Site of James

Placing my knowledge on a web page

RTK 1 While Working Full Time: A Study Method

As I posted on Reddit, I often see posts about experiences on tackling RTK. This is my study method for completing Remembering the Kanji 1 6th Edition in less than 5 months (Dec 9 - April 27th; started immediately after I took N5) while also working a full-time job of 9 hours per day.

Since this was mainly inspired by the post here, I similarly feel posting my methods can at least help someone else planning–or in the process–of doing RTK. I also believe it’s important for one to find one’s own way of study and working, so adapt this method to your needs if you like it. I’ll also share how I addressed the largest (IMO) pitfall of the book: near-synonymous keywords.

The main problem with a full-time job is lack of time. The keys of this method are finding crevices to squeeze in time, and a balance of time taken vs. retention rate.

First, in order to do this, be prepared to make your life revolve around the book. Live it, breathe it, love it. Because cutting out the time herein while also working full time isn’t easy, at least not for myself. I needed to make sacrifices to get it done in the “standard” target of 3-4 months. RTK is one of those books that does it’s job fantastically, and is a unique one-of-one, but explicitly assumes the goal of leaning all Jouyou characters (Page 6). To that end, as you may know, because the kanji aren’t given in any “use” order such as frequency, it means you pretty much must complete all 2200 to really get value of it (example, the common 来 in くる is frame 2029). Make sure you have that dedication.

The Primary 20 Kanji Per Day

The starting point is 20 kanji per day, as Heisig calls it “not too excessive” for someone who “has only a couple of hours to study a day.” I agree.

The key for the day’s primary push is in lunch time in the office. At 11:50 or 11:55, cut down a bowl of cereal or some other very quick lunch and finish up by 12:00 to 12:15. After 12:15, you are late. I supported this by bringing in a small amount of milk for lunch each day, and having a reserve 3-5 cereal boxes at my desk. Cereal is ideal because it takes 0 time to make, 0 time to clean up, and 0 prep time. While eating lunch, stand up. This is inspired from Agile software development standup meetings: you’re much less likely to drag on if you’re standing than slouching and getting comfy. While standing, use the time to flip through your Anki review for the day (Anki is detailed later–hold on for that part).

Doing this gives the entire lunch hour free and clear to concentrate. I create my stories, write the kanji as detailed in Session 5 of RTK on 2.5” x 2.5” square half-width index cards. Yes, buy 2200 index cards, as I’ll explain.

The Review Set

Doing 20 kanji a day is nice, but they need to be remembered. Because all kanji have been done on index cards, it gives a very clear picture on exactly what day a kanji was learned. I define the “Review Set” as the set of 20 kanji learned two days earlier (why two? I’ll explain in the next section).

The second daily portion is this review set. Throughout the day, where any free time is had, review the Review Set. Look at the keyword, and guess the kanji whilst recalling your story. To do that, keep them in your pocket to review. During these flashes of time, flip through the Review Set:

  • While walking to the bathroom
  • On to meetings, waiting in meetings
  • Waiting something to download.
  • When driving to and from work, when stopped at red lights

The in-the-car part (also the ease of telling which day they were learned on) is one of the key factors for using purely index cards and not using something like strictly Anki. They’re far easier to handle than whipping our your phone while driving (don’t do it) or getting out a 8.5” x 11” notebook, or even full-size (2.5” x 5”) index cards. Doing this, the 20 in the review set should easily be finished by the end of the workday.

Index cards also encourage natural pen-and-paper study, which is important. On Anki, it’s not as easy to do so.

Anki Reviews

The third daily portion is Anki reviews. I used this deck. One key in a good RTK deck–as I and others have discovered–is the inclusion of crowed-sourced stories from the Kanji Koohi forums ( If you simply cannot think of a good story, this deck has the two most-upvoted stories for it, which is very very useful for backup. A good deck also shows stroke order, to make sure you’re seeing it outside of the pages of RTK. If you would like to make your own deck, that’s cool too.

First and foremost, a card in Anki should be on a lapse of 5 days–that is, the first time you see a kanji in Anki is 5 days after you first studied it out of RTK. Why? Recall that the Review Set is on a two-day lapse. With a 5-day lapse in this manner, we have already invoked spaced repetition, by seeing it once 2 days after, then 3 days after that. This drills in the kanji as soon as possible. As a result, Anki reviews should be fairly quick and easy as kanji are already “half-learned.”

Because there are often 80-100 reviews per day, I partially tackled this by finding morning time in four ways:

  • Start as soon as I opened my eyes in the morning. Instead of going on Reddit, crack open Anki on your phone, and make a dent before getting out of bed!
  • Breakfast should already be made and require no time in the morning, if possible. I make overnight oats the night before (delicious, healthy, cheap, zero effort) which just require me to retrieve them from the fridge.
  • While eating breakfast, I lay in bed doing reviews.
  • My job is flexible in that I can flex my work hours. For example, 8-5 or 9-6, or maybe 10-7. I utilized this it to push more time into the morning for reviews. Also, set an alarm for maybe 15 or 20 minutes before you need to get up, to allow for wake-up and review time–a helpful LPT in general.

