What Is Video Metadata? Types, How to Edit It, and How to Strip It

Video metadata is the embedded information inside a video container file—or attached alongside it—that describes its technical properties, content, rights, and administrative history. It enables media players, video editors, search engines, and content management systems to interpret, catalog, index, and play video assets correctly without needing to decode the entire audiovisual stream.

📋 TL;DR • Three core types: administrative (rights, timestamps, GPS), descriptive (title, tags, chapters, transcripts), and structural (resolution, codec, bitrate, frame rate). • Container write limits: MP4 and MOV support full tag editing in ExifTool; MKV, WebM, and AVI are read-only there (use FFmpeg or mkvpropedit instead); and C2PA provenance credentials can be deleted but not arbitrarily edited. • Lossless editing: Use ExifTool (exiftool -all= video.mp4) or FFmpeg (ffmpeg -i in.mp4 -map_metadata -1 -c copy out.mp4) to modify or strip tags in seconds without re-encoding. • Upload stripping: YouTube and Vimeo discard camera and GPS metadata on ingest, relying on on-page schema and platform studio fields for search ranking. • Playback optimization: Video CDNs read structural metadata to package adaptive bitrate ladders and deliver the optimal stream to each device.

Video continues to account for the majority of global bandwidth. At the end of 2025, video traffic accounted for around 75 percent of all mobile data traffic (Ericsson, 2026). As archives scale into thousands of files, organizing assets, scrubbing private GPS tags, and ensuring rapid web streaming depends on accurate file metadata.

What is video metadata? Meaning and container architecture

A video file is not a single flat recording. It is a container (such as MP4, MOV, or WebM) packaging multiple streams together: video elementary streams, audio channels, subtitles, and dedicated data blocks. Video metadata consists of the structured bytes inside that container describing everything except the raw pixels and audio samples.

Metadata resides in three primary locations:

1. Container headers: Embedded directly in the file, such as the moov atom in MP4/MOV files or EBML elements in MKV/WebM files.
2. Sidecar files: External text or XML files, including WebVTT/SRT for captions and transcripts and Adobe XMP sidecars (.xmp).
3. On-page web markup: Structured Schema.org VideoObject JSON-LD embedded in HTML, enabling search crawlers to index videos without fetching gigabytes of media.

In MP4 and QuickTime MOV containers, structural metadata lives in the moov atom (Movie Box), indexing chunk offsets and frame timings. Moving the moov atom to the front—a process called FastStart—lets browsers stream immediately via HTTP range requests rather than waiting for full downloads.

The 3 main types of video metadata

All video metadata falls into administrative, descriptive, and structural categories:

1. Administrative metadata

Administrative metadata records logistical, legal, and hardware facts. Cameras write hardware serials, capture settings, timestamps, and geolocation tags. Editors add client codes and copyright licenses.

Administrative tags also cover authenticity. Under the C2PA (Coalition for Content Provenance and Authenticity) standard, signed manifests verify whether media was captured by a lens or generated synthetically. Similarly, version 1.7 of the IPTC Video Metadata Hub (VMH), released November 10, 2025, added four properties for embedding the AI prompt data and models used to produce synthetic media (IPTC, 2025).

2. Descriptive metadata

Descriptive metadata identifies what happens on screen, powering catalog search and recommendation algorithms:

• Title and description: Plaintext summaries of the asset.
• Tags and categories: Taxonomical labels grouping related content.
• Chapters and timestamps: Navigational markers segmenting key sections.
• Transcripts: Time-aligned spoken text enabling deep search indexing.

3. Structural metadata

Structural metadata acts as the hardware decoding blueprint:

• Codec and profile: The compression format (e.g., H.264, HEVC, AV1).
• Resolution and aspect ratio: Pixel dimensions and display proportions; see resolution explained and how to choose your aspect ratio.
• Frame rate and bitrate: Playback speed (24, 30, 60 fps) and data rate; review video bitrate settings.
• Audio layout: Channels (stereo, 5.1) and sample rate (48 kHz).

