📌 Pinterest Scraper – Extract Pins, Boards, Profiles & Video Pin Data

The Pinterest Scraper extracts pins, boards, profiles and video pins from Pinterest search results and returns them as structured JSON. Give it a keyword or a Pinterest search URL, choose which result type you want, and it collects the pin content, creator details, board references, media assets and Pinterest's own metadata into one dataset you can analyse, store or feed into a downstream pipeline.

Pinterest is a visual discovery engine with strong commercial intent — people search it to plan purchases, projects and events. That makes Pinterest data valuable for trend research, competitor monitoring, influencer discovery and content strategy. This Pinterest scraper turns that discovery surface into structured records, with optional lightweight content analysis, sentiment scoring and content categorisation attached to each pin.

---

📊 What Data Can You Extract with This Pinterest Scraper?

Records are shaped by the result type you request. Across pins, boards and profiles, the scraper returns the following field groups.

Category	Fields	What it gives you
🧭 Record identity	type, entity_type, schema, id, url	What kind of record it is, its Pinterest ID, canonical URL, and the output schema version
📌 Pin content	title, pin	The pin title plus the nested pin object containing description and pin-level detail
👤 Creator data	creator	The nested creator object with username, full name and follower count
🗂️ Board context	board_ref, board	The board a pin belongs to, and full board records when scraping the boards result type
🖼️ Media assets	media	Image variants with their URLs and dimensions for each pin
🔍 Provenance	source_url, seed_type, seed_value	Exactly which search URL or keyword produced the record, so results stay traceable
😊 Sentiment analysis	sentiment_score, sentiment_score_normalized, sentiment_confidence, sentiment_label	Heuristic sentiment scoring of the pin description when enabled
🏷️ Content classification	content_category_label, content_category_path, content_category_confidence, content_category_match_type	Category assignment with a hierarchical path and match confidence
📦 Raw extras	extra, profile	Additional Pinterest metadata, and full profile records for the profiles result type

The seed_type and seed_value pair is quietly one of the most useful things in the output. When you run a dozen keywords in one job, every record still knows which keyword or URL produced it — so you can group, compare and rank results per query without keeping a separate mapping.

---

🌟 Key Features of the Pinterest Scraper

Feature	Description
🔎 Keyword and URL input	Feed plain keyword strings via queries, Pinterest search URLs via startUrls, or both in the same run
🗃️ Four result types	Switch between all-pins, videos, boards and profiles with a single type setting
🖼️ Full media variants	The media object carries image variants with width, height and direct URLs for each pin
🗂️ Board enrichment	enrich_boards fetches full board details — owner, preview images, cover images — so board_ref.extra is complete
🏷️ Content analysis	Optional content_analysis attaches hashtags, extracted links, word count and a has_price flag to each pin
😊 Sentiment scoring	Optional heuristic sentiment on pin descriptions, returned as a score, normalised score, confidence and label
🍪 Cookie support	Supply a Pinterest cookie header from a logged-in session to improve reliability and result depth
🎚️ Per-search result limits	limit caps how many records are saved per keyword search, so validation runs stay small and cheap
🔄 Automatic proxy handling	Proxy rotation is handled internally — there is no proxy field to configure

---

🚀 Why Choose This Pinterest Scraper?

One actor covers four Pinterest entity types. Rather than stitching together separate tools for pins, video pins, boards and creator profiles, the type field switches the whole run. The record shape adapts — pins carry pin and media, boards carry board, profiles carry profile — while identity and provenance fields stay consistent across all of them.

Enrichment is optional and explicit. enrich_boards, content_analysis and sentinent_analysis are independent toggles. You decide whether the extra requests and processing are worth it for a given job, rather than paying for analysis you will not read.

Every record is traceable to its seed. source_url, seed_type and seed_value mean a multi-keyword run produces a single dataset you can still segment perfectly by query. That matters for competitive research where the comparison between keywords is the whole point.

Structured media, not just a thumbnail URL. The media object exposes image variants with their dimensions, so you can pick the right resolution for a moodboard, a report or a downstream image pipeline without guessing at URL patterns.

