Facebook Ad Library Scraper

Extract ads from the Facebook Ad Library — creative, landing pages, platforms, impressions, spend, and advertiser page details — without a Facebook account or API key.

Search by keyword, pull every ad an advertiser page is running, or search for a keyword inside one advertiser's ads. Output is one clean row per ad with dates as YYYY-MM-DD, ready for CSV, Excel, or JSON.

---

Quick start

Search the Ad Library for a keyword

{ "keyword": "running shoes", "country": "US", "activeStatus": "active", "limitPerSource": 100 }

Get all ads from advertiser pages

{ "pages": ["https://www.facebook.com/ZapierApp", "nike", "254515547747520"], "limitPerSource": 0 }

Search for a keyword inside an advertiser's ads

{ "keyword": "free trial", "pages": ["https://www.facebook.com/ZapierApp"], "keywordWithinPages": true }

Monitor a competitor every week — schedule this and set a run tag per run:

{
    "pages": ["https://www.facebook.com/ZapierApp"],
    "datePreset": "last7d",
    "limitPerSource": 0,
    "includePageSummary": true,
    "runTag": "2026-W39"
}

Political ads about a topic, exact phrase, Instagram only

{
    "keyword": "climate change",
    "searchType": "keyword_exact_phrase",
    "adType": "political_and_issue_ads",
    "publisherPlatforms": ["instagram"],
    "country": "US",
    "datePreset": "last30d"
}

---

Features

• No login required — scrapes the public Ad Library over plain HTTP, no browser
• Three ways in — keyword search, advertiser pages, or raw Ad Library URLs
• Search within an advertiser — turn on keywordWithinPages to find only one advertiser's ads about a topic
• Advertiser pages in any form — page URL, handle (ZapierApp), numeric page ID, or an Ad Library link
• Every Ad Library filter — country, active status, ad category (political, housing, employment, credit), media type, date range, platforms, languages, exact-phrase matching
• Per-page summary rows — total ads, active ads, platform breakdown, date range for each advertiser
• EU transparency — EU reach with country / age / gender breakdown, targeting, payer and beneficiary, no login needed
• Rich ad-level fields — publisher platforms, impressions band, spend and reach (political ads), collation count (number of near-identical variants), page likes and categories, AI-generated flag
• Clean creatives — headline, body, CTA, landing page, image, video and thumbnail; carousels nested under cards
• Two output shapes — one row per ad (default) or one row per creative card
• Cost control — per-source and global ad limits
• Errors in the dataset — a page that cannot be resolved produces an error row, never a silent empty run
• Proxy-ready — optional Apify or custom proxies, usually not needed

---

Input

Sources

Field	Type	Description
keyword	string	A search of the whole Ad Library for this term, like the Ad Library search box.
pages	string[]	Advertiser pages: every ad each one is running. Each entry may be a page URL, a handle, a numeric page ID, or an Ad Library URL containing view_all_page_id.
urls	string[]	Searches copied from the Ad Library website, one per entry ({ "url": "…" } objects are accepted too). Everything in the URL is kept; filters you change are added only when the URL does not set them. Facebook page URLs here behave like entries in pages.
keywordWithinPages	boolean	Default false: every entry above is its own source. true: the keyword stops being a search of its own and instead filters the pages' ads, e.g. "Model Y" inside Tesla's ads.

Provide at least one of keyword, pages, or urls. Each one is scraped independently and the results are combined in one dataset, so the default input (a keyword, a page, and a URL) yields three sets of ads. Empty entries are ignored.

Filters (apply to keyword and pages)

Field	Type	Default	Description
country	string	ALL	ISO 3166-1 alpha-2 code (US, GB, DE) or ALL
activeStatus	string	all	all, active, inactive
adType	string	all	all, political_and_issue_ads, housing_ads, employment_ads, credit_ads. Political and issue ads carry spend and reachEstimate.
mediaType	string	all	all, image, video, meme, none
datePreset	string	—	last24h, last7d, last14d, last30d, last90d. Ads that were running at any point in the window.
startDateMin / startDateMax	string	—	YYYY-MM-DD. Explicit window; overrides datePreset. Same "running at any point" semantics as Facebook's own date filter.
publisherPlatforms	string[]	[]	Any of facebook, instagram, audience_network, messenger, threads. Empty = all.
contentLanguages	string[]	[]	ISO 639-1 codes (en, es, de). Empty = all.
searchType	string	keyword_unordered	keyword_unordered (any of the words) or keyword_exact_phrase.
sortBy	string	impressions_desc	Sent to Facebook as the sort preference. Facebook's server-rendered results currently ignore it and come back in Facebook's own order.

