Pinterest Search Scraper: Product Page Details Extractor

Pinterest Search Scraper — Extract Pins, Links and Product Data

Pinterest Search Scraper: Product Page Details Extractor turns a Pinterest keyword search into structured JSON: every pin's outbound destination link and domain, a per-run link summary (unique domains, counts, top destinations), and — for shopping pins — best-effort product data (price, currency, brand, availability) parsed from the pin's own structured data. Unlike scraping frameworks that return raw HTML, this scraper returns typed JSON — ready for your model, your database, or your pipeline without any parsing. This guide covers every input and output field plus three concrete deployment strategies for turning search-based pin data into a running pipeline.

🧭 What Does Pinterest Search Scraper: Product Page Details Extractor Do?

This Actor searches Pinterest for a keyword (e.g. "nike shoes") and streams back every pin found, including its outbound destination link, normalized domain, and — when the pin looks like a shopping pin — parsed price, currency, brand, and availability from the pin's closeup page. It runs logged-out; no Pinterest account, login, or cookies are required. At the end of a run it appends one aggregated link-summary row and mirrors the same data to a key-value store record.

• Keyword-based pin search with live, response-driven streaming (no wait for the whole run to finish before rows appear)
• Destination-link extraction — link/domain are already present in Pinterest's own search payload, at zero extra request cost
• Domain include/exclude filtering (domainFilter, excludedDomains) and a "linked pins only" mode (includeOnlyLinkedPins)
• Best-effort product data for shopping pins: price, currency, brand, availability, product name — parsed from application/ld+json @type=Product
• A derived linkSummary row: unique domain count, per-domain pin counts, top destinations, linked/product coverage fractions
• Video URL harvest for video pins (videoUrls)
• Automatic retry with a fresh browser session (up to 3 attempts) if a session gets blocked

⚡ Features & Capabilities

The Actor combines real-time pin search with two layers of enrichment — link intelligence and product-data extraction — that most keyword-search scrapers skip entirely.

Core features

• Live keyword search — searches pinterest.com/search/pins/ for searchTerm and pushes pins to the dataset as Pinterest's own internal /resource/BaseSearchResource/get/ responses arrive, not after a fixed scroll budget
• Destination link + domain on every row — hasLink, destinationUrl, destinationDomain are populated straight from the search payload for every pin, whether or not product data is requested
• Domain filtering — domainFilter (substring allow-list) and excludedDomains (substring block-list) apply before a row is counted against maxPins
• Product data (best-effort) — for pins that look like shopping pins (is_shopping_ad, is_eligible_for_pdp, rich_metadata.products, or simply carrying an outbound link), the Actor fetches the pin's closeup page in the same browser session and parses the ld+json Product block for productName, productPrice, productCurrency, productBrand, productAvailability
• Link summary row — one rowType: "linkSummary" row per run with totalPins, linkedPinCount, linkedPinFraction, uniqueDomainCount, topDomains, domainCounts, productPinCount, productPinsWithPrice, productLookupsPerformed; also written to the LINK_SUMMARY key-value store record
• Video pin support — pinType: "videos" filters to video pins only, and videoUrls harvests variant video URLs from videos.video_list
• Backward-compatible input — base-actor keys query, filter, and limit are still accepted as fallbacks for searchTerm, pinType, and maxPins
• Manual stealth + Residential proxy support — a stealth init script masks the most-fingerprinted automation tells (navigator.webdriver, plugins, languages), and the Actor retries with a fresh session up to 3 times before giving up

Pinterest Search Scraper: Product Page Details Extractor within the Scrapio data stack

This Actor covers pins discovered through a Pinterest keyword search, with destination-link and product-data enrichment. For pins pulled from a specific Pinterest board or profile (with board metadata and owner enrichment attached), use Pinterest Pins Videos Search Scraper With Board Collection instead — it fetches a board/profile's real pins rather than treating the URL as a search term.

Why do developers and data teams scrape Pinterest?

Pinterest search results sit at the intersection of visual discovery and commerce — a large share of shopping-intent pins link straight to a retailer's product page. That makes the destination link, domain, and product fields this Actor extracts unusually direct signals for several distinct audiences.

