Facebook Videos Scraper: Page Videos, Views
Pricing
from $4.22 / 1,000 videos
Pull a public Facebook page's whole Videos tab, reels included, with no login: give links, usernames or page ids. Each video: exact plays and post plays, reactions by type, comment and share counts, title, upload date, length, HD and SD MP4 files, captions. JSON, CSV, API. Pay per video.
One per line: a public page or profile (facebook.com/natgeo, facebook.com/natgeo/videos/, facebook.com/profile.php?id=100044623170418, facebook.com/people/Name/100044623170418/, a username such as natgeo or @natgeo, a numeric profile id or a page's numeric (delegate) id such as 23497828950). Its Videos tab is read: every video and reel the page posted, newest first. A link to one single video or reel is not read (a free status row says so). Up to 500 values. Also read as startUrls, profileUrls, facebook_urls, username, pageId, profileId, pageNames, pageIds, pageTargets, urls, delegatePageId or delegate_page_id.
One https link to a public text or CSV file of up to 1 MB, or to a Google Sheet shared with "anyone with the link" (its first tab, or the tab in the link, is read as CSV). Every cell that names a Facebook page or profile is read as one more value of "pages". startUrls items in Apify's { "requestsFromUrl": ... } form are read the same way.
How many videos to deliver for each page, newest first (50 when empty, up to 100000: a page's whole Videos tab). Videos left out by a filter do not count. Also read as resultsLimit, maxResultsPerPage, maxVideosPerTab or maxVideosPerProfile.
An upper bound on the videos the whole run delivers, across every page. Empty: no bound beyond the videos per page and your charge limit. Also read as maxResults, maxVideos or maxTotalItems.
Only videos published on or after this day (UTC): YYYY-MM-DD, an ISO time, today, yesterday, or a relative "7 days", "2 weeks", "3 months", "1 year" counted back from the run's start. The Videos tab lists videos newest first, so the walk stops at the first older ones. Also read as onlyNewerThan, publishedAfter, dateFrom or newerThan.
The same window as "Published on or after", given as a number of days back from the run's start (0 is today). Send one of the two.
Only videos published on or before this day (UTC), in the same forms. The newer videos before it are read free (they are not charged) within the run's unpaid work allowance, so an end date far back in a busy page's list can end the run early with a free row saying so. Also read as onlyOlderThan, publishedBefore, dateTo or olderThan.
Leave out videos with fewer plays than this (Facebook's exact play count). Videos left out are free. Also read as minViews.
Leave out videos with more plays than this. Videos left out are free.
Leave out videos shorter than this many seconds. Videos left out are free.
Leave out videos longer than this many seconds. Videos left out are free.
Leave out videos with fewer reactions than this. Videos left out are free. With Video details off, Facebook's list sometimes gives 10,000 or more as 10,000: a minimum above 10,000 is surest with the details on.
Only videos whose title, text or hashtags contain at least one of these words or phrases, as whole words, case and accents ignored. Videos left out are free. Also read as keywords or captionContains.
Leave out videos whose title, text or hashtags contain any of these words or phrases (the same matching). Videos left out are free.
Keep only reels, or only videos that are not reels (empty: both; the Videos tab lists both). Judged on Facebook's own link of each video, which its details give, so it needs "Video details" on. Videos left out are free. Also read as videoType; onlyReels (true) keeps reels.
Keep only videos that carry Facebook captions (subtitles), judged on their details (needs "Video details" on). Videos left out are free.
On (the default): each video's own page is read and delivered at no extra charge: reel or video, comment count, share count, the HD video file, Facebook's captions (link, language, every track), the music of a reel, the exact time. Off: the rows carry what the Videos tab list gives (plays, reactions, time, text, size, SD file, thumbnail) and the run is faster. Also read as includeEngagement.
Read Facebook's own captions file of each delivered video that has one and deliver it as text and timed cues (transcript, transcriptCues). No extra charge; needs "Video details" on. A video without captions gets transcriptStatus no_captions. Also read as fetchTranscripts.
Deliver the comments of each delivered video as rows of their own, up to the number below, in the order chosen below. Charged as the Comment event per comment row. Commenters' names are delivered; their profile ids and links only with personal data on.
With comments on: how many comment rows (replies included) to deliver for each video (20 when empty). Also read as maxCommentsPerItem or comments_limit.
With comments on: read the replies under each comment too (charged as Comment rows, inside the comments per video).
With replies on: how many replies to read under each comment (10 when empty).
Add each Page's About tab and its Meta Ad Library facts to its page row: category, address, phone, e-mail, websites, business hours, price range, the rating line, and whether the Page runs ads now (with its active ads count). No extra charge; read only for a page that delivered a video.
Off: a person's profile (a public figure's included) keeps its name, and its profile id, username, link and picture are empty, and its video links are Facebook's watch links (facebook.com/watch/?v=), which name no profile; mentions and commenters that are people keep their names without ids or links; links in the text that reach a person's profile are left out. On: they are delivered, and phone numbers and e-mail addresses written in the text get fields of their own. You are responsible for using personal data lawfully, the GDPR included. Texts are always delivered as Facebook serves them.
full: every field of a video. compact: 26 fields most buyers sort, chart and download by (ids, link, kind, title, text, time, length, plays, counts, engagement rate, thumbnail, video file, transcript, page).
How many pages to read at the same time (1 to 3).
The pause between two calls of one page, in milliseconds (200 to 10000; 300 when empty).
End the run after this many seconds with what was read so far (60 to 86400). Empty: the run's own timeout.
Apify Proxy by default: the Videos tab pages, each video's details and the comments go through Apify's datacenter pool, and the Videos tab's list through Apify residential exits (the only exits Facebook answers it on), in the country named here (US when empty); naming the RESIDENTIAL group changes nothing else. Your own proxy links are possible (every call through them), and no proxy uses the run container's own address (Facebook then gives the first 9 videos of each page only).
Comment order
commentsOrder
Optional
With comments on: oldest first (the default), newest first, or Facebook's most relevant first. Also read as comment_filter (most_relevant, newest, all_comments).