Limits

Field	Type	Default	Description
limitPerSource	integer	10	Max ads per keyword, page, or URL. 0 = all.
maxItems	integer	0	Max ads across the whole run. 0 = no limit. The run stops as soon as it is reached.

Output options

Field	Type	Default	Description
outputMode	string	ad	ad = one row per ad (first card flattened, all cards under cards). card = one row per creative card.
includePageSummary	boolean	false	Add one summary row per advertiser page (see below).
includeEuTransparency	boolean	false	Fetch each ad's details: EU reach with country / age / gender breakdown, targeting, payer and beneficiary, Instagram handle and followers, page verification. One extra request per ad.
includeRawData	boolean	false	Attach the untouched Facebook ad object as rawData.
runTag	string	—	Copied into the runTag column of every row.
proxyConfiguration	object	—	Optional proxy settings.

---

Output

Ad rows (default, outputMode: "ad")

{
    "adArchiveID": "899524914329334",
    "adUrl": "https://www.facebook.com/ads/library/?id=899524914329334",
    "pageName": "Wealthsimple",
    "pageID": "268452016650171",
    "pageUrl": "https://www.facebook.com/wealthsimple/",
    "pageProfilePictureUrl": "https://scontent.../pp.jpg",
    "pageLikeCount": 123456,
    "pageCategories": ["Financial service"],
    "isActive": false,
    "startDate": "2021-10-20",
    "endDate": "2022-01-12",
    "publisherPlatforms": ["FACEBOOK", "INSTAGRAM", "AUDIENCE_NETWORK", "MESSENGER"],
    "impressions": ">1M",
    "impressionsIndex": 39,
    "spend": "CA$350K - CA$400K",
    "currency": "CAD",
    "reachEstimate": ">1M",
    "categories": ["POLITICAL"],
    "isPolitical": true,
    "paidForBy": "Wealthsimple Technologies Inc.",
    "disclaimer": null,
    "collationCount": 1,
    "collationID": "165980345735977",
    "isAiGenerated": false,
    "stateMediaLabel": null,
    "targetedCountries": [],
    "totalActiveTimeSeconds": null,
    "displayFormat": "VIDEO",
    "format": "video",
    "isDCO": false,
    "primaryText": "Receive one free stock (up to $4.5k) just for signing up!",
    "headline": "Buy & sell stocks in just a few taps",
    "body": "Receive one free stock (up to $4.5k) just for signing up!",
    "cta": "Install Now",
    "ctaType": "INSTALL_MOBILE_APP",
    "destinationUrl": "http://itunes.apple.com/app/id1403491709",
    "displayDomain": "itunes.apple.com",
    "linkDescription": null,
    "imageUrl": "https://scontent.../thumb.jpg",
    "videoUrl": "https://video.../video.mp4",
    "videoThumbnailUrl": "https://scontent.../thumb.jpg",
    "cardCount": 1,
    "cards": [{ "cardIndex": 0, "headline": "Buy & sell stocks in just a few taps", "body": "…", "cta": "Install Now", "ctaType": "INSTALL_MOBILE_APP", "destinationUrl": "…", "displayDomain": "itunes.apple.com", "linkDescription": null, "imageUrl": "…", "videoUrl": "…", "videoThumbnailUrl": "…" }],
    "sourceUrl": "https://www.facebook.com/wealthsimple",
    "position": 1,
    "totalResults": 1032,
    "runTag": "2026-W39",
    "scrapedAt": "2026-09-22T16:00:00.000Z"
}

Field	Description
adArchiveID, adUrl	Ad Library ID and direct link
pageName, pageID, pageUrl, pageProfilePictureUrl, pageLikeCount, pageCategories	Advertiser page
isActive, startDate, endDate	Run status and dates (endDate is null while still running)
publisherPlatforms	Where the ad runs: FACEBOOK, INSTAGRAM, AUDIENCE_NETWORK, MESSENGER, THREADS
impressions, impressionsIndex	Impressions band (e.g. 10K - 15K) and Facebook's ordinal index; only shown by Facebook for some ads
spend, currency, reachEstimate	Political / issue ads only
categories, isPolitical, paidForBy, disclaimer	Ad category labels and political disclaimers
collationCount, collationID	Number of near-identical variants of this ad, and their group ID
isAiGenerated	Facebook's "digitally created" flag
stateMediaLabel, targetedCountries, totalActiveTimeSeconds	Transparency labels when Facebook provides them
displayFormat, format, isDCO	Facebook's format (IMAGE, VIDEO, CAROUSEL, DCO, DPA…), our derived image / video / carousel / text, and whether the ad uses dynamic creative templates
primaryText	Post text above the creative (null for DCO, where Facebook sends template variables)
headline, body, cta, ctaType, destinationUrl, displayDomain, linkDescription, imageUrl, videoUrl, videoThumbnailUrl	First creative card
cardCount, cards	All creative cards (carousel cards or DCO variants), deduplicated
sourceUrl	The keyword URL, page, or URL you supplied
position, totalResults	Rank within its source, and the total number of matching ads Facebook reports (capped at 50,001)
runTag, scrapedAt	Your tag and the scrape timestamp
rawData	The untouched Facebook object, only when includeRawData is on

