eBay Scraper By Buyer Interest & Seller Feedback

eBay Scraper — Listings, Buyer Interest and Seller Feedback

Extract eBay listings from any search page or direct item link and get them back as structured JSON — title, price, was-price, brand, and category data alongside live watchers count, a derived demand score, and seller feedback count/percentage. Resellers, dropshippers, pricing analysts, and market researchers use this data to rank listings by real buyer interest instead of price alone, and to screen out low-trust sellers before sourcing. Paste your URLs and every field lands in one flat row per listing.

What is eBay Scraper By Buyer Interest & Seller Feedback?

eBay Scraper By Buyer Interest & Seller Feedback is an Apify Actor that turns eBay search-result pages and individual listing pages into structured data rows, adding buyer-demand and seller-trust signals pulled from the same page load. No eBay account, login, or API key is required — the Actor reads public search and listing pages exactly as a visitor's browser would.

• Discovery by URL — paste eBay search/category URLs to cast a wide net, or direct /itm/ listing URLs for a laser-focused pull; mix both freely in one run.
• Full listing identity and pricing — title, subtitle, price, was-price, availability, sold count, brand, EAN/MPN/UPC, and category breadcrumb, in the same pass.
• Live buyer-interest signals — real watchers count plus a derived demandScore that combines watchers and sold count.
• Seller-trust signals — seller feedback count, positive-feedback percentage, and return-policy terms (day window and who pays return shipping).
• Built-in quality filters — minWatchers and minSellerFeedbackPercent drop low-interest or low-trust listings before they ever reach your dataset.
• Structured export — one consistent JSON schema per listing, downloadable as JSON, CSV, or Excel through Apify's standard dataset export.

What data can I extract with eBay Scraper By Buyer Interest & Seller Feedback?

Every run returns one flat row per listing covering identity, pricing, stock, seller, buyer-interest, and trust data:

Field	Example Value	Use Case
url	https://www.ebay.com/itm/195678901234	Canonical link back to the live listing
itemNumber	195678901234	Unique eBay item ID for deduping/joins
title	"Electric Massage Recliner Chair with Heat"	Product identification
subTitle	"Zero Gravity, Bluetooth Speaker, Remote Control"	Extra positioning copy sellers add
whyToBuy	["Free shipping", "Top Rated Plus"]	Seller-highlighted selling points
price	499.99	Numeric price for sorting/filtering
priceWithCurrency	"US $499.99"	Display-ready price
wasPrice	599.99	Pre-discount price for markdown tracking
wasPriceWithCurrency	"US $599.99"	Display-ready was-price
available	12	Remaining stock count
availableText	"More than 10 available"	Raw stock phrasing as shown on eBay
sold	184	Cumulative units sold
image	https://i.ebayimg.com/images/g/abc/s-l1600.jpg	Primary listing photo
seller	"cozyhome-outlet"	Seller username/store handle
itemLocation	"Ontario, CA, United States"	Ships-from location
brand	"Homall"	Structured brand for filtering
type	"Massage Chair"	Product-type item specific
categories	["Home & Garden", "Furniture", "Massage Chairs"]	Breadcrumb category path
ean	"0192837465019"	Barcode identifier, when disclosed
mpn	"MC-2200BLK"	Manufacturer part number, when disclosed
upc	"Does not apply"	UPC, or eBay's own "not applicable" text
watchersCount	47	Live buyer-interest count
watchersText	"47 watchers"	Raw text as rendered on the page
demandScore	225.0	Derived interest index (see below)
sellerFeedbackCount	3512	Total feedback ratings for the seller
sellerFeedbackPositiveText	"99.2% positive"	Raw feedback text
sellerFeedbackPositivePercent	99.2	Clean float for trust-threshold filtering
returnPolicyText	"30 days returns. Buyer pays for return shipping."	Full policy sentence
returnPolicyDays	30	Explicit return window, when stated
returnShippingPaidBy	"Buyer"	Who covers return shipping, when stated
scrapedAt	"2026-07-25T14:02:11.123456+00:00"	Timestamp of extraction

Buyer-interest and seller-trust signals

watchersCount and sellerFeedbackCount/sellerFeedbackPositivePercent are read from the same item-page HTML as everything else — no extra request per listing. Watchers are genuinely conditional data: eBay only renders the watchers widget once a listing crosses its own internal interest threshold, so watchersCount/watchersText come back null on lower-interest listings rather than a faked 0. demandScore is not an eBay metric — it's a value this Actor derives (3 × watchersCount + 1 × sold, weighting live watcher intent above cumulative sales) so listings can be ranked on one number. Use sellerFeedbackPositivePercent together with returnPolicyDays/returnShippingPaidBy to screen out sellers before you commit to sourcing from them.

