Facebook Keyword Search Scraper

Search public Facebook posts by keyword and get clean, structured data: full post text, author, date, total reactions, comments, shares, video views, media and links.

• No Facebook account needed. No login, password or cookies.
• Public content only. Returns what any logged-out visitor can see.
• Many keywords in one run. Keywords are searched in parallel.
• Optional filters. Newest posts first, last 24 hours / 7 days / 30 days, or a custom date range.
• Simple pricing. You pay only for posts you receive.

---

What it does

1. You enter one or more keywords (brands, products, topics, hashtags).
2. The Actor searches Facebook posts for each keyword.
3. It collects public posts up to your limit per keyword and removes duplicates.
4. Every post is saved to the dataset, ready to export as JSON, CSV or Excel, or to use through the API, n8n, Make or Zapier.

---

Use cases

• Brand monitoring: see what is being posted about your brand or products.
• Competitor tracking: follow posts mentioning competitors and compare engagement.
• Campaign and hashtag tracking: collect posts for a campaign keyword or hashtag.
• Market research: find public conversations about a topic, product or industry.
• Influencer and page discovery: find active pages and authors posting about a subject.

---

Input

{
    "queries": ["electric cars", "EV charging"],
    "maxResultsPerQuery": 50
}

Field	Default	Description
Search keywords (queries)	required	Keywords or phrases to search. Each one is searched separately. Up to 100 per run.
Maximum posts per keyword (maxResultsPerQuery)	100	Upper limit of posts for each keyword (1-5000).
Strict keyword filter (keywordFilter)	false	Keep only posts whose text contains the keyword.
Sort results (sortBy)	relevance	recent = newest posts first. See Filters.
Posted within (postedWithin)	any	Keep only posts from the last 24 hours, 7, 30, 90 days or 12 months.
Posted on or after / before (postedAfter, postedBefore)	empty	Custom date range, YYYY-MM-DD.
Search type (searchType)	posts	Type of results.
Proxy configuration (proxyConfiguration)	Residential	Residential is recommended.
Parallel keywords (maxConcurrency)	3	How many keywords are searched at the same time (1-5).

Tips

• Put all your keywords in one run. One run with 5 keywords is faster and cheaper than 5 runs with 1 keyword each.
• The limit is per keyword. 2 keywords x 50 = up to 100 posts. It is a maximum, not a guarantee: if Facebook shows fewer public posts for a keyword, you get what is available.
• Strict keyword filter: Facebook also matches hashtags, page names and link previews. Turn this on if you only want posts that mention the keyword in their text.
• No quotes needed. Use electric cars, not "electric cars".

---

Filters (optional)

Field	Values	What it does
Sort results (sortBy)	relevance (default), recent	recent asks Facebook for the newest posts first
Posted within (postedWithin)	any (default), 24h, 7d, 30d, 90d, 365d	Keeps only posts from that period
Posted on or after (postedAfter)	YYYY-MM-DD	Start date (ignored if Posted within is set)
Posted on or before (postedBefore)	YYYY-MM-DD	End date

Example, newest posts from the last 7 days:

{ "queries": ["Tesla"], "maxResultsPerQuery": 50, "sortBy": "recent", "postedWithin": "7d" }

• Date filters are always applied exactly on the post date, so every post you receive is inside your range.
• Most recent first shows about 50-70 of the newest posts per keyword (Facebook's limit for public search). For busy keywords that covers roughly the last day, so a higher limit may not be reached. The order is newest first in general, not strictly by minute; sort by timestamp if you need an exact order.
• With filters on, a keyword stops early when no more matching posts are found, so a run may return fewer posts than the limit. You pay only for posts you receive.
• Daily monitoring tip: run once a day with sortBy: recent and postedWithin: 24h.

---

Output

One dataset item per post:

{
    "type": "post",
    "postId": "1610706737767863",
    "authorName": "Example Page",
    "url": "https://www.facebook.com/...",
    "profileUrl": "https://www.facebook.com/...",
    "description": "Full post text...",
    "snippet": "Short preview of the post text...",
    "timestamp": "2026-10-05T09:12:44.000Z",
    "likesCount": 5646,
    "commentsCount": 285,
    "sharesCount": 25,
    "viewsCount": 166177,
    "mediaType": "video",
    "images": ["https://..."],
    "attachmentUrl": "https://example.com/...",
    "query": "electric cars",
    "searchType": "posts",
    "position": 1,
    "scrapedAt": "2026-10-07T10:35:00.000Z"
}

Field	Meaning
postId	Facebook post ID
authorName, profileUrl	Who published the post (page or profile)
url	Link to the post
description	Full post text
snippet	First 300 characters of the text
timestamp	When the post was published (UTC)
likesCount	Total reactions (like, love, haha, etc. combined)
commentsCount	Total comments
sharesCount	Total shares
viewsCount	Video views (videos only)
mediaType	text, photo, video, link, ...
images	Image links in the post
attachmentTitle, attachmentUrl	Shared link, if any
query	Keyword that found this post
position	Order of the post within that keyword's results
scrapedAt	When the post was collected

• A count is null when Facebook does not show it publicly. 0 means Facebook reported zero.
• Fields with no data are left out.
• Posts from different keywords can be mixed in the dataset. Use query to group them.

---

Duplicates

• The same post (same postId) is returned only once per keyword.
• Different posts are always kept, even from the same author.
• If two keywords find the same post, it appears once under each keyword.

---

Using it with the API or n8n

HTTP Request (runs the Actor and returns the posts in one call):

• Method: POST
• URL: https://api.apify.com/v2/acts/datafinch~facebook-keyword-search-scraper/run-sync-get-dataset-items?timeout=300
• Header: Authorization: Bearer <YOUR_APIFY_TOKEN>
• Body (JSON):

{
    "queries": ["electric cars", "EV charging"],
    "maxResultsPerQuery": 50
}

The call waits up to 5 minutes. For larger runs, start the run asynchronously and fetch the dataset when it finishes (the Apify node in n8n does this with "wait for finish").

To get only new posts, use the filters in the input instead of filtering afterwards, for example "sortBy": "recent", "postedWithin": "24h".

---

FAQ

Why did I get fewer posts than my limit? Facebook shows a limited number of results to logged-out visitors. Many keywords return roughly 60-100 posts, even with a higher limit.

Why are results slightly different between runs? Facebook's ranking changes over time, so the same keyword can return a slightly different set of posts.

Are the numbers exact? Reactions, comments, shares and views are the exact numbers Facebook provides, not rounded values like "1.2K".

Does it work for private groups or profiles? No. Only public content visible without logging in is collected.

Something isn't working. Please open an issue in the Issues tab with your run link. We usually respond quickly.