---

📥 Input

{
  "startUrls": [
    { "url": "https://www.pinterest.com/search/pins/?q=home%20decor" }
  ],
  "queries": ["home decor", "scandinavian interior"],
  "type": "all-pins",
  "limit": 500,
  "enrich_boards": true,
  "content_analysis": false,
  "sentinent_analysis": false
}

🔧 Pinterest Scraper Input Fields

Field	Type	Required	Default	Description
startUrls	array	No	[{"url": "https://www.pinterest.com/search/pins/?q=home%20decor"}]	Pinterest search URLs. Only URLs containing a ?q= query are usable
queries	array	No	["home decor"]	Plain keyword strings; each is expanded into a Pinterest search
type	string	No	"all-pins"	What kind of results to collect. One of all-pins, videos, boards, profiles
limit	integer	No	50000	Maximum number of records to save for each keyword search. Use a smaller number for fast validation
enrich_boards	boolean	No	true	Fetch full board details (owner, preview images, cover_images) so board_ref.extra is complete. Adds one cached request per unique board
content_analysis	boolean	No	false	Attach a lightweight content_analysis object (hashtags, extracted links, word count, has_price) to each pin
sentinent_analysis	boolean	No	false	Attach a lightweight heuristic sentiment score for the pin description
cookies	string	No	—	Optional Pinterest cookie header from a logged-in browser session. Improves reliability and result depth

Result type options

Value	Meaning
all-pins	All pins — visual posts
videos	Video pins
boards	Boards — pin collections
profiles	Profiles — creator pages

Note that sentinent_analysis is spelled as shown; the key name was kept as provided in the original input schema.

💡 Input Examples

Quick keyword validation run

{
  "queries": ["minimalist kitchen"],
  "type": "all-pins",
  "limit": 50,
  "enrich_boards": false
}

Creator discovery via profiles

{
  "queries": ["food photography", "recipe developer"],
  "type": "profiles",
  "limit": 200
}

Full analysis run on video pins

{
  "startUrls": [
    { "url": "https://www.pinterest.com/search/pins/?q=diy%20furniture" }
  ],
  "type": "videos",
  "limit": 1000,
  "content_analysis": true,
  "sentinent_analysis": true,
  "enrich_boards": true
}

---

📤 Output

A real pin record from an actual run:

{
  "type": "pin",
  "schema": { "version": 2, "entity_type": "pin" },
  "entity_type": "pin",
  "id": "225039312627881892",
  "url": "https://www.pinterest.com/pin/225039312627881892/",
  "title": "Cozy Reading Nook",
  "source_url": "https://www.pinterest.com/search/pins/?q=home+decor",
  "seed_type": "url",
  "seed_value": "https://www.pinterest.com/search/pins/?q=home+decor",
  "pin": { "title": "Cozy Reading Nook", "description": "Neva Round Daybed in Recycled Faux Fur color: Pampas ..." },
  "creator": { "id": "225039450038600465", "username": "erinturnage", "full_name": "Erin Turnage", "follower_count": "..." },
  "board_ref": { "id": "225039381319533959", "name": "Reading Room Inspo", "url": "https://www.pinterest.com/erinturnage/..." },
  "media": { "images": { "thumb": { "width": 236, "height": 445, "url": "https://i.pinimg.com/236x/2c/24/91/2c2491..." } } },
  "extra": { "node_id": "UGluOjIyNTAzOTMxMjYyNzg4MTg5Mg==", "is_eligible_for_filters": false }
}

🧾 Pinterest Output Fields — Core Record

Field	Type	Description
type	string	Record type, e.g. pin
schema	object	Output schema version and entity type
entity_type	string	The Pinterest entity this record represents
id	string	Unique Pinterest identifier for the item
url	string	Canonical URL of the scraped item
title	string	Title of the item
source_url	string	The search URL the item was scraped from
seed_type	string	Whether the record came from a URL seed or a keyword seed
seed_value	string	The exact URL or keyword that produced the record
pin	object	Nested pin object with description and pin-level detail
creator	object	Nested creator object with username, full name and follower count
board_ref	object	Reference to the board the pin belongs to
media	object	Image variants with URLs and dimensions
extra	object	Additional Pinterest metadata for the item
board	object	Full board record, returned for the boards result type
profile	object	Full profile record, returned for the profiles result type

