Kakaku.com Japan Lowest Prices — Shops, Rank, Rating

Unofficial — independent tool, not affiliated with, endorsed by, or sponsored by Kakaku.com. It reads only publicly visible pages. Support, reliability guarantees and the full disclaimer are at the bottom of this page.

What it does: Type a product keyword and get what it costs new in Japan on Kakaku.com — lowest price, how many shops sell it, popularity rank, review score.

You enter: one or more keywords — a model number like LUMIX TZ99, or the Japanese name カメラ (camera). A weekly ranking page can be given instead, on its own.

You get: one row per keyword: lowest price (typical and lowest–highest) over the products read, shops per product, review score, the top match's rank (1 of 2,246), the site's own price band table over every priced hit, and the site's hit counts.

Price: $0.02 per keyword — a weekly ranking page counts as one keyword. +$0.002 per row if you also want one row per product. No results = no charge.

Example: enter LUMIX TZ99 → 1,129 hits · 1 catalogue product and 39 shop listings read · lowest price ¥70,074 at 40 shops · rank 1 of 2,246 · review 4.13 from 53 people · the typical priced hit is only ¥2,588 because cases and filters match the same words (real run, 2026-09-08)

Unofficial — not affiliated with Kakaku.com. Reads public pages only.

Pricing — $0.02 per keyword

Event	Price	When
Keyword price summary	$0.02	Per keyword, and per weekly ranking page
Individual product	$0.002	Only if you turn on Include individual products

A default run (1 keyword, summary only) costs $0.02 and makes 2 requests to the site. You pay per keyword; there is no monthly fee. A keyword that returns zero listings is never charged, and neither is one the site answers with its "0 件" page.

Two limits meet a long list, and the time one arrives first: the run's default time limit is 300 seconds, which at ≥1.6 s per request is about 80 keywords at one page each with the rank lookup on, while the platform's Maximum cost per run ($3.00 by default) buys 150 summaries. Keywords the run has no time left for are skipped — not read, not charged, and named in the log. Raise either limit in the run options, or split the list across runs.

Input

Field	Example	Notes
keywords	["LUMIX TZ99"]	Model numbers as printed (ILCE-7M4) or Japanese names (カメラ). Each keyword costs $0.02. Give keywords, rankingUrls, or both — at least one of the two
rankingUrls	["https://kakaku.com/camera/digital-camera/ranking_0050/"]	Optional, and usable on its own without any keyword: Kakaku.com's weekly best-seller lists. Each address costs $0.02, like a keyword. Only this exact address form is read
category	"0003_0001"	Optional. One category only: 0003_0001 digital cameras, 0003 cameras, 0005 phones, 0001 computers. Changes the hit count, the price bands and which products are measured
pagesPerKeyword	1	1–5. One page = up to 40 products in popularity order = 1 request. Default 1
sortBy	"standard"	standard (popularity, the site's own order), price_asc, price_desc. Use the price orders only together with a category — see below
priceMin / priceMax	50000 / 100000	Optional yen limits. They narrow the hit count and the price bands as well as the products measured
includeTopProductRank	true	On by default. Adds the top match's rank / rankOf from its product page (1 extra request per keyword, no extra charge)
includeIndividualItems	false	Turn on to also get one row per product read (+$0.002 each)
convertToUsd	true	Adds US$ figures at today's rate. A failed rate lookup never fails the run

Which number covers what

Two different price numbers come back, and they answer two different questions. Both carry a label so you never have to guess:

Field	Label	What it covers
priceJpy	priceJpyBasis: "sampled_catalog_products"	The lowest price of each catalogue product actually read (up to 40 per page). count says how many products that was
priceBands / bandMedian	priceBandsBasis: "all_priced_hits_including_accessories"	The site's own 56-band count over every priced hit for the keyword — shop listings, cases, filters and spare batteries included

That is why the example above shows ¥70,074 next to ¥2,588: the first is the camera, the second is the middle of everything the words LUMIX TZ99 match on the site. pricedListings (the bands' total) next to totalListingsFound tells you how much of the result set carries a price at all — 1,129 of 1,129 here, 96% for a broad word like カメラ.

productsSampled and mallListingsSampled split what was read: catalogue products (the ones with a shop count, a rank and a review score) and single-shop listings, which are counted but never returned as rows and never followed — those links point into a part of the site that robots.txt closes.

When a keyword comes back empty

keywordStatus is ok as soon as a catalogue product was read, not_found when the site declares 0 hits, and unknown when there are hits but all of them are shop listings, so there is no lowest price, shop count or rank to report. The last two come with a one-sentence hint naming the next thing to try (null otherwise). Neither a not_found keyword nor a failed run is charged.

Output example (type: "keyword_price_summary")

Measured on 2026-09-08 (real run, {"keywords": ["LUMIX TZ99"], "pagesPerKeyword": 1, "includeTopProductRank": true, "includeIndividualItems": false}). The whole row, nothing shortened — including all 56 price bands exactly as the site counted them.