Targeting and identity data

categories, brand, type, and itemLocation are the fields to segment on: filter a dataset down to one category branch, one brand, or one shipping region after the run, or use them to spot-check that a search URL actually returned the product line you meant. ean/mpn/upc support exact cross-listing/cross-marketplace matching when a seller discloses them, and itemNumber is the stable join key across repeated runs.

How does eBay Scraper By Buyer Interest & Seller Feedback differ from the official eBay API?

eBay's own Browse API can return a watchCount field and seller feedback data via getItem, but the default rate limit is 5,000 calls/day per app, and a higher tier needs eBay's Application Growth Check — a real ceiling on bulk demand-screening for a new app (eBay Developers Program, API Call Limits).

Feature	eBay Browse API	This eBay Scraper
Access requirement	Registered developer app, OAuth token	None — no eBay account or API key
Default call volume	5,000 calls/day per app	maxListings up to 100,000 per run
Watchers + feedback	Separate getItem call per listing	One page fetch per listing, all fields included
Search-page discovery	search returns light summaries; detail needs one call per item	Paste a search URL, every listing is discovered and fetched
Setup	Register app, manage OAuth tokens	Paste URLs and run
Output shape	Raw API JSON, varies by fieldgroup	One flat, consistent JSON row per listing

What's the eBay Browse API's default rate limit?

5,000 calls per day per application, covering getItem — the method carrying watchCount and seller feedback (eBay Developers Program, API Call Limits). Since getItem returns one listing at a time, screening a few thousand listings a day for buyer interest already uses most of that quota, before counting the calls needed to discover those listings. This Actor covers the gap by reading the same public page a buyer would load, one fetch per listing, with no per-day API quota beyond your own Apify usage.

Reach for the official API for a production eBay integration on eBay's own infrastructure. Reach for this Actor for buyer-interest and seller-trust data from a list of URLs, with no developer app to register.

How to use data extracted from eBay?

Resellers and dropshippers

Paste a category or search URL, set minWatchers and minSellerFeedbackPercent to your bar, and the dataset that lands is already screened to "hot and trustworthy" candidates. Sort the result on demandScore and wasPrice vs price to find listings with both live buyer interest and room for a markdown, then cross-check ean/mpn/upc against your own catalog before sourcing. The output drops straight into a spreadsheet or a sourcing pipeline without opening a single listing manually.

Pricing and monitoring agencies

Agencies tracking eBay pricing for clients can rerun the same set of listing or search URLs on a schedule and diff price, sold, and watchersCount between runs to show a client whether a competitor's listings are gaining interest or getting discounted. itemNumber is the stable key for joining rows across runs, so a week-over-week or month-over-month watcher/price trend report doesn't need any manual reconciliation.

Market research and intelligence

Point the Actor at a category's search results and analyze the returned categories, brand, and price fields in bulk to map which sub-categories and brands dominate a search term, or aggregate demandScore and sold by brand to see which manufacturers actually convert. Because search and direct-listing inputs return the same schema, category-wide sampling and single-listing deep dives can feed the same analysis pipeline.

AI agents and automated pipelines

Because output is one flat JSON schema per listing, an agent can call this Actor as a tool to pull live eBay demand data mid-conversation, ground a recommendation in real watchersCount/sellerFeedbackPositivePercent figures, or feed a RAG pipeline with current listing and seller-trust context rather than stale training data.

🔼 Input sample

None of the five input parameters are required — the Actor runs with listingUrls alone.

Parameter	Required	Type	Description	Example Value
listingUrls	No	array (string list)	eBay search/category URLs or direct /itm/ listing URLs; mix both. Also accepts the base urls field.	["https://www.ebay.com/sch/i.html?_nkw=massage+recliner+chair"]
maxListings	No	integer (0–100,000)	Cap on how many listing pages to process after URL discovery; 0 = no limit. Also accepts the base maxItems field.	20
minWatchers	No	integer (0–100,000, default 0)	Keep only listings with at least this many watchers; 0 = no filter.	10
minSellerFeedbackPercent	No	number (0–100, default 0)	Keep only listings from sellers at or above this positive-feedback percentage; 0 = no filter.	98
networkSettings	No	object (proxy editor)	Optional Apify Proxy configuration. Also accepts the base proxyConfiguration field.	{"useApifyProxy": true}

{
  "listingUrls": [
    "https://www.ebay.com/sch/i.html?_nkw=massage+recliner+chair",
    "https://www.ebay.com/itm/195678901234"
  ],
  "maxListings": 50,
  "minWatchers": 10,
  "minSellerFeedbackPercent": 98,
  "networkSettings": { "useApifyProxy": true }
}

Pitfall to avoid: setting minWatchers or minSellerFeedbackPercent above 0 drops any listing where that signal genuinely can't be read — eBay only renders the watchers widget above its own internal interest threshold, so a real but lower-interest listing is excluded the same as one with true zero interest. If your dataset looks thinner than expected, lower the threshold back to 0 and check watchersCount/sellerFeedbackPositivePercent on the unfiltered rows first.

