All Collections
About/Basics
Intro to CrowdTangle
Understanding and Citing CrowdTangle Data
Understanding and Citing CrowdTangle Data
C
Written by Christina Fan
Updated over a week ago

CrowdTangle exclusively tracks public content. Since we focus on public Instagram accounts, Facebook Pages, public Facebook groups, and don’t include data like reach or impressions, there can often be misconceptions about what CrowdTangle data is showing. We wanted to lay out some of the most common misconceptions and help make sure CrowdTangle data is reported on accurately.

When thinking about CrowdTangle as a tool for explaining the Facebook ecosystem, it’s important to remember that engagement data is imperfect. Engagement data is only one lens into what’s happening on Facebook and Instagram, and it’s important to look at a variety of metrics to get a holistic picture.

Use this hub to:

What Data CrowdTangle Tracks

TL;DR

  • CrowdTangle is a public insights tool from Facebook that helps publishers, journalists, researchers, fact-checkers and more follow, analyze, and report on what's happening across social media.

  • We make public content from Pages, groups, Instagram accounts, and popular subreddits more discoverable, and we make engagement data on that content, (i.e. shares, video views, comments, and reactions) easy to sort through at scale.

  • CrowdTangle only tracks data on public content.

  • We do not track any non-public content. We not track things like reach (the number of people exposed to a post) or Instagram Stories.

Insights from CrowdTangle data can be shared publicly, but you may not share the raw data publicly (such as CSVs). We ask you to properly cite CrowdTangle -- even when sharing on social media -- as outlined here.

How to use CrowdTangle data in your stories

Use CrowdTangle and our data to report on what’s happening across social media around key news events. Here are some articles that can help in your reporting:

How to accurately cite CrowdTangle data in your article or publication

Make sure to cite CrowdTangle when using our data that readers know where your information is coming from and what data is and isn’t included in your reporting.

  • Always cite CrowdTangle -- even if you’re just posting using our data on social media.

  • Short form citation is “Data from CrowdTangle, a public insights tool owned and operated by Facebook.”

  • Make sure to contextualize the data. This means explicitly stating that we only track engagement data on public accounts, and this doesn’t include ads.

Common misconceptions about CrowdTangle data

CrowdTangle gives important insights into how public content performs and how the most influential public accounts are using social media. That covers a lot, but not everything. Here are some common misconceptions we’ve noticed.

  • We don’t differentiate between organic and inorganic engagement.

  • We don’t measure video watch time.

  • We don’t track reach (the number of people exposed to a post) or impressions (the number of times a post was seen).

Glossary of terms

There’s a lot of specific social data terms we use when talking about CrowdTangle data and social media data overall. Here’s a guide.

  • CrowdTangle tracks interactions, which is defined as reactions, comments and shares. Interactions are also known as engagement.

  • Overperforming is calculated by benchmarking how many interactions that account’s posts usually get after a certain period of time.

  • Reach: The unique number of people who saw a post at least once. This means that CrowdTangle data will not enable you to say that X post reached Y number of people. Do not conflate Reach with the total number of followers of a page, or members of a group.

More tools for tracking social data

  • There are many other tools available to track social media data, both from Facebook and third parties.

  • Many of these tools are geared towards page owners looking to optimize a customer journey or understand their audience.

  • All data available to researchers is listed at research.fb.com/data.

Did this answer your question?