You will likely not get all reviews done by the end of the work day. After work, just make sure the Anki reviews are finished before going to bed. You can do this by doing them while eating dinner, for instance. Otherwise, you quickly run the risk of falling faaaar behind, as often reviews are 80-110 per day. You should be 40%+ percent after work done on average days.

I worked by suspending all 2200 cards, and un-suspending up to the Kanji frame number I’ve learned. This makes sure no kanji I haven’t see are shown accidentally shown in Anki, by working too fast.

This 5-day lapse may get out of sync if you don’t study 20 newies one day, for example. It will take some occasional grooming to keep it in line, but I found it often evens itself out if you can sometimes push more than 20 a day.

I had a custom setting as pointed out by Matt of Matt vs. Japan in a video I can conveniently no longer find: Set the lapse interval to 25% instead of 0%. Why? Because we already have been reviewing kanji in multiple different ways, and using the strengths of RTK of long-term memory, it’s a good trade-off to have a card keep momentum instead of resetting purely to nothing.

The Sunday Review

Monday-Saturday, aim for 20 kanji a day, plus Review Set, plus Anki as outlined above. Sunday is break day. This is directly taken from the post that inspired my method at the beginning. On Sunday, take all the index cards for that week (100-115), shuffle them, and do a formal review one-by-one like the Review Set, but write them all down once from your story, and keep track your retention rate. A failure is when you can’t write it perfectly from your story. Aim for 80%-85%.

The ones I recorded were (Format: Failed/Total : Retention %):

  • Week of 2/4: 7/138: 95% (First week of 2-day lapse reviews)
  • Week of 2/11: 8/80: 90%
  • Week of 2/18: 17/98: 82%
  • Week of 2/25: 19/117: 83%
  • Week of 3/4: 23/115: 80%
  • Week of 3/11: 9/109: 91%
  • Week of 3/18: (Unknown): 85%
  • Week of 3/25: 28/70: 60%
  • Week of 4/1: (Unknown): 68%
  • Week of 4/15: 16/106: 84%
  • Week of 4/22: 23/98: 76%

As an aside - What happened on 3/25 and 4/1? I discovered my Anki deck was too fast. That is, a kanji a reviewed yesterday I was seeing the next day. This ruined the 2 and 5 day lapse, compounded with parts lessons being one of the harder ones, according to Heisig (Lesson 51 - “This is probably the most difficult lesson of the book”). I fixed this and things resumed as normal. The last week I can attribute to the fact that only the obscure ones remained, as well.

It’s worth noting to not get hung up on this retention number. Because early next week the failed Sunday cards will be appearing on Anki. You’ll wonder how you ever forgot them by the end of the next week. It’s simply a good checkpoint and chance to reset any loose strings.

Notes on Burnout

Naturally, this is all very intensive each day. In regards to burning out, it’s better to complete the course in, say 2 weeks later than getting burnt out halfway through and giving up by pushing too hard. Keep a nice pace, even if that means 15 a day. Just don’t give up, or else all your effort is simply lost. More than anything, a comfortable pace is the key.

  • It’s okay to not study new Kanji occasionally (just don’t fall behind on Anki)
  • If you’ve done 20 a day and want to, say, do 10 more in the evening because you’re “feeling it”, don’t do it. Save that fire for the next day.

RTK’s Biggest Pitfall (IMO)

I find the most difficult part of RTK is the near-synonymous keywords. The ones that gave me the most trouble learning were general (総), widespread (氾), public (公), universal (普), and overall (統). To break these up, write them all side-by-side in a notebook, and write out the connotation your mind thinks of for each, in a few sentences. Doing so with each next to each other will allow you to learn the differences between each, and contrast the stories of them all.

For example: “universal” implies everyone on the universe is doing it, while “widespread” just implies a single part of a country.

Now, this is essentially exactly what Heisig says to do in the beginning of the book: the connotations of the keywords are very important–who knew? But keep it in mind.

That’s It

Now, could things be tweaked to help weekly retention? Sure, but this method prioritizes time saving and speed. Another basis is the fact that if you learn 120 things and have a 80% retention rate, you’ve simply learned more things than someone who has learned 100 things with 90% retention.

Does this work for everyone? No. Are there more efficient ways? Maybe. But this was my method and I’m very satisfied with the results. During this, I read manga (currently Komi-san Volume 7) before bed, and I can safely point to RTK as me being able to understand mostly every page.

At the end of the book, in Anki I sat at at 1482 Mature (67%), and 718 Young+Learn. Not bad.

  • Aim for doing 20 per day, or as long as an hour allows. My key to this is lunch hour at the office.
  • Review the “Review Set” of the kanji from two days ago each day. Don’t forget to bring these to work.
  • Anki reviews are offset by 5 days. That means this is three days after the review set.
  • Sunday Reviews are important to keep track on progress and adjust for the next week
  • Life adjustments: More time in the morning for reviews, make sure nobody takes your precious lunch hour for pointless meetings, carry RTK with you when possible, always have index cards on my hands.

Please enjoy your kanji!