{
  "type": "keyword_price_summary",
  "keyword": "LUMIX TZ99",
  "keywordStatus": "ok",
  "sortUsed": "standard",
  "categoryUsed": null,
  "totalListingsFound": 1129,
  "pricedListings": 1129,
  "sampledListings": 40,
  "productsSampled": 1,
  "mallListingsSampled": 39,
  "priceJpy": {
    "min": 70074,
    "p25": 70074,
    "median": 70074,
    "p75": 70074,
    "max": 70074,
    "average": 70074,
    "count": 1
  },
  "priceJpyBasis": "sampled_catalog_products",
  "priceUsd": {
    "min": 453.8,
    "p25": 453.8,
    "median": 453.8,
    "p75": 453.8,
    "max": 453.8,
    "average": 453.8
  },
  "exchangeRateJpyUsd": 0.006476,
  "shopCount": { "min": 40, "median": 40, "max": 40 },
  "rating": { "average": 4.13, "ratedProducts": 1 },
  "priceBands": [
    { "from": null, "to": 199, "count": 0 },
    { "from": 200, "to": 399, "count": 0 },
    { "from": 400, "to": 599, "count": 8 },
    { "from": 600, "to": 799, "count": 4 },
    { "from": 800, "to": 999, "count": 58 },
    { "from": 1000, "to": 1199, "count": 89 },
    { "from": 1200, "to": 1399, "count": 108 },
    { "from": 1400, "to": 1599, "count": 59 },
    { "from": 1600, "to": 1799, "count": 73 },
    { "from": 1800, "to": 1999, "count": 61 },
    { "from": 2000, "to": 2249, "count": 19 },
    { "from": 2250, "to": 2499, "count": 63 },
    { "from": 2500, "to": 2999, "count": 128 },
    { "from": 3000, "to": 3499, "count": 113 },
    { "from": 3500, "to": 3999, "count": 121 },
    { "from": 4000, "to": 4499, "count": 30 },
    { "from": 4500, "to": 4999, "count": 13 },
    { "from": 5000, "to": 5999, "count": 26 },
    { "from": 6000, "to": 6999, "count": 9 },
    { "from": 7000, "to": 7999, "count": 1 },
    { "from": 8000, "to": 8999, "count": 1 },
    { "from": 9000, "to": 9999, "count": 1 },
    { "from": 10000, "to": 12499, "count": 0 },
    { "from": 12500, "to": 14999, "count": 0 },
    { "from": 15000, "to": 17499, "count": 0 },
    { "from": 17500, "to": 19999, "count": 0 },
    { "from": 20000, "to": 22499, "count": 0 },
    { "from": 22500, "to": 24999, "count": 0 },
    { "from": 25000, "to": 27499, "count": 0 },
    { "from": 27500, "to": 29999, "count": 0 },
    { "from": 30000, "to": 32499, "count": 0 },
    { "from": 32500, "to": 34999, "count": 0 },
    { "from": 35000, "to": 37499, "count": 0 },
    { "from": 37500, "to": 39999, "count": 0 },
    { "from": 40000, "to": 42499, "count": 0 },
    { "from": 42500, "to": 44999, "count": 0 },
    { "from": 45000, "to": 47499, "count": 0 },
    { "from": 47500, "to": 49999, "count": 0 },
    { "from": 50000, "to": 54999, "count": 0 },
    { "from": 55000, "to": 59999, "count": 0 },
    { "from": 60000, "to": 64999, "count": 0 },
    { "from": 65000, "to": 69999, "count": 0 },
    { "from": 70000, "to": 74999, "count": 2 },
    { "from": 75000, "to": 79999, "count": 71 },
    { "from": 80000, "to": 84999, "count": 28 },
    { "from": 85000, "to": 89999, "count": 13 },
    { "from": 90000, "to": 94999, "count": 8 },
    { "from": 95000, "to": 99999, "count": 9 },
    { "from": 100000, "to": 124999, "count": 5 },
    { "from": 125000, "to": 149999, "count": 3 },
    { "from": 150000, "to": 174999, "count": 3 },
    { "from": 175000, "to": 199999, "count": 2 },
    { "from": 200000, "to": 299999, "count": 0 },
    { "from": 300000, "to": 399999, "count": 0 },
    { "from": 400000, "to": 499999, "count": 0 },
    { "from": 500000, "to": null, "count": 0 }
  ],
  "priceBandsBasis": "all_priced_hits_including_accessories",
  "bandMedian": 2588,
  "topProduct": {
    "id": "J0000046783",
    "name": "LUMIX DC-TZ99",
    "maker": "パナソニック",
    "category": "デジタルカメラ",
    "lowestPrice": 70074,
    "shopCount": 40,
    "rating": 4.13,
    "reviewCount": 53,
    "releaseDate": "2025-02-20",
    "url": "https://kakaku.com/item/J0000046783/",
    "rank": 1,
    "rankOf": 2246
  },
  "hint": null,
  "truncatedForTimeLimit": false,
  "sourceUrl": "https://search.kakaku.com/LUMIX%20TZ99/?sort=standard&page=1",
  "checkedAt": "2026-09-08T12:25:36.245453+00:00"
}

