mean stop lessons for a day or two and clear reviews only.
The firehose is now opt-in
WaniKani used to hand new users every available lesson at once, which is how 500-review queues appear three weeks into an account. Since the Today's Lessons update the default is a curated daily batch with a cap you set yourself, anywhere from 0 to 100. Set it once and the decision stops being a daily test of willpower.
You do not have to track the rest by hand. Wanilog watches your Apprentice count and your busiest upcoming review day, and a peak that climbs week after week means the cap is too high. The daily review budget runs the division against your own remaining items and target date.