Structural metadata enables adaptive bitrate streaming. When a CDN ingests a master video, it reads these values to generate multi-rendition HLS and DASH manifests.

Container support matrix: Which formats allow metadata editing?

Container formats vary substantially in how they handle metadata editing. Based on ExifTool version 13.59 (ExifTool, 2026), video container formats divide into distinct write-capability tiers:

If your media library contains legacy AVI files, ExifTool treats them as read-only. Remux them into MP4 or MOV containers before attempting ExifTool edits, or write the tags with FFmpeg.

🚀 Ready to put those container specs to work? Once your codec, resolution, and bitrate fields are accurate, the next decision is who reads them at delivery time. Compare hosting that reads your file’s metadata automatically →

How to see video metadata

Inspect existing tags and streams using operating system tools, desktop analyzers, or command-line utilities.

1. Built-in OS tools

• Windows File Explorer: Right-click the video file, select Properties, and open the Details tab to view title, tags, dimensions, bitrate, and frame rate.
• macOS Finder: Select the file and press Cmd+I to open Get Info, expanding “More Info” for codecs, dimensions, and audio channels.
• macOS Terminal (mdls): Inspect Spotlight attributes directly:

mdls -name kMDItemPixelWidth -name kMDItemPixelHeight -name kMDItemCodecs video.mp4

2. Dedicated desktop utilities

Desktop utilities make the distinction between descriptive tags and stream-level properties visible without opening an editor.

VLC’s Media Information panel exposes descriptive tags and codec details in one view.

• VLC Media Player: Press Ctrl+I (Windows/Linux) or Cmd+I (macOS) to view “Media Information.” The General tab shows descriptive tags, while Codec Details lists stream parameters.
• MediaInfo: Developed by MediaArea, MediaInfo reports deep structural detail, including chroma subsampling (4:2:0 vs. 4:2:2), bit depth (8-bit vs. 10-bit), and container profiles.

3. Command-line inspection

For automated pipelines, command-line inspection yields instant, parseable data:

Using ExifTool:

exiftool -a -u -g1 video.mp4

Using FFprobe:

ffprobe -v quiet -print_format json -show_format -show_streams video.mp4

How to edit MP4 metadata and add metadata to video files

Editing metadata does not require re-encoding. Because tags reside in container headers separate from compressed video frames, modifications are lossless and finish in milliseconds. In our testing on a 4 GB ProRes master, rewriting the title and copyright fields with ExifTool completed faster than the file could be copied, because only the header region is touched and the compressed streams are left byte-identical on disk.

Video metadata tools comparison

Professional video editing software makes metadata management a core part of the workflow.

Editing MP4 metadata with ExifTool

Consult the ExifTool SourceForge tag documentation for comprehensive tag parameters.

Add title, artist, and copyright to a single file:

exiftool -Title="Product Overview 2026" \
         -Artist="Acme Corporation" \
         -Copyright="2026 Acme Corp. All rights reserved." \
         -overwrite_original video.mp4

Batch add metadata to all MP4 files in a folder:

exiftool -Artist="Acme Corporation" -overwrite_original -ext mp4 /path/to/videos/

Adding metadata with FFmpeg (Lossless)

Inject metadata using -metadata flags while passing -c copy to bypass re-encoding:

ffmpeg -i input.mp4 \
       -metadata title="Product Overview 2026" \
       -metadata artist="Acme Corporation" \
       -c copy \
       output.mp4

How to remove metadata from a video (Privacy and file scrubbing)

Stripping metadata is essential when handling footage recorded on smartphones, drones, or studio rigs. In practice the risk is rarely the title field: it is the coordinates, camera serials, and scratch project paths that ride along silently from capture through export, and they stay in the file until someone removes them on purpose.

Why remove video metadata?

1. Location privacy (GPS): Smartphones embed exact latitude, longitude, and elevation into container headers. Sharing raw clips can expose home addresses or confidential filming sites.
2. Internal project leaks: NLE exports often retain camera serial numbers, local drive paths, and scratch audio tracks.
3. Header bloat: Redundant XMP history blocks create unnecessary overhead on web servers.

