Facebook Likes Scraper

Extract public Facebook likes and reactions from post or reel URLs without Facebook cookies. The actor returns structured public reaction previews: reaction type, display name, profile URL, identifiers when Facebook exposes them, source, and scrape time. It is designed for social listening, competitor engagement analysis, research, and agent workflows that need a clean dataset rather than a browser session.

Use this actor for public Facebook post and reel reaction data. Do not use it for private posts, closed groups, complete private audience graphs, or Facebook Page follower counts. Facebook itself determines which reaction identities it exposes publicly; the actor never claims to reveal hidden likes.

What you get

Field	Meaning
postUrl	Public Facebook post or reel that was inspected
reactionType	Like, love, care, wow, sad, angry, haha, or another public label
reactorName	Public display name when Facebook shows one
reactorProfileUrl	Public profile URL when available
reactorId	Source-provided Facebook identifier, when available
reactionCount	Count-only fallback metric; null for individual reactor previews
recordType	public_reactor or aggregate_metrics so agents can distinguish the shape
source	Native Apify path, ScrapeCreators, or SociaVault

The default source mode uses the native, no-cookie Apify reaction path first. If it cannot return public reactor previews, the actor tries configured provider endpoints. Providers are deliberately allowed to return only aggregate_metrics when they expose a verified reaction count. They are never used to fabricate a person’s identity.

Pricing

This actor uses Pay per event + usage. It charges $0.00005 per actor start and $0.01 for each persisted public reaction or aggregate fallback record. Apify platform compute and network usage are charged separately by Apify. Set maxResults before running: a run capped at 50 records has a maximum event charge of $0.50005, excluding platform usage.

Input

Provide one or more public Facebook post or reel URLs in postUrls. maxResults is a global cap from 1 to 500; maxResultsPerPost limits how much a single URL can consume in a batch. Keep sourceMode as auto for the recommended native-first behavior. Choose apifyNative if you require only individual public reactor previews. scrapeCreators and sociaVault are advanced fallback modes and can only produce count-level data if their documented post response includes a reaction count.

Example input:

{
  "postUrls": ["https://www.facebook.com/WBBPosts/posts/pfbid02j21mpxEygikxxqKh3AaQxsct5vByRpuZVTgpBMELoKArJT6oJjvek8Y3TdRdDgvYl"],
  "maxResults": 25,
  "sourceMode": "auto"
}

Use from JavaScript

import { ApifyClient } from 'apify-client';

const client = new ApifyClient({ token: process.env.APIFY_TOKEN });
const run = await client.actor('khadinakbar/facebook-likes-scraper').call({
  postUrls: ['https://www.facebook.com/reel/753347914167361'],
  maxResults: 10,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(items);

Use from Python

from apify_client import ApifyClient

client = ApifyClient('YOUR_APIFY_TOKEN')
run = client.actor('khadinakbar/facebook-likes-scraper').call(run_input={
    'postUrls': ['https://www.facebook.com/reel/753347914167361'],
    'maxResults': 10,
})
for item in client.dataset(run['defaultDatasetId']).iterate_items():
    print(item)

For AI agents

Use this actor when an agent needs public engagement identities or reaction labels for specific Facebook posts. It returns one structured dataset item per publicly exposed reactor, not a prose summary. Inspect recordType: a public_reactor record has an individual public preview; an aggregate_metrics record is a count-only fallback. The OUTPUT key reports COMPLETE, PARTIAL, VALID_EMPTY, INVALID_INPUT, UPSTREAM_FAILED, or CONFIG_ERROR and includes billing and warning diagnostics.

Limits and responsible use

Only process data that is publicly visible and relevant to your legitimate research or analytics workflow. Facebook can change what it renders publicly, limit visible reaction previews, or return no identities for a post. A valid post with no visible reactions is a truthful VALID_EMPTY result, not a hidden error. If one post in a batch is blocked but other records are saved, the actor returns PARTIAL and preserves the useful data.

The actor uses owner-configured provider secrets only for fallback post metrics. API keys are never accepted through input or included in datasets, logs, OUTPUT, or RUN_SUMMARY. The fallback cannot bypass Facebook privacy controls, and neither fallback provider is advertised as a source of private liker identities.

FAQ

Why did I get fewer people than the post’s reaction total? Facebook often exposes only a public preview of reactors. The total can be larger than the preview, and availability can vary by post, region, audience, and Facebook changes.

Does this need my Facebook cookie? No. The default route is designed for public, no-cookie data. It does not accept Facebook credentials.

Why is reactionCount sometimes null? Individual public_reactor rows represent one person, so a count would be misleading. reactionCount is only used on a clearly labeled count-only fallback row.

Can I scrape any Facebook URL? No. Use a public facebook.com post or reel URL. Pages, groups, search pages, malformed URLs, and private content are outside the actor’s contract.

Legal note

You are responsible for ensuring your use complies with applicable law, Facebook policies, and the rights of the people whose public data you process. Do not use this actor to make high-impact decisions about individuals or to circumvent platform privacy controls.