Everyday Japanese Kanji: The Ones to Learn First (N5–N4)

Here is a sign you will meet within ninety seconds of arriving at any Japanese station:

東口  ·  西口  ·  出口

Three pairs of characters. Let’s take them apart.

• 口 (kuchi, or guchi when it follows another word) literally means “mouth”. On a building it means an opening — a way in or out.
• 東 (higashi) is east. 西 (nishi) is west. 出 (deru) means to go out.

So: 東口 (higashi-guchi) is the east exit. 西口 (nishi-guchi) is the west exit. 出口 (deguchi) is simply “the exit”. You have just learned four characters, and you will see them repeatedly throughout a typical trip.

That is the whole idea of this article: the everyday Japanese kanji you meet without trying are the ones worth learning first. Most people who give up on Japanese give up at kanji — and usually at the number rather than the characters themselves. Our own beginner course covers around 130 kanji, and by the time you reach upper-intermediate the figure is closer to 1,200. Seen as a list, that is crushing. Seen as a street, it is quite manageable, because the street repeats itself.

And you do not have to be in Japan to start. Everything below works from a desk in India — Street View, a station map and a few photographs are enough. There is a seven-day version at the end that gets you to around sixty characters without a plane ticket.

A note on the labels used below. The JLPT (Japanese-Language Proficiency Test) is an internationally recognised Japanese-language proficiency test. It runs from N5, the beginner level, up to N1, the most advanced. One thing worth knowing early: the JLPT does not publish an official kanji list for any level. The level labels everyone uses — including us, below — come from widely used reference lists compiled by teachers and publishers, and those lists occasionally disagree with each other by a level. So when this article calls a character “commonly classified as N5”, that is exactly what it means: it appears on the beginner lists most schools work from, not on an official roster. If you are not sure which level you should be aiming at, our free level test will tell you in a few minutes.

Layer 1: Eight characters that help you decode Japanese addresses

Before learning any place name, learn the containers they sit in. Japanese addresses are written largest unit first, and eight characters mark the administrative divisions you will meet most often. Not all of them appear in any one address — a given address uses whichever divisions apply to it — but between them they cover most of what you will see.

That order is the reverse of an Indian address, which starts with the flat and ends with the state. The Tokyo Metropolitan Government Building, for example, is 東京都新宿区西新宿2-8-1 — Tokyo Metropolis, then Shinjuku ward, then the Nishi-Shinjuku district, then block 2, sub-block 8, building 1. Written the English way, the same address comes out backwards: 2-8-1 Nishi-Shinjuku, Shinjuku, Tokyo. The eight characters:

• 都 (to) — metropolis. Only one place uses it: 東京都 (Tōkyō-to), Tokyo Metropolis.
• 道 (dō) — circuit. Also only one: 北海道 (Hokkaidō).
• 府 (fu) — urban prefecture. Two only: 京都府 (Kyōto-fu) and 大阪府 (Ōsaka-fu).
• 県 (ken) — prefecture. The other 43.
• 市 (shi) — city
• 区 (ku) — ward
• 町 (chō or machi) — town
• 村 (son or mura) — village

Learn those eight and many Japanese addresses suddenly have a visible skeleton. You may not be able to pronounce the name in the middle, but you will recognise which part is the prefecture, which is the city and which is the ward or town. That is genuine reading, and it takes about twenty minutes.

Layer 2: The station, where this method pays best

Station vocabulary is the single most useful set in this whole article. When we counted the characters on station signage, around three-quarters of them sat in the beginner-to-intermediate range, and only a handful were rare enough to fall outside the common reference lists altogether — the best ratio of any group we looked at. (Our counting method is set out near the end.) It is one of those uncommon cases where the practical vocabulary and the exam vocabulary are close to the same list.

Start with these. Nearly all are commonly classified as N5 or N4:

Kanji	Reading	Meaning	Where you see it
駅	eki	station	東京駅 (Tōkyō-eki), 京都駅 (Kyōto-eki)
口	kuchi / guchi	mouth → opening	東口, 西口, 北口 (kita-guchi), 南口 (minami-guchi)
出 / 入	de / iri	exit / enter	出口 (deguchi), 入口 (iriguchi)
行	yuki / iki	bound for	大阪行 (Ōsaka-yuki) — this train goes to Osaka
発 / 着	hatsu / chaku	departure / arrival	発 10:15   着 12:40
特急	tokkyū	limited express	On the front of the faster trains
急行	kyūkō	express	急 = hurry, 行 = go
電車	densha	train	電 = electric, 車 = vehicle
番線	bansen	platform number	3番線 (san-bansen) = platform 3

Look again at what 口 is doing there. Every textbook introduces it as “mouth”, and a beginner dutifully memorises “mouth”. On a station wall it means the way out, and once you have stood between 東口 and 西口 at opposite ends of a station, you will never forget 口 or the direction characters again.