🧾 Pinterest Output Fields — Optional Analysis

Field	Type	Description
sentiment_score	number	Raw heuristic sentiment score for the pin description
sentiment_score_normalized	number	Normalised sentiment score
sentiment_confidence	number	Confidence in the sentiment assessment
sentiment_label	string	Human-readable sentiment label
content_category_label	string	Assigned content category
content_category_path	array	Hierarchical category path
content_category_confidence	number	Confidence in the category assignment
content_category_match_type	string	How the category match was made

---

💻 How to Use the Pinterest Scraper (Step by Step)

Step 1: Choose keywords or search URLs

You can drive the Pinterest scraper two ways. queries takes plain keyword strings and expands each into a Pinterest search — the simplest option and the one most people should start with. startUrls takes Pinterest search URLs directly, which is useful when you have already refined a search in the browser and want to reproduce it exactly. Only URLs containing a ?q= query parameter are usable; board and profile URLs pasted directly will not work as seeds.

Step 2: Pick the result type

type decides the shape of everything that follows. all-pins is the default and returns visual posts. videos narrows to video pins, which matters if you are studying motion content performance. boards returns pin collections, useful for understanding how people group and curate a topic. profiles returns creator pages, which is the starting point for influencer discovery. Run separate jobs if you need more than one type.

Step 3: Set a sensible limit

limit defaults to 50000 records per keyword search, which is a ceiling rather than a target. For your first run set it to something small — 50 or 100 — so you can inspect the record shape and confirm the keyword returns what you expect before committing to a large job. Raise it once the output looks right.

Step 4: Decide on board enrichment

enrich_boards is on by default and fetches full board details so that board_ref.extra contains the owner, preview images and cover images. It adds one cached request per unique board, so on a run that touches thousands of distinct boards it has a real cost. Switch it off for fast, cheap harvests where the board name and URL alone are enough.

Step 5: Add content and sentiment analysis if you need it

content_analysis attaches hashtags, extracted links, word count and a has_price flag to each pin — the has_price signal is a quick way to separate commercial pins from purely inspirational ones. sentinent_analysis adds a heuristic sentiment score for the description. Both default to off. Turn them on when you are doing content strategy or brand-perception work, and leave them off for straightforward data collection.

Step 6: Supply cookies for better depth

The optional cookies field accepts a Pinterest cookie header copied from a logged-in browser session. Pinterest shows more to authenticated sessions than to anonymous ones, so providing cookies improves both reliability and how deep the scraper can go. It is entirely optional — the scraper works without it — but it is the single most effective lever if you find results thinner than expected.

Step 7: Run, review and export

Start the run and watch the log. When it finishes, open the Dataset tab and group by seed_value to compare keyword performance side by side. Export to JSON to preserve the nested pin, creator, board_ref and media objects, or flatten to CSV if you only need top-level fields such as id, url, title and the sentiment columns.

---

🔌 API Access & Integrations

Run the Pinterest scraper over HTTP and get dataset items back in the same request:

curl -X POST "https://api.apify.com/v2/acts/scrapers-hub~pinterest-scraper/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "queries": ["home decor"],
    "type": "all-pins",
    "limit": 100,
    "enrich_boards": false
  }'

With the Python client:

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")

run = client.actor("scrapers-hub/pinterest-scraper").call(run_input={
    "queries": ["scandinavian interior", "japandi living room"],
    "type": "all-pins",
    "limit": 300,
    "content_analysis": True,
    "enrich_boards": True,
})

by_keyword = {}
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    by_keyword.setdefault(item.get("seed_value"), []).append(item)

for keyword, pins in by_keyword.items():
    print(keyword, len(pins), "pins")
    print("  sample:", pins[0].get("title"), pins[0].get("url"))

Results can also be routed into Zapier, Make, Google Sheets or Slack, or delivered to your own service with an Apify webhook fired on run completion.