🔽 Output sample

Results are pushed to the Actor's dataset one row per listing as they're scraped, and can be exported as JSON, CSV, or Excel from the Apify run's Storage tab.

{
  "url": "https://www.ebay.com/itm/195678901234",
  "itemNumber": "195678901234",
  "title": "Electric Massage Recliner Chair with Heat and Bluetooth Speaker",
  "subTitle": "Zero Gravity, Full Body, Remote Control",
  "whyToBuy": ["Free shipping", "Top Rated Plus", "30 day returns"],
  "price": 499.99,
  "priceWithCurrency": "US $499.99",
  "wasPrice": 599.99,
  "wasPriceWithCurrency": "US $599.99",
  "available": 12,
  "availableText": "More than 10 available",
  "sold": 184,
  "image": "https://i.ebayimg.com/images/g/abc123/s-l1600.jpg",
  "seller": "cozyhome-outlet",
  "itemLocation": "Ontario, CA, United States",
  "brand": "Homall",
  "type": "Massage Chair",
  "categories": ["Home & Garden", "Furniture", "Massage Chairs"],
  "ean": "0192837465019",
  "mpn": "MC-2200BLK",
  "upc": "Does not apply",
  "watchersCount": 47,
  "watchersText": "47 watchers",
  "demandScore": 325.0,
  "sellerFeedbackCount": 3512,
  "sellerFeedbackPositiveText": "99.2% positive",
  "sellerFeedbackPositivePercent": 99.2,
  "returnPolicyText": "30 days returns. Buyer pays for return shipping.",
  "returnPolicyDays": 30,
  "returnShippingPaidBy": "Buyer",
  "scrapedAt": "2026-07-25T14:02:11.123456+00:00"
}

A listing that fails to load (removed, ended, or blocked mid-run) is still added to the dataset, but as a short placeholder row — {"url", "error": "unavailable", "errorReason", "scrapedAt"} — instead of a full product record, so you can see which URLs need a retry. These placeholder rows, and any listing filtered out by minWatchers/minSellerFeedbackPercent, are not charged; only a fully-extracted, filter-passing row is billed under the row_result event. To export just the billed rows, filter out any record where the error field is present.

How do you filter and target specific listings?

Precision here starts with the input itself: paste a search or category URL to let the Actor discover every listing on the results grid (broad canvas, good for category-wide demand sweeps), or paste direct /itm/ listing URLs when you already know exactly which items to check (laser-focused, no discovery step, faster per listing). The two input types can be mixed in a single listingUrls array.

maxListings is the volume control: it caps how many discovered/pasted listings actually get fetched, so a category sweep can be sampled cheaply (a low cap for a quick read on demand) or run exhaustively (0 = no limit) once you know the URL is worth the full pull.

minWatchers and minSellerFeedbackPercent are the quality thresholds — the two filters that make this Actor a targeting tool rather than a plain page dumper. Set them together to keep only listings that are both in demand and from a trustworthy seller; remember that either filter drops listings where the underlying signal is genuinely absent on the page, not just low.

{ "listingUrls": ["https://www.ebay.com/sch/i.html?_nkw=vintage+camera"], "maxListings": 200 }

{ "listingUrls": ["https://www.ebay.com/itm/195678901234", "https://www.ebay.com/itm/198765432109"] }

{ "listingUrls": ["https://www.ebay.com/sch/i.html?_nkw=graphics+card"], "minWatchers": 25, "minSellerFeedbackPercent": 99 }

▶️ Want to try other eBay scrapers?

Scraper Name	What it extracts
eBay Seller Scraper With Business Seller Info	Seller profile pages: feedback score, store stats (items sold, positive-feedback %, followers, top-rated badge), and the seller's public business disclosure (VAT number, company registration number, registered business name)
Amazon Search Products Scraper: Brand Comparison Details	Amazon search results with brand-level price and listing comparison data
Amazon Bestsellers Scraper with Price Drop Report	Amazon bestseller lists with price-drop tracking over time
Amazon Seller Data Extractor With Business Contact Enrichment	Amazon seller storefronts enriched with business contact details
Alibaba Email Scraper & Trade Assurance Details	Alibaba supplier listings with contact emails and Trade Assurance verification status

How to extract eBay data programmatically

This Actor runs like any Apify Actor: one API call with your input JSON, one auth header, structured dataset JSON back.