That is the real mechanism here, and it is not mnemonics. It is meeting one character doing an obvious job in a place where the job is obvious.

Once those are comfortable, add the harder station set as recognition vocabulary — characters you can spot without being able to write them:

• 改札 (kaisatsu) — ticket gate. You will usually see 改札口 (kaisatsu-guchi).
• 精算 (seisan) — fare adjustment, for when you underpaid at the start.
• 運賃 (unchin) — the fare.
• 乗換 (norikae) — transfer to another line.
• 各駅停車 (kakueki-teisha) — the local train, literally “stops at every station”.

Layer 3: The map — and why Kyoto teaches better than Tokyo

This was the most useful thing that came out of our analysis, and it surprised us.

Kyoto’s ward map is a direction-kanji drill that already exists. The city has eleven wards, and six of them are simply a direction plus 京 (kyō, capital):

Ward	Reading	Literally
上京区	Kamigyō-ku	upper capital (上 ue/kami = above)
中京区	Nakagyō-ku	middle capital (中 naka = middle)
下京区	Shimogyō-ku	lower capital (下 shita/shimo = below)
左京区	Sakyō-ku	left capital (左 hidari = left)
右京区	Ukyō-ku	right capital (右 migi = right)
西京区	Nishikyō-ku	west capital (西 nishi = west)

Add 北区 (Kita-ku, north ward), 南区 (Minami-ku, south ward) and 東山区 (Higashiyama-ku, east mountain) and nine of the eleven wards are built from characters a learner meets early. Only 山科区 (Yamashina-ku) and 伏見区 (Fushimi-ku) fall outside that range.

One map gives you 上下左右 (up, down, left, right) and 東西南北 (east, west, south, north) together, attached to real places you could walk through. No flashcard deck does that.

Tokyo is less predictable for a beginner — and it is the map everybody reaches for first. Nine of its 23 wards are written with characters commonly classified as beginner-level: 千代田 (Chiyoda), 文京 (Bunkyō), 台東 (Taitō), 品川 (Shinagawa), 目黒 (Meguro), 大田 (Ōta), 中野 (Nakano), 北 (Kita) and 足立 (Adachi). The others include characters that are considerably more advanced, or absent from the common reference lists altogether — the 墨 in 墨田 (Sumida), the 渋 in 渋谷 (Shibuya), the 葛 and 飾 in 葛飾 (Katsushika).

There is nothing wrong with learning the Tokyo wards, and if you are going there you probably should. But for a beginner working specifically on directional and positional kanji, the Kyoto ward map offers a noticeably tidier set of recurring patterns.

The twelve characters that keep coming back

Across every place name and sign we examined, these repeated most often — and eleven of the twelve are commonly classified as beginner-level:

中 (naka, middle) · 川 (kawa, river) · 田 (ta, rice field) · 東 (higashi, east) · 京 (kyō, capital) · 北 (kita, north) · 山 (yama, mountain) · 大 (ō, big) · 西 (nishi, west) · 野 (no, field) · 南 (minami, south) · 口 (kuchi, opening)

Once you have these, names start coming apart by themselves:

• 中野 (Nakano) — middle field
• 山口 (Yamaguchi) — mountain mouth
• 大田 (Ōta) — big rice field
• 川崎 (Kawasaki) — river cape

Many Japanese place names preserve descriptions of the landscape — rivers, fields, mountains, directions and other geographical features. Once you can see those elements, a great many names stop being opaque blocks and start being descriptions of somewhere.

Layer 4: Everyday Japanese kanji on the street

Three sets you will see on almost any shopping street. In our count, each of them ran at roughly four-fifths beginner-to-intermediate characters.

Is it open?

• 営業中 (eigyō-chū) — open for business
• 準備中 (junbi-chū) — preparing, i.e. not open yet. A very common disappointment.
• 定休日 (teikyūbi) — regular closing day

Notice that 中 — the same character you just learned from 中野 and 中京区 — is doing the work in the first two. Here it means “in the middle of doing something”.

What does it cost?

