Yahoo! Auctions Japan Monthly History — Legacy Task/API

Legacy entrance for saved Tasks and API calls. Keep using this Actor for existing monthly-history integrations and their original input names. For new workflows, use Yahoo! Auctions Japan Sold Price Summary & Monthly History with includePriceTrend: true. The migration example below explicitly disables individual rows, which are on by default in the main Actor.

Legacy pricing: $0.02 per keyword only when at least two requested months meet minPerMonth and support a trend. Optional individual rows cost $0.002 for every sampled ended auction, including zero-bid and pre-window boundary rows. Those item fees apply independently, even when the trend-summary fee is zero. Individual rows are off by default here. The main Actor instead charges its $0.02 keyword fee when priced sold auctions are available, even with too few usable months for a direction; its history-only rows are never returned or charged individually.

This Actor returns a trend_summary with monthly sold-price medians, sample counts and a rising / falling / flat direction when usable. Its default budget is 300 matching ended auctions, not 300 sales. Legacy whole-sample statistics and coverage flags have different meanings from the main Actor's history output.

Accuracy update: price bands now apply locally to allowed source pages. A filtered totalListingsFound is an integer only after a complete, consistent scan; otherwise it is null. Unfiltered source totals remain integers. Saved input names, defaults, record types and fee amounts are retained, but consumers must accept this newly explicit unknown filtered total.

Unofficial — not affiliated with Yahoo! Auctions. Reads public pages only.

Move a saved input to the main Actor

Keep keywords, priceMinJpy, priceMaxJpy and convertToUsd. Rename monthsBack to trendMonthsBack, minPerMonth to trendMinSoldPerMonth, and the legacy maxItemsPerKeyword to trendMaxItemsPerKeyword.

In the main Actor, maxItemsPerKeyword separately controls the normal sold-price sample and any returned individual rows. Set it independently from the history budget. For example, this input requests monthly history with a 300-row history budget, a 30-row normal summary budget, and no individual rows:

{
  "keywords": ["Nikon Z6"],
  "includePriceTrend": true,
  "includeIndividualItems": false,
  "maxItemsPerKeyword": 30,
  "trendMonthsBack": 6,
  "trendMaxItemsPerKeyword": 300,
  "trendMinSoldPerMonth": 5,
  "priceMinJpy": 50000,
  "convertToUsd": false
}

This is an input example, not a claim about current market results. The main Actor returns its monthly history inside price_summary.priceTrend. It uses requested-month statistics and acquisition reach, so moving the same keyword and budgets does not make all legacy summary fields equal. Set includeIndividualItems explicitly: the main Actor defaults it to true and returns individual rows from its normal sample, rather than all legacy history rows.

Keep an existing legacy input

The original field names and defaults remain valid:

{
  "keywords": ["Nikon Z6"],
  "monthsBack": 6,
  "maxItemsPerKeyword": 300,
  "minPerMonth": 5,
  "priceMinJpy": 50000,
  "includeIndividualItems": false,
  "convertToUsd": false
}

monthsBack includes the current calendar month in Japan. The requested window can exceed the history available from the source. Each run reads current public ended-auction search pages; this Actor does not maintain a historical price database between runs.

Only priced ended auctions with at least one bid count toward sold-price statistics. A bid is the public listing signal used here, not proof of payment or a completed transaction. Zero-bid auctions still consume the ended-auction sample budget and can appear in optional individual rows.

Sampling, price bands and source reach

Both price bounds are inclusive and applied locally after fetching allowed public pages. They are not sent as source price-filter parameters. Only matching ended auctions consume maxItemsPerKeyword; narrow bands can require more source pages. Collection can stop at the matching-row budget, source exhaustion, or a page before the requested start month when newest-first order is confirmed. It cannot guarantee every requested month is populated. The existing 300-second run timeout still applies; a narrow band over a broad keyword can time out and fail rather than return a false empty result. Use a specific model keyword to keep the source scan focused.

Without price bounds, totalListingsFound is the integer total reported by the source. With either bound, it is the exact matching count only when all source pages have been inspected consistently. It is null after an early stop or incomplete inspection; a bounded sample with no matches does not establish a zero total. Check acquisition.totalListingsFoundExact, sourceExhausted, sourcePagesConsistent and stopReason when interpreting a filtered total. A failed or unreadable source is not treated as a confirmed empty search.

Read the legacy output

Each successful keyword returns one trend_summary. Optional individual records keep type: "item" and the existing fields: id, title, priceJpy, bidCount, condition, isFixedPrice, endTime, month, url, and priceUsd when available.

• months[] covers only the requested months. Each entry has month, soldCount, medianJpy, p25Jpy, p75Jpy and usedInTrend. A month below minPerMonth remains visible but does not affect the direction.
• trendUsable needs at least two usable months. trendDirection and changePct compare the first and last usable monthly medians; an absolute change below 3% is flat. When unusable, the direction and change are null, and trendUnusableReason explains the sample's limitation.
• sampledListings counts all retained ended-auction rows. soldTotal and overallPriceJpy use priced rows with at least one bid across that whole retained sample. They can include rows before the requested window from the boundary page. The sum of monthly sold counts can therefore differ from soldTotal.
• oldestEndTime and newestEndTime describe the retained sample, including zero-bid and pre-window rows. They are not restricted to the requested monthly buckets.
• windowCovered preserves the legacy definition: the earliest requested month containing a priced auction with at least one bid reaches the requested start month. An older boundary row alone does not make it true. This is not the main Actor's acquisition-coverage flag. budgetExhausted means the retained sample reached its configured budget.
• acquisition separately reports actual source inspection: pages and rows read, the unfiltered source total, matching rows observed, duplicate handling, confirmed ordering, stopping reason and observed time reach. Its requestedWindowReached describes observed source reach, independent of legacy windowCovered and monthly sold counts.
• overallPriceUsd and exchangeRateJpyUsd are added when conversion succeeds. checkedAt is the UTC observation time; monthly buckets use the returned Japanese end month. mixedKeywordLikely flags a wide sampled price spread, not a verified product match.

For example, a two-month request retaining an August sale at ¥100,000, a July zero-bid auction and a June boundary sale at ¥1,000 has sampledListings: 3, soldTotal: 2 and an overall median of ¥50,500. Its requested monthly buckets contain only the August sale, and windowCovered is false. With default minPerMonth: 5, no trend fee is due; enabling individual rows still delivers and charges all three rows. This is a synthetic compatibility example, not a market observation.

Pricing

A usable legacy trend costs $0.02 per keyword (trend-analyzed). Fewer than two usable months means no trend-summary fee. Optional ended-auction rows cost $0.002 each (item-scraped), independently of trend usability, and are off by default. No matching rows means no item fee. There is no monthly subscription fee.

Public pages only, no login, requests spaced out. No seller information is collected.

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.
• Legacy trend and item fees are separate. $0.02 only for a usable trend; each optional sampled ended-auction row costs $0.002 even when no trend is usable.
• 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

• Yahoo! Auctions Japan Sold Price Summary & Monthly History
• Yahoo! Auctions Japan Bulk Listings — Live or Ended
• Yahoo! Auctions Japan Deals — Below Past Sold Prices
• Mercari Japan Price Summary — Live or Sold Out
• Daikokuya Japan Buyback — What a Shop Actually Paid
• Hifido Japan Used Audio — What It Actually Sold For
• Japan Used Camera & Lens — Dealer Price vs Sold Price
• Used iPhone & Android Prices Japan — Dealer vs Sold

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 Yahoo! Auctions. 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.