Method 1: Remove metadata using Windows File Explorer

1. Right-click the video file and select Properties.
2. Go to the Details tab.
3. Click “Remove Properties and Personal Information”.
4. Select “Create a copy with all possible properties removed” and click OK.

[!NOTE] Windows Explorer only removes basic Windows Shell properties. It does not reliably scrub embedded GPS tracks or C2PA provenance blocks. Use command-line tools for thorough scrubbing.

Method 2: Lossless complete strip using ExifTool

ExifTool deletes all non-essential metadata while keeping video and audio bitstreams intact.

Scrub all metadata tags:

exiftool -all= -overwrite_original video.mp4

Scrub only GPS location coordinates:

exiftool -gps:all= -overwrite_original video.mp4

Scrub C2PA provenance manifests:

exiftool -JUMBF:all= -overwrite_original video.mp4

Method 3: Lossless complete strip using FFmpeg

Instruct FFmpeg to drop the global metadata it would otherwise copy using -map_metadata -1 (add a stream-scoped specifier such as -map_metadata:s:v -1 when per-stream tags also need clearing):

ffmpeg -i input.mp4 -map_metadata -1 -c copy scrubbed_output.mp4

Because -c copy avoids transcoding, the command completes in seconds with zero image degradation.

File metadata vs. platform metadata: What survives upload?

A frequent misconception is that YouTube and Vimeo index the metadata embedded inside uploaded video files.

They do not.

When you upload media to YouTube or Vimeo, automated ingest systems strip virtually all embedded file-level metadata (including camera serials, EXIF tags, and GPS coordinates) during transcoding. This protects creator privacy and standardizes internal storage.

Platforms rely entirely on platform metadata: titles, descriptions, tags, chapter timestamps, and uploaded caption files. For complete optimization across platform algorithms, review our guide to YouTube’s platform metadata.

Embedded file metadata remains critical for:

• Internal DAM systems: Searching and cataloging high-resolution studio masters.
• Direct client downloads: Delivering commercial video assets and training files.
• Open RSS video podcasts: Supplying titles and episode details directly to podcast players.
• Owned website embeds: Enabling video players and CDNs to read technical stream specifications.

Video metadata management: Organizing libraries and workflows

Without systematic video metadata management, expanding media libraries result in misplaced assets, duplicate shooting costs, and rights compliance issues. We have seen teams re-shoot footage that already existed in the archive simply because nobody could search for it, which is a tagging problem rather than a storage problem.

1. Standardizing asset taxonomy

Establish uniform metadata conventions across media ingest:

• Controlled vocabularies: Standardize category and project naming to avoid disjointed searches (e.g., adopting “Promo” rather than mixing “promo” and “teaser”).
• Date conventions: Enforce ISO 8601 formatting (YYYY-MM-DD) in file names and timestamps.
• Licensing records: Embed rights holders and expiration dates into administrative tags before archival.

2. Ingest automation with DAM and MAM platforms

Media Asset Management platforms (such as Frame.io, iconik, and CatDV) use metadata to automate production:

• Proxy workflows: Reading structural metadata to automatically generate lightweight editing proxies.
• Speech-to-text tagging: Transcribing spoken dialogue into searchable time-indexed tags.
• Lifecycle archiving: Moving master files to cold cloud storage once project completion tags are detected.

Proper metadata hygiene ensures your team can smoothly optimize your web video content.

Video metadata optimization: Video SEO and schema markup

Search crawlers like Googlebot do not inspect binary MP4 atoms to index web pages. For search visibility, focus on on-page schema markup for video.

Video metadata SEO checklist

1. Title: Include your primary keyword in the first 60 characters of the page heading.
2. Written summary: Provide a 150–300 word description summarizing the video beneath the player.
3. Thumbnail: Supply a high-contrast 16:9 thumbnail hosted on a stable URL.
4. Captions and transcripts: Host full written transcripts on the page so search engines can index spoken keywords. Learn formatting rules in our guide to closed captions and WebVTT subtitles.
5. Structured schema: Implement Schema.org VideoObject markup in JSON-LD.

VideoObject JSON-LD template with Key Moments

Google uses VideoObject markup to generate rich search snippets, duration badges, and chapter key moments:

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "VideoObject",
  "name": "What Is Video Metadata and How to Manage It",
  "description": "A guide explaining video metadata types, container limits, and command-line management.",
  "thumbnailUrl": [
    "https://example.com/thumbnails/16x9/video-metadata.jpg"
  ],
  "uploadDate": "2026-09-16T08:00:00+00:00",
  "duration": "PT4M35S",
  "contentUrl": "https://example.com/videos/video-metadata-explainer.mp4",
  "embedUrl": "https://example.com/embed/video-metadata",
  "hasPart": [
    {
      "@type": "Clip",
      "name": "Definition of Video Metadata",
      "startOffset": 0,
      "endOffset": 65,
      "url": "https://example.com/embed/video-metadata?start=0"
    },
    {
      "@type": "Clip",
      "name": "The 3 Main Types of Metadata",
      "startOffset": 66,
      "endOffset": 160,
      "url": "https://example.com/embed/video-metadata?start=66"
    },
    {
      "@type": "Clip",
      "name": "How to Strip GPS Tags",
      "startOffset": 161,
      "endOffset": 275,
      "url": "https://example.com/embed/video-metadata?start=161"
    }
  ]
}
</script>

Key moments in Google Search: Clip vs. SeekToAction

Google Search documentation distinguishes between two key-moment implementations (Google Search Central, 2026):

• Clip markup: Creators manually define chapter boundaries with startOffset and endOffset. Clip is supported in all languages where Google Search is available.
• SeekToAction markup: Google algorithmically links to timestamps using player URL patterns. However, Google restricts SeekToAction to 12 languages: English, Spanish, Portuguese, Italian, Chinese, French, Japanese, German, Turkish, Korean, Dutch, and Russian.

For international publishers, Clip ensures universal chapter display. For CMS setup, see our guide to video SEO for WordPress.

How video delivery networks use structural metadata

Once uploaded, structural metadata directly governs player delivery:

1. FastStart streaming: Reading moov atom offsets allows web browsers to begin playback immediately via HTTP range requests without waiting for complete downloads.
2. Player scaling: Reading width and height values ensures the player renders the correct aspect ratio, preventing pillarboxing.
3. Adaptive ladder generation: Edge servers inspect source codecs, color spaces, and audio layouts to generate multi-bitrate HLS/DASH streams; see how a video CDN uses your file’s specs.

Generic third-party players often fail to optimize these delivery mechanics. If you host video on business pages, examine why YouTube embeds can hurt website performance and evaluate hosting options in our video hosting platforms comparison.

How SmartVideo uses structural metadata for flawless playback

When hosting video through Swarmify SmartVideo, structural metadata management happens automatically.

SmartVideo parses container headers upon ingest:

• Instant format detection: Identifies source resolution, frame rate, and codec to serve the optimal stream per device.
• Dynamic bitrate optimization: Prevents stalls by adapting stream renditions in real time based on user connection speed.
• Global CDN delivery: Caches video across a worldwide edge network for instant, buffer-free playback.

SmartVideo provides predictable pricing based on views and storage with unlimited bandwidth on every plan:

• Startup: $19/month (billed annually) for 10,000 monthly views and 1 TB storage.
• Growth: $59/month (billed annually) for 50,000 monthly views and 3 TB storage.
• Pro: $99/month (billed annually) for 150,000 monthly views and 5 TB storage.

SmartVideo automates metadata handling and stream delivery so you can focus entirely on producing quality content.

Frequently asked questions

Putting metadata management on autopilot

Metadata work rewards a small amount of discipline at ingest. Decide which administrative fields your team must record, standardize the descriptive vocabulary, verify the structural values with MediaInfo or FFprobe before publishing, and strip the private tags from anything that leaves your network. Everything after that is repetition, which is exactly the part a script or a hosting platform should own rather than a person.

Try SmartVideo Free for Your Site →