🏢 E-commerce, affiliate and dropshipping research

Affiliate marketers and product researchers run a category keyword (e.g. "leather sofa") through the Actor, then filter on isProductPin and productPrice to build a live shortlist of trending shopping pins with real price/brand data. destinationDomain shows exactly which retailers are winning that keyword's Pinterest traffic, and domainFilter lets a researcher restrict a run to competitor or partner domains (e.g. ["amazon.com", "etsy.com"]) before it's even pulled. The output lands as one row per pin, ready to load into a spreadsheet or a product-tracking database without any HTML parsing.

📊 AI training data and RAG indexing

title, description, and the product fields (productName, productBrand, productPrice) are the high-information text and structured-data fields for this dataset. For RAG, indexing title + description alongside destinationDomain lets an agent answer "what products are trending for X on Pinterest and where do they link" without re-scraping. For training data, the product fields return as consistent typed primitives (string, float, string) across every row where a lookup succeeded, which keeps a training set free of ad-hoc HTML cleanup.

📱 Competitive and market intelligence

Run the same searchTerm on a schedule and track domainCounts from the linkSummary row over time — a rising count for a specific competitor domain signals they're gaining Pinterest visibility (organic or paid) for that keyword. Tracking productPrice across runs for pins tied to a known competitor domain also surfaces price changes without visiting the retailer's site directly.

🔬 Research and academic use

Researchers studying visual-commerce behavior, platform-to-retailer traffic patterns, or e-commerce link ecosystems can use destinationDomain and linkedPinFraction as a public, reproducible dataset. This Actor only returns data already visible to a logged-out visitor — no private account data, no login-gated content.

🎥 Product and SaaS development

Teams building price-monitoring tools, affiliate-link directories, or social-commerce dashboards can run this Actor as the ingestion layer: destinationUrl feeds a link-resolution service, productPrice/productCurrency feed a price-history table, and the linkSummary row feeds a per-keyword dashboard tile without any additional aggregation code.

🍚 Input Parameters

All parameters are read from .actor/input_schema.json. Only searchTerm is required.

Parameter	Required	Type	Description	Example Value
searchTerm	Yes	string	Keyword to search on Pinterest. Shopping-style queries work best.	"nike shoes"
pinType	No	string (enum)	"all" or "videos". Default "all".	"all"
maxPins	No	integer	How many pins to extract (1–10,000). Default 100.	100
includeOnlyLinkedPins	No	boolean	Keep only pins with an outbound link. Default false.	false
domainFilter	No	array of string	Keep only pins whose destination domain contains one of these (substring match). Default [].	["amazon.com", "etsy.com"]
excludedDomains	No	array of string	Drop pins whose destination domain contains one of these. Default [].	["pinterest.com"]
includeLinkSummary	No	boolean	Append the linkSummary row (and LINK_SUMMARY KV record). Default true.	true
includeProductData	No	boolean	Fetch and parse product price/brand/availability for likely product pins. Default true.	true
maxProductLookups	No	integer	Cap on closeup product-detail fetches per run (0 = unlimited). Default 30.	30
proxyConfiguration	No	object	Apify proxy configuration. Residential strongly recommended. Default {"useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"]}.	{"useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"]}

Example input:

{
  "searchTerm": "nike shoes",
  "pinType": "all",
  "maxPins": 100,
  "includeOnlyLinkedPins": false,
  "domainFilter": ["amazon.com", "etsy.com"],
  "excludedDomains": ["pinterest.com"],
  "includeLinkSummary": true,
  "includeProductData": true,
  "maxProductLookups": 30,
  "proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Supported URL types and input formats

This Actor takes a keyword, not a Pinterest URL — searchTerm is used to build a pinterest.com/search/pins/?q=... request internally. There is no direct pin, board, or profile URL input on this variant (for board/profile URLs, see the related board-collection scraper below). For backward compatibility, the base actor's raw keys query, filter, and limit are also accepted and take priority over searchTerm/pinType/maxPins when present in the input JSON.

• Broad category search, all domains: {"searchTerm": "leather sofa", "maxPins": 200}
• Video pins only, filtered to two marketplaces: {"searchTerm": "skincare routine", "pinType": "videos", "domainFilter": ["amazon.com", "sephora.com"]}
• Linked pins only, no product lookups (fast run): {"searchTerm": "home office desk", "includeOnlyLinkedPins": true, "includeProductData": false}

📦 Output Format

Output is typed, normalized JSON pushed live to the dataset — one row per pin plus one aggregated summary row per run. Export directly from the Apify platform (JSON, CSV, Excel, or via the API) with no post-processing required.