• 割引 (waribiki) — discount
• 半額 (hangaku) — half price
• 無料 (muryō) — free
• 税込 (zeikomi) — tax included
• 現金 (genkin) — cash

Things that actually matter

• 注意 (chūi) — caution
• 危険 (kiken) — danger
• 禁止 (kinshi) — prohibited
• 立入禁止 (tachiiri-kinshi) — no entry
• 非常口 (hijōguchi) — emergency exit
• 男 (otoko) and 女 (onna) — men and women, on every toilet door in the country
• 交番 (kōban) — police box

There is your 口 for the third time, in 非常口. A character you meet three times in one afternoon does not need revising.

One set to leave for later: restaurant menus. 丼 (donburi, rice bowl), 麺 (men, noodles) and 寿司 (sushi) all sit outside the common beginner reference lists, and menu vocabulary came out weakest of every group we looked at. Wonderful for a holiday, less efficient if you are working towards an exam.

Where this method stops working

This is the part most articles on the subject leave out, and it will save you more time than anything above.

Four things signs will never teach you

We checked which beginner-level characters our whole collection of signs and place names failed to contain. Twenty-nine never appeared at all, and they fall into remarkably clean groups:

• Numbers: 一 (ichi), 二 (ni), 四 (yon), 六 (roku), 七 (nana), 八 (hachi), 九 (kyū), 十 (jū), 百 (hyaku, hundred), 万 (man, ten thousand)
• Time: 今 (ima, now), 年 (toshi, year), 月 (tsuki, month), 午 (go, noon), 毎 (mai, every), 間 (aida, interval)
• People: 人 (hito, person), 友 (tomo, friend), 父 (chichi, father), 母 (haha, mother)
• Speaking and listening: 聞 (kiku, to hear), 話 (hanasu, to speak), 語 (go, language), 読 (yomu, to read)

The pattern is not a coincidence. Signs and place names are nouns and directions. They never count, never tell you the date, never mention your mother, and never speak. Everything social and conversational about Japanese — which is to say, the entire part you would use to actually talk to someone — is invisible on a street.

Use your surroundings for the concrete nouns they teach so well. Get numbers, time, family and verbs from structured study, because nothing outside a classroom is going to hand them to you.

What about verbs?

Signs are full of verbs. Almost none of them are in a form you can use yet.

Look at what a street actually hands you. 押 (osu, push) and 引 (hiku, pull) sit on doors as bare stems, with nothing attached. 止まれ (tomare, stop) on a road sign is a plain imperative — a form most courses leave until much later, and one you would almost never say to a person. 禁止 (kinshi, prohibition) and 営業 (eigyō, trading) are verbs that have been turned into nouns. ご注意ください (go-chūi kudasai, please take care) is a polite request built around a noun rather than a conjugated verb.

What you will not find on a sign anywhere in Japan is the dictionary form, the polite ます form, the past tense or the て-form — the four things a beginner course spends months on. 食べる, 食べます, 食べました and 食べて appear on no station board.

There is a second, quieter problem, and it is specific to kanji. Signs strip the okurigana — the kana tail that follows a character and carries its ending. 押 on a door is really 押す (osu). 止 in 通行止 is really 止まる (tomaru). And that tail is doing real work: 出 on its own could be 出る (deru, to go out) or 出す (dasu, to take something out) — two different verbs sharing one character, told apart only by the kana the sign has removed. The same is true of 上がる and 上げる, 下がる and 下げる, 入る and 入れる. A street will show you 出 a hundred times and never once tell you which verb it belongs to.

So signage teaches you to recognise the characters that verbs are built from, while telling you nothing about how verbs actually behave. That is a perfectly good division of labour, as long as you know it is happening.

What this helps with, and what it doesn’t

It helps a lot with:

• Recognising kanji by shape, and telling lookalikes apart — 語, 話 and 読 stop blurring together once you have met them in the wild
• Meanings of concrete nouns — places, directions, buildings, facilities
• Compound logic — that 出 + 口 makes “exit”, that 営業 + 中 makes “open”. This is arguably the most valuable habit the method builds, because it is how most Japanese vocabulary is assembled
• Survival reading — stations, shops, warnings, addresses
• Momentum — kanji stops feeling like an undifferentiated wall

It helps a little with:

• Readings — you meet on’yomi in compounds and kun’yomi in place names, but unpredictably, and with nothing to check yourself against
• Vocabulary breadth — real, but narrow: place, commerce and safety, and almost nothing else

It does not help at all with:

• Grammar — particles, sentence structure, word order. Signs are noun phrases; they have no grammar to learn from
• Verb conjugation, as above
• Listening — a sign is silent. You cannot hear 営業中 by looking at it
• Speaking, and writing from memory — recognising 駅 on a wall is a long way from producing it on a page or in a sentence
• Stroke order and stroke count — a sign shows you the finished character and none of the moves that made it. You cannot learn to write from something already written
• Okurigana — stripped from signage, so verb and adjective endings stay invisible
• Kun’yomi readings — signage is compound-heavy, so it feeds you on’yomi almost exclusively. The native readings arrive through place names, unreliably
• Politeness levels — the register of signage is nothing like the register of conversation
• Numbers, counters, dates and times
• Abstract vocabulary — opinions, feelings, reasoning, anything you would actually discuss
• Stroke order and handwriting — a sign shows you the finished character, never how it is built

The risk in that last list is a subtle one, and worth naming. Recognition comes quickly and feels like real progress. A learner who can read fifty signs can feel quite advanced — right up to the moment they need to ask 出口はどこですか (deguchi wa doko desu ka — where is the exit?) and discover that no sign ever taught them how to build that sentence. The characters were the easy half.

Meanings are reliable. Readings are not.

Most Japanese characters carry more than one pronunciation, and they generally fall into two families. The on’yomi is a Japanese adaptation of the character’s historical Chinese pronunciation, and tends to appear in compound words. The kun’yomi attaches the character to an existing Japanese word with a matching meaning, and tends to appear when the character stands alone. 山 is read san in one family and yama in the other. Both are correct; which one surfaces depends on the word, and many characters have several of each.

Place names are a special case. They were often fixed centuries ago, and they do not reliably follow those patterns. When we worked through the 47 prefecture names, most could be reconstructed from readings a learner would already know — but several could not:

宮城 (Miyagi) · 茨城 (Ibaraki) · 神奈川 (Kanagawa) · 奈良 (Nara) · 鳥取 (Tottori) · 愛媛 (Ehime) · 大分 (Ōita)

愛媛 is the clearest example: “Ehime” cannot be assembled from the readings of its two characters. The kanji fit the name as a whole rather than syllable by syllable. 神戸 (Kōbe) behaves similarly. And the further you travel from central Japan, the less the standard patterns apply:

• Hokkaidō. Many place names on the island come from Ainu, the language of Hokkaidō’s indigenous people, written with kanji chosen for their sound rather than their meaning. 登別 (Noboribetsu) comes from Ainu nupur-pet, “dark-coloured river” — the pet element meaning “river” is what the 別 is transcribing, and the same 別 recurs in a number of Hokkaidō names. On its own that character means “separate”, which tells you nothing whatsoever about the town.
• Okinawa. In the Okinawan language, nishi means north, where the cognate word nishi in standard Japanese means west. The town of 西原 (Nishihara) takes its name from lying north of Shuri, the old Ryukyuan capital. This is a feature of Okinawan and of historical naming rather than of modern standard Japanese — you should not expect 西 to mean north anywhere else — but it is a good illustration of how much local history a place name can be carrying.

None of this makes the method useless. It means you should use signs and place names for meaning and shape, and get your readings from vocabulary study, where they behave far more predictably.

A seven-day version you can start tonight, from India

You do not need to be in Japan for any of this. Street View, a map and station websites are enough.

1. Day 1 — The eight address containers: 都 道 府 県 市 区 町 村. Write your own address — Bengaluru, or anywhere — in the Japanese largest-to-smallest format.
2. Day 2 — Directions and position: 東 西 南 北 上 下 左 右 中. All commonly classified as beginner-level.
3. Day 3 — The Kyoto ward map. Nine wards, built largely on what you learned yesterday plus 京 and 山.
4. Day 4 — Station survival: 駅 口 出 入 行 発 着 電 車. Then find a photo of a station exit board online and read it.
5. Day 5 — The landscape twelve. Take ten station names and pull them apart.
6. Day 6 — Shop and safety signs: 営業中 準備中 割引 半額 注意 危険 禁止 非常口.
7. Day 7 — Review, then deliberately test the edges. Try to read a menu and a train timetable. Notice exactly where you stall — that is your syllabus for next week.

Seven days, around sixty characters, learned in context rather than in isolation. That is a very good start. It is not a course.

How we analysed the examples

The comparisons above come from a count we ran ourselves rather than from published research, so it is worth being clear about how we did it and what it can and cannot tell you.

