CrowdTangle alternative: Social Fetch vs CrowdTangle

CrowdTangle was Meta's public insights tool for Facebook and Instagram. Newsrooms, researchers, fact-checkers and many others used its dashboards, saved lists, search and alerts to see which public posts were spreading. Meta announced the shutdown in March 2024, and as of August 14, 2024 it is no longer available.

Meta's replacement is the Meta Content Library, a web interface, and the Content Library API. Together they cover public Facebook Pages, posts, profiles, groups, events, comments and Marketplace listings, and public Instagram accounts, posts and comments. They are research tools, not a new CrowdTangle for everyone.

Meta's eligibility rules cover two groups: people at academic institutions that are accredited, degree-granting and not-for-profit, and people at not-for-profit institutions whose primary purpose is scientific or public-interest research. You apply through Meta's Research Tools Manager; an independent partner reviews the application (Meta's docs name ICPSR at the University of Michigan and France's CASD), which Meta says takes 2 to 3 weeks, followed by another 2 to 3 weeks of processing.

If you qualify, it's a good deal: there are no fees for the web interface or for the API in Meta's Secure Research Environment, and API queries return up to 100,000 results. If you work at a brand, an agency, a startup or most newsrooms, you don't meet those criteria. Columbia Journalism Review's reporting on the switch made the same point about news organisations.

Content Library API analysis runs inside Meta's Secure Research Environment or an approved third-party cleanroom, in Python or R. You can export research outputs under agreed terms, and downloading a public data subset is limited to the web interface for approved applicants. That fits a study. It doesn't fit a product feature, a client dashboard or an alert that has to reach Slack.

Social Fetch returns JSON to your own servers. What you do with it is up to you and your counsel: Meta's terms restrict automated collection without its permission, and Social Fetch can't grant that permission.

On Facebook (21 endpoints): Pages and public profiles by URL (GET /v1/facebook/profiles), their posts, reels, photos and events (/v1/facebook/profiles/posts), single posts, comments and replies (/v1/facebook/posts/comments), video transcripts, public groups and group posts (/v1/facebook/groups/posts), events, Marketplace listings and the Facebook Ad Library (/v1/facebook/ad-library/ads/search). On Instagram (20 endpoints): profiles, posts, reels, comments, highlights, hashtag search and transcripts. Most routes cost 1 credit.

The same key covers TikTok, X, YouTube, Reddit and the rest of the 25 platforms. Route lists: Facebook hub, Instagram hub. For group research specifically, see the Facebook group data API page.

Be clear about the gaps before you migrate a CrowdTangle workflow. Social Fetch has no keyword search across all public Facebook posts: you look up known Pages, groups and posts by URL (on Instagram, hashtag search covers some discovery). It has no archive: you get what a Page's public feed shows when you call it, not a stored history of past posts and their engagement. There are no dashboards, saved lists, leaderboards or Chrome extension.

For alerts, Monitors watch Instagram posts, reels and hashtags and send a signed webhook when something new appears, but Monitors don't cover Facebook yet; for Facebook Pages you schedule the posts call yourself and diff the results. For a wider survey of options, see best Facebook data APIs in 2026.

CrowdTangle is gone. If you're an academic or non-profit researcher, apply for the Meta Content Library: it's free, has search and an archive, and is the closest thing to CrowdTangle's data. If you're not eligible, or you need the data in your own product, Social Fetch returns public Facebook Pages, posts, comments and groups, plus Instagram and other platforms, self-serve and per request. It doesn't replace CrowdTangle's search, archive or dashboards.