---

💡 Best Use Cases for Pinterest Data

🎨 Visual Trend Research

Pinterest is where aesthetic trends surface before they reach mainstream retail. Scrape a category keyword monthly and track how title phrasing and content_category_label distributions shift over time. The media object gives you the imagery to build a visual trend deck, while seed_value keeps each month's snapshot cleanly separated.

🛍️ E-commerce Product Discovery

With content_analysis enabled, the has_price flag and extracted links separate commercial pins from inspirational ones. Combine that with the destination links in the pin object to see which retailers are getting traction in a category, and use board_ref to understand what other products people group alongside them.

🤝 Influencer and Creator Discovery

Run with type set to profiles to build a list of creators active in your niche, or scrape pins and aggregate the creator object across records to find who appears most often. Follower count in the creator object gives a first-pass sizing signal, and each creator's pins tell you whether their aesthetic actually fits your brand.

🏢 Competitor Content Monitoring

Search for competitor brand names and product terms, then examine which of their pins are surfacing, which boards they appear on via board_ref, and how their descriptions read. sentiment_label on those descriptions gives a rough read on the tone competitors use, which is useful input when positioning your own content.

🗂️ Board and Curation Analysis

Scraping with type set to boards reveals how people organise a topic — the board names, the themes they cluster around, and how many collections exist for a given search. This is category taxonomy research done from real user behaviour rather than from internal assumptions.

🎬 Video Pin Performance Research

Set type to videos to isolate video pins. Comparing which topics generate video content versus static pins tells you where motion is worth the production cost, and the media object gives the associated assets for review.

📈 Keyword Comparison and Content Planning

Because every record carries seed_type and seed_value, running ten candidate keywords in one job and grouping the output by seed gives you an immediate comparison of result volume and content character per keyword. That is a fast, evidence-based way to decide which terms to build a Pinterest content calendar around.

---

⚙️ Tips for Better Pinterest Scraping Results

• Start with a small limit. The default of 50000 per search is a ceiling. Run 50 first, check that the pin, creator and media objects contain what you need, then scale up.
• Use specific keywords. "Home decor" returns an enormous, generic set. "Japandi living room shelving" returns a smaller set that is far more useful for content or product research.
• Turn off enrich_boards for large harvests. It adds a request per unique board. If you only need the board name and URL from board_ref, disabling it makes big runs noticeably faster.
• Supply cookies when depth matters. A logged-in cookie header improves reliability and how far the scraper can go. Refresh it if results start thinning out, since session cookies expire.
• Only ?q= URLs work as seeds. If you paste a board or profile URL into startUrls it will not be usable. Use queries, or a proper search URL.
• Group by seed_value before analysing. Mixing several keywords in one dataset is efficient, but any per-keyword comparison must group on the seed first or the numbers will be meaningless.

---

🛠️ Troubleshooting

My startUrls entry returned nothing. Only Pinterest search URLs containing a ?q= query parameter are usable as seeds. Board URLs, pin URLs and profile URLs will not produce results. Copy the URL from an actual Pinterest search, or use the queries field with a plain keyword instead.

I am getting fewer results than my limit. limit is a maximum, not a target. A narrow keyword simply may not have that many pins available in search results. Broaden the keyword, add more entries to queries, or supply a cookies header to increase result depth.

board_ref.extra is empty. That detail is only populated when enrich_boards is enabled. Set it to true and re-run if you need the board owner, preview images and cover images.

Sentiment or category fields are missing. The sentiment fields only appear when sentinent_analysis is enabled, and the content analysis object only when content_analysis is enabled. Both default to false. Note the spelling of sentinent_analysis, which was kept as provided.

Results feel shallow or the run slows down. Pinterest returns more to authenticated sessions. Adding a valid cookies header from a logged-in browser session is the most effective fix. Also consider lowering limit and splitting a large keyword list across several runs.