Output for pins

{
  "rowType": "pin",
  "query": "nike shoes",
  "node_id": "1234567890123456789",
  "type": "pin",
  "title": "Nike Air Max 90 - White/Black",
  "description": "Classic Nike Air Max 90 sneakers in white and black colorway.",
  "domain": "nike.com",
  "link": "https://www.nike.com/t/air-max-90-mens-shoes-6n3vDV",
  "dominant_color": "#f2f2f2",
  "created_at": "Tue, 04 Mar 2025 18:22:10 +0000",
  "board": { "id": "987654321", "name": "Sneakers", "url": "/username/sneakers/" },
  "reaction_counts": { "1": 42 },
  "native_creator": {},
  "pinner": { "id": "555000111", "username": "sneakerhead", "full_name": "Sneaker Head" },
  "image_url": "https://i.pinimg.com/originals/aa/bb/cc/aabbcc.jpg",
  "image_width": 1000,
  "image_height": 1500,
  "pinUrl": "https://www.pinterest.com/pin/1234567890123456789/",
  "destinationUrl": "https://www.nike.com/t/air-max-90-mens-shoes-6n3vDV",
  "destinationDomain": "nike.com",
  "hasLink": true,
  "videoUrls": null,
  "isProductPin": true,
  "productName": "Nike Air Max 90",
  "productPrice": 129.99,
  "productCurrency": "USD",
  "productBrand": "Nike",
  "productAvailability": "InStock",
  "scrapedAt": "2026-07-25T09:14:02Z"
}

Output for link summary

{
  "rowType": "linkSummary",
  "query": "nike shoes",
  "totalPins": 100,
  "linkedPinCount": 72,
  "linkedPinFraction": 0.72,
  "uniqueDomainCount": 18,
  "topDomains": [
    { "domain": "nike.com", "count": 21 },
    { "domain": "amazon.com", "count": 9 },
    { "domain": "footlocker.com", "count": 6 }
  ],
  "domainCounts": { "nike.com": 21, "amazon.com": 9, "footlocker.com": 6 },
  "productPinCount": 14,
  "productPinsWithPrice": 11,
  "productLookupsPerformed": 30,
  "scrapedAt": "2026-07-25T09:14:44Z"
}

Schema stability and export options

Field names stay stable across runs — every pin row carries the full set of fields above regardless of whether a value was found; unmatched product fields return null rather than being omitted, so downstream code can rely on a consistent key set. If Pinterest changes its internal payload shape, new upstream fields are additive; existing keys documented here keep their meaning. Results can be exported from the Apify platform as JSON, CSV, Excel, or fetched programmatically via the Apify API/SDK — no delivery mechanism beyond the standard Apify dataset/key-value store is built into this Actor.

💡 Pinterest Search Scraper: Product Page Details Extractor Strategy Guide

🎯 Strategy 1: Real-time enrichment pipeline

Trigger a run per incoming keyword (e.g. a new product category added to your catalog), then read the dataset as rows stream in rather than waiting for run completion. For each row, append destinationUrl, destinationDomain, and — when isProductPin is true — productPrice/productBrand/productAvailability to your source record. Because rows are pushed live via Actor.push_data, a consumer polling the dataset gets partial results within seconds of the run starting, not only at the end.

🎯 Strategy 2: Scheduled monitoring and alerting

Schedule the same searchTerm to run daily or weekly using an Apify Schedule. After each run, diff the new linkSummary row's domainCounts against the previous run's — a domain's count jumping or a new domain entering topDomains signals a shift in who is winning that keyword's Pinterest traffic. For product pins already seen before (matched on destinationUrl), diff productPrice between runs to alert on price changes.

🎯 Strategy 3: Bulk dataset build

For a research or training dataset, iterate a list of keywords, calling the Actor once per keyword (each call is a separate, independent run), and aggregate every run's dataset into one CSV or a shared database table keyed on node_id. maxProductLookups bounds the extra product-page fetches per run, keeping cost and runtime predictable across a large keyword list; set it to 0 only if you need every product-like pin looked up.