What we looked at. Two sets. The first was 230 place names: all 47 prefectures, 122 ward names from twelve cities (Tokyo, Osaka, Yokohama, Nagoya, Kyoto, Sapporo, Kobe, Fukuoka, Kawasaki, Saitama, Hiroshima and Sendai), and 50 of the country’s busiest railway stations. The second was 144 words drawn from everyday signage, grouped by where you meet them: stations and transport, shops and prices, safety and warnings, buildings and directions, food, and the address system. Place names are a fixed, public, countable set. The signage words are not — we chose them as a fair sample of what a visitor actually reads in a day, and a different sensible person would have chosen a somewhat different list.

How we counted. Each distinct character was counted once per group, however many times it appeared in that group — so 中, which turns up in 中野, 中京区 and 営業中 alike, counts once towards the group it is being measured in. We then checked each character against widely used JLPT reference lists, the kind most textbooks and schools work from. “Beginner to intermediate” throughout means a character commonly classified as N5, N4 or N3.

What this is not. Because the JLPT publishes no official kanji list, any level figure depends on which reference list you pick, and the lists disagree at the margins. The numbers below are rounded for that reason — the pattern they show is solid, the second decimal place would be false precision.

Where you see it	Commonly classified N5–N3
Kyoto’s 11 ward names	about 90%
Buildings and directions	about 80%
Safety and warning signs	about 80%
Shops and prices	about 80%
Stations and transport	about 75%
The 50 busiest station names	about 60%
Tokyo’s 23 ward names	about 60%
The 47 prefecture names	about half
Restaurant menus	under half

Two patterns shaped this whole article. First, everyday signage is better beginner material than place names — roughly seven in ten characters on signs are beginner-to-intermediate, against about half for place names. Second, and less obviously, the famous names are the weakest material. Prefecture names came second from bottom; menus came last.

Put both sets together and you would meet somewhere around sixty per cent of a typical N5 list and about half of a typical N4 list — but under a quarter of N3. That is the honest boundary of this approach. It is a beginner method, and it thins out quickly above N4.

The reason it thins out is worth knowing, because it is not random. Signage kanji are overwhelmingly concrete: places, directions, objects, buildings, instructions. N3 is roughly where Japanese starts demanding abstract characters instead — 思 (think), 考 (consider), 感 (feel), 議 (discuss), 論 (argue), 経 (experience). Nobody puts those on a station wall, because nothing on a station wall is an idea. Once the syllabus moves from the world to thinking about the world, the street has nothing left to teach you.

Where a teacher takes over

The honest summary: everyday Japanese kanji will give you a good share of the beginner characters, almost entirely in nouns and directions, with readings you cannot fully trust. It is an excellent on-ramp and a poor destination.

What it cannot give you is sequence. At Saachi, kanji is built in deliberately — around 130 characters across 100+ hours at N5, about 340 over another 100 hours at N4, then roughly 700 at N3 and 1,200 at N2 — each character attached to the grammar and vocabulary that make it usable rather than merely recognisable. Sessions run 2.5 hours, twice a week. Every level includes extensive listening practice, role-play activities, conversation practice with your teacher and classmates, interaction with native Japanese speakers, written and conversation tests after every five lessons, and a student presentation at the end of the course — so the characters you learn get spoken aloud rather than only spotted.

Our batches run from 18 to 35 learners, which people sometimes overlook as a strength: it means a room full of classmates at your own level for the pair work, role plays and presentations. Kanji sticks when you use it with someone, not when you stare at it alone.

If you have been circling Japanese for a while and are not sure where you would fit, the quickest answer is our free JLPT level test — a few minutes, and it tells you which level to aim at instead of guessing. If you would rather just see what a class feels like, we run a free demo: write to us through the enquiry form or message us on WhatsApp at +91 95914 57722.

Keep reading

• How to Prepare for JLPT N4: A Step-by-Step Study Plan for Indian Learners — where the jump to 340 kanji happens, and how to pace it.
• Can You Learn Japanese from Anime? What Actually Works — the same question asked of a different shortcut, with a similarly mixed answer.
• How to Choose the Right JLPT Level: N5 to N2 Explained for Indian Learners

Japan’s map has been teaching its own writing system to anyone who looks carefully for about thirteen hundred years — the edict standardising place names into two auspicious characters dates to 713 AD. Use it for what it is good at, know exactly where it runs out, and get the rest from a teacher. Speak Japanese with confidence.