---

❓ Frequently Asked Questions About Pinterest Scraping

What can the Pinterest Scraper extract? Pins, video pins, boards and creator profiles from Pinterest search results, including titles, nested pin content, creator details, board references, media image variants and Pinterest's own metadata.

Do I need a Pinterest account to use this scraper? No. The scraper runs without credentials. Supplying an optional cookies header from a logged-in session improves reliability and result depth, but it is not required.

How do I scrape Pinterest by keyword rather than URL? Use the queries field. Each plain keyword string is expanded into a Pinterest search automatically, so you never need to construct the URL yourself.

Can I scrape Pinterest boards and profiles too? Yes. Set type to boards for pin collections or profiles for creator pages. The corresponding board or profile object appears on those records.

How many pins can I collect in one run? limit controls the maximum records saved per keyword search and defaults to 50000. Actual volume depends on how much Pinterest returns for your search terms.

What is enrich_boards for? It fetches full board details — owner, preview images and cover images — so board_ref.extra is complete. It costs one cached request per unique board, so disable it when you want faster, cheaper runs.

What does content_analysis add to each pin? A lightweight object containing hashtags, extracted links, word count and a has_price flag. It is the quickest way to separate commercial pins from purely inspirational content.

How accurate is the sentiment analysis? It is a lightweight heuristic scoring of the pin description, returned as sentiment_score, sentiment_score_normalized, sentiment_confidence and sentiment_label. Treat it as a directional signal for aggregate analysis rather than a precise per-pin judgement.

Do I need to configure a proxy for Pinterest scraping? No. Proxy handling is managed automatically inside the actor, which is why there is no proxy field in the input.

Can I extract the images themselves? The media object contains image variants with their URLs, widths and heights. You can download those URLs separately; the scraper stores the references rather than the binary files.

How do I tell which keyword produced a given record? Check seed_type and seed_value. Every record records whether it came from a URL or a keyword, and which one, so multi-keyword runs stay fully separable.

Can I schedule recurring Pinterest scrapes? Yes. Use Apify's Scheduler to run the same input on a cadence. Storing each run's dataset and comparing over time is how you turn one-off scraping into genuine trend tracking.

What export formats are available? JSON, CSV, Excel and XML from the Console. JSON is recommended because it preserves the nested pin, creator, board_ref, media and extra objects that CSV flattening loses.

Why is the sentiment input field spelled sentinent_analysis? The key name was kept exactly as it appeared in the original input schema for backwards compatibility. Use that spelling when calling the actor via the API.

Does the Pinterest scraper collect private boards or private pins? No. It only reads content that is publicly visible in Pinterest search results. Private boards and secret pins are not accessible to it.

---

🆘 Support & Feedback

Found a bug, a keyword that behaves unexpectedly, or a field that is not parsing correctly? Open a ticket in the Issues tab on the actor page with the run ID and your input.

Need a custom build — different result types, extra enrichment, or delivery straight into your analytics stack? Email scraperhubapi@gmail.com with the details.

If this Pinterest scraper is useful, a rating and a short review on the actor page helps others find it and guides what gets built next.

---

⚖️ Disclaimer

This Pinterest scraper collects only publicly available data from Pinterest search results — the same pins, boards and profiles any visitor can see. It does not access private boards, secret pins, direct messages or any authenticated-only content beyond what an ordinary session would show.

You are responsible for how you use this actor and the data it produces, including compliance with Pinterest's terms of service and with copyright law. Pins and the images they reference are typically the intellectual property of their creators; collecting a reference to an image is not the same as acquiring the right to republish it.

Where records include personal data — creator usernames, full names, profile details — you act as the data controller for anything you store and must comply with GDPR, CCPA and any other privacy regulation that applies. Only retain what you have a lawful basis to process, and honour access and deletion requests from the individuals concerned.

If you believe data collected through this actor should be removed, contact scraperhubapi@gmail.com with the details and the request will be addressed.