EU transparency and payer details (includeEuTransparency: true)

Adds these columns to every ad row (or card row), plus the full adDetails object:

Field	Description
euTotalReach	Estimated Meta accounts in the EU that saw the ad at least once
payer, beneficiary	Who paid for and who benefits from the ad, as declared for EU delivery
igUsername, igFollowers, pageVerification	Advertiser's Instagram handle and followers, and Facebook verification status (BLUE_VERIFIED…)
adDetails.euTransparency	targetsEu, totalReach, ageAudience {min,max}, genderAudience, locationAudience[] (included/excluded countries, regions, cities), reachBreakdown[] rows of {country, ageRange, male, female, unknown}
adDetails.payerBeneficiary	All payer / beneficiary pairs
adDetails.advertiser	pageLikes, pageCategory, igVerified, isPoliticalPage, isRestricted
adDetails.isTakenDown, violationTypes, isViolatingEuRules	Enforcement flags

Ads never shown in the EU have euTransparency: null. The data comes from the same request the Ad Library website makes when you click See ad details; it needs no login. If Facebook changes that query, the run emits one AD_DETAILS_QUERY_INVALID error row and continues without details.

Card rows (outputMode: "card")

Same ad-level fields, but one row per creative card with a cardIndex column instead of cardCount / cards. Use this for creative analysis where each carousel card matters.

Page summary rows (includePageSummary: true)

One per advertiser page, marked rowType: "pageSummary". Shown in the Pages view of the Output tab.

{
    "rowType": "pageSummary",
    "sourceUrl": "https://www.facebook.com/ZapierApp",
    "pageName": "Zapier",
    "pageID": "309796912374932",
    "pageUrl": "https://www.facebook.com/ZapierApp/",
    "pageProfilePictureUrl": "https://scontent.../pp.jpg",
    "pageLikeCount": 56539,
    "pageCategories": ["Software"],
    "pageIsDeleted": false,
    "totalAds": 1032,
    "adsCollected": 50,
    "activeAdsCollected": 12,
    "platformBreakdown": { "FACEBOOK": 50, "INSTAGRAM": 31, "MESSENGER": 9 },
    "earliestStartDate": "2024-01-24",
    "latestStartDate": "2026-06-04",
    "runTag": "2026-W39",
    "scrapedAt": "2026-09-22T16:00:00.000Z"
}

totalAds is the count Facebook reports for your filters (so with activeStatus: "active" it is the number of ads currently live). The other counts describe the ads actually collected.

Error rows

If a source fails, the dataset gets a row with errorCode and error instead of a silent empty result. Shown in the Errors view.

errorCode	Meaning
INVALID_PAGE	An entry in pages is not a Facebook page URL, handle, or ID
PAGE_ID_NOT_FOUND	The Facebook page could not be resolved to an advertiser ID (private, deleted, or not a page)
DUPLICATE_SOURCE	Two entries resolved to the same page; its ads appear once
NO_AD_DATA	The Ad Library returned a page without ad data
CHALLENGE_FAILED	Facebook's anti-bot challenge could not be solved
AD_DETAILS_QUERY_INVALID	Facebook rejected the ad-details query (emitted once; the run continues without EU data)
REQUEST_FAILED	Network or HTTP failure after retries

---

How it works

The Ad Library server-renders the first page of results as JSON inside the HTML. The scraper:

1. Sends a plain HTTP request with browser-like headers (no browser, so runs are cheap and fast)
2. Solves Facebook's lightweight __rd_verify challenge if it is returned
3. Parses the embedded search_results_connection JSON
4. Follows the end_cursor to the next page until the limits are reached
5. For advertiser pages, first resolves the page's numeric ID, then queries the Ad Library with view_all_page_id

---

Pricing

Pay per event. You are charged for what lands in your dataset, not for rows:

Event	When it is charged
ad	Once per ad, in both output modes. A six-card carousel in "one row per creative card" mode is still one ad.
ad-details	Once per ad whose EU transparency details were fetched successfully (only with includeEuTransparency).
apify-actor-start	Once per run, by the platform.

Error rows and page-summary rows are free. Set Maximum cost per run in the run options to cap spending; the actor stops cleanly when the cap is reached and you keep everything scraped up to that point.

Proxy

The Ad Library is a public transparency tool and works without a proxy in normal use. If you scrape very large volumes and hit blocks, add residential proxies:

"proxyConfiguration": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }