THE VIDEO VAULT

WHO WATCHED
THAT?

Over 10,000 videos Fortune 100 and FTSE 100 companies put on YouTube, and how each one actually did. Every video is measured against its own channel’s usual views, not cat videos. The ones that blow past it are outliers.
Filter it. Watch it. Steal what works.

ALL VIDEOS

Loading the library…

    Public YouTube data. Performance is relative to each channel’s usual views: a 4.2× outlier got 4.2 times its channel’s median.

    Want help applying this to your team? Book a call

    HOW WE MEASURE

    Outperformance
    Views divided by the channel’s usual (median) views for that format, Shorts against Shorts, long-form against long-form.
    Engagement
    Likes plus comments per 1,000 views, and how that compares with the channel’s norm.
    Confidence
    How much to trust the number, based on how many videos the channel has, how big its usual audience is, how new the video is, and whether stats are hidden. We hide low-confidence results (thin or young channel baselines) by default; tick “Show low-confidence” to see them.
    Likely paid
    Flagged when views far outrun subscribers or the channel’s norm while engagement stays low, or the title carries an ad length like :15.

    Flags are a guess from public data, not proof. YouTube doesn’t publish watch time, retention or how much of a video’s audience came from ads.