Strategy comparison at a glance

Strategy	Best for	Run pattern	Output format
Real-time enrichment	Enriching inbound records as they arrive	One run per record, read live	Streamed JSON rows
Scheduled monitoring	Tracking domain/price shifts over time	Recurring scheduled runs, diffed	JSON rows + linkSummary diff
Bulk dataset build	Research or training datasets	One run per keyword, aggregated	CSV / database table

🌴 Related Pinterest Scrapers & Tools

Scraper Name	What it extracts
Pinterest Pins Videos Search Scraper With Board Collection	Real pins from a Pinterest board or profile URL, with board metadata and optional owner enrichment
Amazon Search Products Scraper	Amazon search-result products (ASIN, price, rating, review count, image) by keyword
Amazon Bestsellers Scraper By Category	Amazon Best Sellers products plus the full category/browse-node tree
eBay Scraper: Sold Listings & Price Trend Analytics	eBay sold/completed listings with a derived price-trend summary (average, min/max, per-condition)
eBay Seller Scraper — Full Listings, Specifics & Shipping	Full per-seller inventory with item condition, shipping, specifics and description

How to integrate Pinterest Search Scraper: Product Page Details Extractor with your stack

Pinterest Search Scraper: Product Page Details Extractor works with any language or tool that can make an HTTP request to the Apify API. The examples below use the official Apify SDKs.

Python

import csv
from apify_client import ApifyClient

client = ApifyClient("<APIFY_API_TOKEN>")

keywords = ["nike shoes", "leather sofa", "skincare"]

rows = []
for term in keywords:
    run_input = {
        "searchTerm": term,
        "maxPins": 100,
        "includeProductData": True,
        "maxProductLookups": 30,
        "proxyConfiguration": {"useApifyProxy": True, "apifyProxyGroups": ["RESIDENTIAL"]},
    }
    run = client.actor("Scrapio/pinterest-search-scraper-product-page-details-extractor").call(run_input=run_input)
    for item in client.dataset(run["defaultDatasetId"]).iterate_items():
        if item.get("rowType") == "pin":
            rows.append(item)
        elif item.get("rowType") == "linkSummary":
            print(f"{term} -> {item['uniqueDomainCount']} unique domains, {item['productPinCount']} product pins")

with open("pinterest_pins.csv", "w", newline="", encoding="utf-8") as f:
    writer = csv.DictWriter(f, fieldnames=list(rows[0].keys()) if rows else [])
    writer.writeheader()
    writer.writerows(rows)

Node.js

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: '<APIFY_API_TOKEN>' });

const run = await client.actor('Scrapio/pinterest-search-scraper-product-page-details-extractor').call({
    searchTerm: 'nike shoes',
    maxPins: 100,
    includeProductData: true,
    domainFilter: ['amazon.com', 'etsy.com'],
});

const { items } = await client.dataset(run.defaultDatasetId).listItems();
const pins = items.filter((row) => row.rowType === 'pin');
console.log(`Fetched ${pins.length} pins, ${pins.filter((p) => p.isProductPin).length} with product data`);

Async and scheduled pipelines

Runs are asynchronous by default — call() blocks until the run finishes, or use client.actor(...).start() to kick off a run and poll client.run(runId).get()/waitForFinish() from your own process. For recurring extraction, use an Apify Schedule against this Actor rather than a webhook — combine it with the dataset's iterate_items()/listItems() to pull only new rows since the last check.

🎯 Who Needs Pinterest Search Scraper: Product Page Details Extractor? (Use Cases & Industries)

🏢 E-commerce, affiliate and dropshipping research

An affiliate site owner runs searchTerm: "standing desk" weekly, keeps only rows with isProductPin: true, and builds a comparison table from productName, productPrice, and destinationUrl — refreshed without manually re-checking each retailer's page.

📊 AI/ML and RAG teams

A RAG pipeline ingests title + description + destinationDomain per pin so an assistant can answer "what's trending on Pinterest for skincare and which brands/sites show up" using only structured Pinterest data, with no HTML to strip first.

📱 Marketing and competitive intelligence

A brand's marketing team tracks domainCounts for their own domain and named competitors across weekly runs of their category keywords, flagging when a competitor's share of pins for that keyword rises.