With includeIndividualItems on you also get one type: "item" row per product read: productId, name, maker, category, lowestPrice, shopCount, rating, reviewCount, releaseDate, url, rankInSearch, page. A ranking address instead returns one type: "ranking_summary" row: rankingUrl, updatedOn, aggregateWindow (the week the site counted), rankedProducts, pricedProducts, priceJpy, rating, and top — the first 10 products with their rank, maker, lowest price and score.

What this Actor does not do

• No sold prices. These are shop prices you could pay today, not what anything actually sold for. Kakaku.com publishes no sale records
• No shop-by-shop price list. The row says how many shops sell a product, not which ones or at what price each
• No price history. The site's price-history endpoint is closed to automated readers by its robots.txt, so it is never touched
• No shop listing links. Single-shop offers are counted (mallListingsSampled) but their links leave the site through a path robots.txt closes, so no such address is ever read or returned
• No reviews, images or product descriptions. Only facts and figures: names, makers, prices, shop counts, ranks, scores, release dates
• No login-only pages, and nothing is kept. Every run reads public pages live

Notes on the data

• "0 hits" arrives as an HTTP 404 with a complete page — measured 2026-09-08, a keyword the site does not carry returns 404 and 20,746 bytes of ordinary HTML. That is reported as not_found, not as a failure. A 404 from a ranking page or a product page is treated as a real miss instead
• The hit count includes shop listings. totalListingsFound is the site's own number (1,129件) and counts catalogue products plus single-shop offers, which is why a model keyword can show thousands of hits and a single product. It also drifts by a few hundred between requests on very broad words
• One result page holds 40 products (plus up to 10 [PR] cards, which are always dropped). A page is 250–310 KB, a product page up to 571 KB
• Cheapest-first is a trap on model keywords. Measured 2026-09-08: LUMIX TZ99 sorted cheapest-first returns 40 cards of ¥400–800 cases and filters and not one camera. Popularity order is the default for that reason, and sortUsed always says which order produced the row
• Ranks are weekly. A ranking row carries updatedOn and aggregateWindow — e.g. updated 2026-09-08 from the week 2026-09-01 to 2026-09-07. Ranks can tie, and a ranked product with no shop price today shows ¥― on the site and lowestPrice: null here; those rows are counted but stay out of the price figures (pricedProducts says how many had a price)
• A brand-new product can have a price and no score yet (rating: null, reviewCount: null); rating.ratedProducts says how many of the products read had a score at all
• Release dates are converted to 2025-02-20 form; a product announced only by month keeps 2025-02, and one with no date at all stays empty
• Every page is Shift_JIS and is read as cp932 with replacement for the handful of characters that break strict decoding. The keyword itself travels UTF-8 percent-encoded inside the address
• Requests are throttled to ≥1.6 s and the run carries two wall-clock marks. At 90 seconds the extra pages and the rank lookup are dropped, and the rows that were cut say so (truncatedForTimeLimit: true); at 240 seconds no new keyword is started, so the run finishes inside the platform's 300-second limit instead of being killed — the keywords it did not reach are listed in the log and are never charged

If something goes wrong

• Wrong number or a failed run? Open a ticket on the Issues tab. I read every one and reply within 2 business days (Japan time).
• You never get a fake "empty" result. If the site can't be read, the run fails and says so.
• No results = no charge. You only pay for results you actually get.
• Checked every week. An automatic test runs this tool weekly; if the site changes, I fix it.
• Public pages only. No login, no personal data, and it goes easy on the site.

More tools by the same author

• BookOff Japan Used Manga, Books, CDs — Price & Stock
• @cosme Japan Beauty Ranking — Top 50 Ratings & Prices
• Digimart Japan Used Guitar & Instrument Prices
• Fujiya Camera Japan Used Camera Prices by Condition
• HobbyLink Japan Gunpla & Figure Prices + Stock Status
• Iosys Japan Used iPhone & Phone Prices by Condition
• JACKROAD Japan Watch Prices — New, Used, Vintage, In Stock
• KOMEHYO Japan Luxury Resale Prices by Condition
• Mercari Japan Sold Prices — What Items Really Sell For
• Yahoo! Auctions Japan Sold Prices — Median, Range, Bids

All tools (Japan marketplaces, real estate, jobs, racing, prediction markets): https://apify.com/jpmarketdata

Disclaimer

Unofficial, independent tool — not affiliated with, endorsed by, or sponsored by Kakaku.com. Product names and logos belong to their owners and only say where the data comes from. Data is read from public pages, for market research; check before you act on it.