Amazon Book & T-Shirt Niche Analyzer: BSR, Reviews & Search Volume

Give it a niche keyword and get back one Amazon niche report: how often shoppers search it on Amazon, how many products compete, the price, rating and review spread of the top 20 results, the Best Sellers Rank of the top products, how old they are, how many are self-published, a USPTO trademark screen of the phrase, and plain demand / competition labels with the numbers behind them. Built for self-publishers of low-content books (for example on Kindle Direct Publishing), Kindle authors and print-on-demand t-shirt sellers (for example on Merch by Amazon), and for AI assistants (MCP). Pay per niche. Failed niches are never charged.

What's in each report

• Amazon search volume: monthly searches for the phrase on Amazon (US, UK, DE, FR, IT, ES, CA, AU, IN, MX, NL)
• Market size: total results Amazon reports for the search, sponsored products on page 1
• Top 20 results: median / min / max price, average rating, median and max reviews, share with under 100 and under 20 reviews, "bought in past month" figures Amazon shows, Best Seller and Amazon's Choice badges
• Top product pages (10 by default, up to 20): main-category Best Sellers Rank (median, best, how many under 100,000), sub-category ranks, publication / first-available date (median age, new in the last 90 and 365 days), publisher (self-published share for books), median page count
• Trademark screen: the niche phrase checked against live USPTO marks in the classes for your product type (books IC 016/009, shirts IC 025), with risk level and flagged phrases
• Signals: demand (strong / moderate / weak) from median Best Sellers Rank and competition (low / medium / high) from median reviews, each with its basis spelled out
• Products list: rank, ASIN, title, link, price, rating, reviews, BSR, sub-category ranks, author / brand, publisher and date for the top 20

Use cases

• Low-content book research: compare journals, planners and logbooks before you design one
• Print-on-demand t-shirts: check how crowded a t-shirt phrase is and whether it's trademarked
• Kindle niches: see how established the top books are and how many are self-published
• AI agents: one call gives Claude or ChatGPT the numbers to rank a list of niche ideas

Input

Field	What it does
niches	One report per niche keyword (up to 50 per run)
productType	paperback (Books department, default), kindle (Kindle Store), t-shirt (adds "t-shirt" to the search) or any
marketplace	Amazon store: US (default), GB, DE, FR, IT, ES, CA, AU, IN, MX, JP, NL
productsToAnalyze	Product pages read per niche, 3-20 (default 10). Same price for any value
trademarkCheck	USPTO screen of the niche phrase (default on)

{ "niches": ["gardening journal", "dog mom"], "productType": "paperback", "marketplace": "US" }

Output (one row per niche)

{
  "type": "niche", "niche": "gardening journal", "status": "ok", "amazonSearchVolume": 224, "totalResults": 55304,
  "signals": { "demand": "moderate", "competition": "medium", "selfPublishedShare": 0.4, "newcomerShare365d": 0.2 },
  "searchStats": { "priceMedian": 9.99, "reviewsMedian": 120, "shareUnder100Reviews": 0.45, "sponsoredOnPage": 12 },
  "productStats": { "bsrCategory": "Books", "bsrMedian": 180000, "bsrBest": 61631, "pagesMedian": 120 },
  "trademark": { "risk": "none", "classesChecked": ["016", "009"] },
  "products": [ { "rank": 1, "asin": "1441332782", "title": "The Gardener's Logbook", "bestSellersRank": 61631 } ]
}

Values above are illustrative of the shape; every run returns live numbers. Rows with status partial, error or no_results are not charged. A run summary (cost, timing, warnings) is saved as OUTPUT in the key-value store.

How the labels work

Label	Basis	Thresholds
demand	median main-category Best Sellers Rank of the analyzed products	Paperback: strong < 50,000 ≤ moderate < 300,000 ≤ weak. Kindle: 30,000 / 200,000. T-shirts: 100,000 / 1,000,000
competition	median review count of the top 20 organic results	low < 50 ≤ medium < 300 ≤ high

The labels are a starting point, not a forecast. We don't estimate sales or royalties: Amazon doesn't publish them, and BSR-to-sales formulas vary widely, so you get the measured numbers instead.

Pricing

$0.15 per niche report (pay per event, no start fee). A niche is charged only when the Amazon search, the search volume lookup and most product pages came back. Searches with no results are not charged.

• 1 niche = $0.15. 50 niches = $7.50.
• Apify free plan: up to 2 niches per run.
• Your run's maximum charge is respected: the run stops before going over it.

Use from AI agents (MCP)

Add this Actor to Claude, Cursor or any MCP client through Apify's hosted MCP server: https://mcp.apify.com/?tools=vncntlabs/amazon-book-tshirt-niche-analyzer. Ask for example: "Compare these 5 journal niches on Amazon and tell me which has the least competition." A run takes about 3-5 minutes.

FAQ

Where does the data come from? Amazon search results, product pages and Amazon search volumes come through DataForSEO's licensed APIs. This Actor doesn't scrape Amazon or log in anywhere. Trademark data comes from the USPTO's public bulk files.

How fresh is it? Search results and product pages are read live in each run. Search volumes are monthly averages.

Why do some products have no BSR or date? Amazon doesn't show them on every product page, and now and then a product page returns no data. The report says how many pages were read.

Is the trademark screen legal advice? No. It flags live USPTO marks that match words in your niche phrase. Check anything flagged with the USPTO or a lawyer before you publish.

Is this affiliated with Amazon? No. VNCNT Labs is not affiliated with, endorsed or sponsored by Amazon. Amazon, Kindle, Kindle Direct Publishing (KDP) and Merch by Amazon are trademarks of Amazon.com, Inc. or its affiliates, named here only to describe what the tool helps with.

Related tools by VNCNT Labs

• POD Trademark Checker to screen whole listings
• Keyword Search Volume for Google and Amazon volumes in bulk
• Amazon Product Scraper for full product data

Questions or a missing field? Open an issue on the Actor's page or write to hello@vncntlabs.com.