Python example

import requests

TOKEN = "YOUR_APIFY_API_TOKEN"
ACTOR = "SimpleAPI~ebay-scraper-by-buyer-interest-seller-feedback"

run = requests.post(
    f"https://api.apify.com/v2/acts/{ACTOR}/run-sync-get-dataset-items",
    params={"token": TOKEN},
    json={
        "listingUrls": ["https://www.ebay.com/sch/i.html?_nkw=drone"],
        "minWatchers": 10,
        "minSellerFeedbackPercent": 98,
    },
)

for row in run.json():
    print(row["title"], row["watchersCount"], row["sellerFeedbackPositivePercent"])

MCP for AI agents

Apify Actors, including this one, are reachable through Apify's own Actors MCP Server without any actor-specific code:

$npx -y @apify/actors-mcp-server --actors SimpleAPI/ebay-scraper-by-buyer-interest-seller-feedback

Point an MCP-compatible client — Claude Desktop, Claude Code, or Cursor — at this server (with an APIFY_TOKEN in the environment) and an agent can call the Actor as a tool to pull live watcher counts and seller feedback for a product line mid-conversation, without a custom integration.

Export to spreadsheets or CRM

From the run's Storage tab, export the dataset as CSV and map columns directly: title → product name, price/wasPrice → pricing columns, watchersCount/demandScore → a lead-scoring or priority column, seller/sellerFeedbackPositivePercent → a supplier-vetting column in your CRM or sourcing sheet.

Is it legal to scrape eBay listings?

Yes — scraping publicly listed eBay pages for business and product data (prices, stock, categories, seller store handles) is generally permitted, since these records exist precisely to be found by shoppers and search engines. This Actor does not collect private buyer information; seller is a public store/username, not a private individual's personal details, so this falls under standard ToS and database-rights considerations rather than GDPR/CCPA personal-data rules. Always review eBay's own Terms of Service before large-scale or commercial use, and consult legal counsel for commercial applications involving bulk storage of any personal data you do encounter.

❓ FAQ

Are removed or ended eBay listings flagged in the output?

Yes — if a listing page can't be fully read (ended, removed, or blocked by eBay's bot defenses), the Actor still adds a row for that URL, but as a short placeholder: {"error": "unavailable", "errorReason": ...} instead of the full product record, so you can see exactly which URLs to retry or drop.

Can I get watchers and seller feedback along with the main listing data?

Yes, in the same row — watchersCount, sellerFeedbackCount, and sellerFeedbackPositivePercent are parsed from the same page fetch used for title/price/stock, with no extra request per listing and no pagination caveat beyond maxListings.

How accurate is the watcher and seller feedback data?

It's exactly what eBay renders on the page at request time. watchersCount is null (never a faked 0) on listings below eBay's own display threshold, and feedback percentage reflects the seller's current public score — both change as buyers interact with the listing, so treat each run as a point-in-time snapshot.

How many listings can I get per run?

maxListings accepts 0–100,000, where 0 means no cap and the Actor processes every listing discovered from your input URLs; the schema's prefill is 20 for quick test runs.

How does the minWatchers / minSellerFeedbackPercent filter work?

Both are threshold filters applied after each listing is scraped: a listing is only saved to the dataset if its watchersCount and sellerFeedbackPositivePercent both meet or exceed the values you set (default 0 = no filtering). Listings where the underlying signal can't be read are treated as failing the bar, not passing it.

Does this scraper work with Claude, ChatGPT, and AI agent frameworks?

Yes — it's callable as an HTTP endpoint by any agent framework via the Apify API, and it's also reachable through Apify's Actors MCP Server (npx -y @apify/actors-mcp-server --actors SimpleAPI/ebay-scraper-by-buyer-interest-seller-feedback) for MCP-compatible clients like Claude Desktop and Claude Code.

Can I use this scraper without an eBay API key or developer account?

Yes — you need an Apify account to run the Actor, but no eBay account, login, or developer API key at all. Just paste eBay search or listing URLs.

How is buyer interest actually measured?

Through two real signals read straight from the listing page — watchersCount (live watch-list count, only shown by eBay once a listing crosses its own interest threshold) and sold (cumulative units sold) — combined into a derived demandScore (3 × watchersCount + 1 × sold). demandScore is this Actor's own weighting, not a metric eBay publishes itself, built so listings can be ranked on one number instead of eyeballing both fields separately.

Conclusion

eBay Scraper By Buyer Interest & Seller Feedback turns eBay search results and listing pages into structured JSON that already answers the two questions price alone can't: is anyone actually interested, and can this seller be trusted. Resellers, pricing agencies, and market researchers get watchers, demand score, and seller feedback in the same row as price and stock, with quality filters to keep only the listings worth a closer look. Paste your search or listing URLs into the Actor on Apify and run it to get your first structured dataset back in minutes.