🔬 Researchers

Academic and market researchers studying platform-to-retailer traffic patterns use destinationDomain and linkedPinFraction as a public, reproducible signal of which retailers dominate a given Pinterest search topic — scoped to data already visible to a logged-out visitor.

🎥 Product and SaaS builders

Teams building a price-monitoring or social-commerce dashboard product use this Actor as their Pinterest ingestion layer, feeding productPrice/productCurrency into a price-history table and the linkSummary row into a per-keyword overview tile.

Is it legal to scrape Pinterest?

Scraping publicly accessible web data is generally lawful in the United States — see hiQ Labs v. LinkedIn (9th Cir. 2019, reaffirmed 2022), which held that scraping data not gated behind a login does not violate the Computer Fraud and Abuse Act. This Actor runs logged-out and only reads pin data a visitor could see without signing in.

Separately, Pinterest's Terms of Service restrict automated access; violating a platform's ToS is a contract/civil matter between the user and the platform, not a criminal one, but it can still carry consequences such as IP blocking or account action. This Actor returns pin, link, and product-page data — not private user account data — so data-protection (GDPR/CCPA) obligations generally do not attach to its output; they can still apply if a pinner or native_creator object in the output is treated as personal data in your jurisdiction.

Pinterest Search Scraper: Product Page Details Extractor returns only publicly accessible data. What you do with that data is your responsibility — consult legal counsel for commercial applications involving personal data.

❓ Frequently asked questions

Does Pinterest Search Scraper: Product Page Details Extractor work without a Pinterest account?

Yes. The Actor runs entirely logged-out — no Pinterest account, login, or cookies are required or used at any point in the run.

How does the Actor handle Pinterest's anti-scraping measures?

It launches a fresh headless Chromium session per attempt with a manual stealth init script (masking navigator.webdriver, plugin list, and languages), rotates user agents, and retries up to 3 times with exponential backoff if a session returns an HTTP error or gets redirected to a login wall. Apify Residential proxy is strongly recommended — Pinterest commonly blocks datacenter IPs outright.

Can I run it at scale without getting blocked?

Reliability depends heavily on using Residential proxy; without one, the Actor logs a warning and runs are far more likely to return zero pins. There is no published uptime or success-rate figure for this Actor — treat proxy configuration as the primary lever you control.

How fresh is the data this Actor returns?

Every run performs a live fetch against Pinterest's search endpoint at the time it runs; nothing is cached or served from a prior run. scrapedAt on every row is the actual timestamp of that fetch.

Which fields work best for AI training and RAG indexing?

For RAG, index title and description (the free-text fields) alongside destinationDomain for retrieval context. For training data, productName, productBrand, productPrice, and productCurrency are the most consistently typed fields across rows where a product lookup succeeded — all return as plain strings/numbers with null (never a placeholder) where no product data was found.

Does product data cover every pin?

No — it's best-effort and limited to pins that look like shopping pins. Non-product pins always return null for productName/productPrice/productCurrency/productBrand/productAvailability, and isProductPin is null for pins never looked up (because includeProductData was off, or maxProductLookups was already hit). The run log reports the real product-coverage fraction rather than assuming full coverage.

Does this Actor work with Claude, ChatGPT, and other AI agent tools?

It's callable as an HTTP endpoint via the Apify API by any agent framework that can make a REST call — there is no dedicated MCP server for this Actor. Every response is typed JSON, so an agent can consume dataset rows directly without an HTML-parsing step.

Am I charged for every row in the dataset?

Yes for this Actor — every pin row and the linkSummary row are pushed with charged_event_name="row_result", so each row that reaches the dataset is one billed row_result event. There are no separate uncharged diagnostic rows; a pin that fails mid-processing is skipped and logged rather than pushed.

What happens if Pinterest returns no results or blocks the session?

The Actor retries up to 3 times with a fresh browser session and backoff. If every attempt still pushes zero pins, the run ends with an explanatory status message (session likely blocked, search had no results, or the response format changed) rather than silently succeeding with an empty dataset.

ℹ️ Disclaimer

Pinterest Search Scraper: Product Page Details Extractor extracts only publicly available data from Pinterest. This tool is intended for lawful use cases only. Users are responsible for complying with Pinterest's terms of service and applicable data protection laws in their jurisdiction.