> For the complete documentation index, see [llms.txt](https://docs.generationlab.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.generationlab.org/marketplace/youtube-genz-mobile-viewing-dataset.md). # YouTube GenZ Mobile Viewing Dataset ### Overview The YouTube GenZ Mobile Viewing Dataset provides real-time behavioral data on how Gen Z users consume YouTube content on their mobile devices. The dataset is produced by **Verb AI** (by Generation Lab), which operates a consented panel of real US mobile device users. Viewing sessions are captured passively and in real time as panelists organically browse YouTube — no surveys, no recall bias, no simulated environments. This dataset is ideal for market researchers, media strategists, alternative data analysts, and brand teams seeking ground-truth insights into Gen Z digital behavior. *** ### Methodology #### Data Collection Verb AI maintains a consented opt-in panel of US-based mobile device users. When a panelist watches a YouTube video, the session metadata is captured, enriched with publicly available YouTube API data (video details, engagement metrics, channel info), and written to Snowflake. #### Key Properties * **Passive observation**: No surveys or self-reporting. Data reflects actual viewing behavior. * **Real-time ingestion**: Sessions appear in the dataset within minutes of occurring. * **Mobile-first**: All data originates from mobile devices, reflecting Gen Z's dominant consumption mode. * **Consented panel**: All participants have explicitly opted in to data collection. #### Privacy This dataset contains **no personal identifiers**. Specifically: * No user IDs, device IDs, or IP addresses * No tracking URLs or advertising identifiers * Geographic data is included at the city/region level only, for aggregate analysis * All data is derived from consented panelists *** ### Schema Reference The dataset is served through a single table: `YOUTUBE_VIDEO_SESSIONS`. | Column | Type | Description | | ----------------------- | --------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `VIDEO_ID` | VARCHAR | YouTube video identifier (the `v=` parameter from a YouTube URL). Primary key for joining with external YouTube data. | | `API_VIDEO_ID` | VARCHAR | Video ID as returned by the YouTube Data API. Typically matches `VIDEO_ID`. | | `API_TITLE` | VARCHAR | Title of the video as returned by the YouTube Data API. | | `API_KIND` | VARCHAR | YouTube API resource type (e.g., `youtube#video`). | | `CHANNEL_ID` | VARCHAR | Unique identifier of the YouTube channel that published the video. | | `CHANNEL_TITLE` | VARCHAR | Display name of the YouTube channel. | | `CATEGORY_ID` | VARCHAR | YouTube video category ID (e.g., `10` = Music, `20` = Gaming, `22` = People & Blogs). See [YouTube category reference](https://developers.google.com/youtube/v3/docs/videoCategories/list). | | `DESCRIPTION_TOP_LEVEL` | VARCHAR | Full video description as set by the creator. | | `DESCRIPTION_SNIPPET` | VARCHAR | Truncated snippet of the video description, suitable for previews and search results. | | `VIEW_COUNT` | NUMBER | Total view count of the video at the time of the session (from YouTube API). | | `LIKE_COUNT` | NUMBER | Total like count of the video at the time of the session. | | `COMMENT_COUNT` | NUMBER | Total comment count of the video at the time of the session. | | `DEFINITION` | VARCHAR | Video quality definition (e.g., `hd`, `sd`). | | `DURATION` | VARCHAR | Video duration in ISO 8601 format (e.g., `PT5M30S` = 5 minutes 30 seconds). | | `TAGS` | VARCHAR | Comma-separated list of tags assigned to the video by the creator. | | `THUMBNAIL_URL` | VARCHAR | URL to the video's default thumbnail image. | | `PUBLISHED_AT` | TIMESTAMP | Timestamp when the video was originally published on YouTube. | | `EVENT_TIME` | TIMESTAMP | Timestamp when the viewing session was observed by Verb AI's panel. | | `COUNTRY` | VARCHAR | Country of the panelist at the time of viewing (ISO 3166-1 alpha-2). | | `REGION` | VARCHAR | State or region of the panelist. | | `CITY` | VARCHAR | City of the panelist. | > **Note**: Column availability may expand over time. Check the Snowflake table definition for the latest schema. *** ### Data Coverage & Freshness | Property | Value | | ----------------------- | ----------------------------------------------------------------------------- | | **Update frequency** | Near real-time (minutes from session to table) | | **Geographic coverage** | Global, primarily US-based panel, with international YouTube content observed | | **Historical depth** | Rolling, contact Verb AI for exact date range | | **Approximate volume** | Thousands of sessions per day (varies with panel size) | *** ### Getting Started #### 1. Attach the listing After subscribing through the Snowflake Marketplace, a shared database will appear in your account. You can rename the database to anything you prefer. #### 2. Query the data All data lives in a single table. Reference it as: ```sql SELECT * FROM .YOUTUBE_VIDEO_SESSIONS LIMIT 10; ``` Replace `` with whatever you named the shared database. #### 3. Trial access A **7-day free trial** is available. During the trial you have full access to the dataset with no restrictions on query volume. *** ### Sample Queries #### Top videos by view count Find the most popular videos among Gen Z viewers ranked by total view count. ```sql SELECT VIDEO_ID, API_TITLE AS TITLE, CHANNEL_TITLE, VIEW_COUNT, LIKE_COUNT, COMMENT_COUNT FROM YOUTUBE_VIDEO_SESSIONS ORDER BY VIEW_COUNT DESC LIMIT 100; ``` #### Trending videos in the last 24 hours Discover what Gen Z is watching right now. ```sql SELECT VIDEO_ID, API_TITLE AS TITLE, CHANNEL_TITLE, VIEW_COUNT, LIKE_COUNT, EVENT_TIME FROM YOUTUBE_VIDEO_SESSIONS WHERE EVENT_TIME >= DATEADD('day', -1, CURRENT_TIMESTAMP()) ORDER BY VIEW_COUNT DESC LIMIT 100; ``` #### Top performing channels Identify the most popular creators among Gen Z audiences for influencer partnerships and sponsorship opportunities. ```sql SELECT CHANNEL_ID, CHANNEL_TITLE, COUNT(DISTINCT VIDEO_ID) AS VIDEOS_WATCHED, SUM(VIEW_COUNT) AS TOTAL_VIEWS, SUM(LIKE_COUNT) AS TOTAL_LIKES FROM YOUTUBE_VIDEO_SESSIONS GROUP BY CHANNEL_ID, CHANNEL_TITLE ORDER BY VIDEOS_WATCHED DESC LIMIT 50; ``` #### Content category breakdown Understand which content categories dominate Gen Z viewing time. ```sql SELECT CATEGORY_ID, COUNT(*) AS SESSION_COUNT, COUNT(DISTINCT VIDEO_ID) AS UNIQUE_VIDEOS, AVG(VIEW_COUNT) AS AVG_VIEWS FROM YOUTUBE_VIDEO_SESSIONS GROUP BY CATEGORY_ID ORDER BY SESSION_COUNT DESC; ``` #### Geographic viewing patterns Analyze regional content preferences across the US. ```sql SELECT REGION, CITY, COUNT(*) AS SESSION_COUNT, COUNT(DISTINCT VIDEO_ID) AS UNIQUE_VIDEOS FROM YOUTUBE_VIDEO_SESSIONS WHERE COUNTRY = 'US' GROUP BY REGION, CITY ORDER BY SESSION_COUNT DESC LIMIT 50; ``` *** ### Use Cases * **Trend detection**: Identify emerging content trends and viral videos before they peak by tracking real-time Gen Z viewing patterns. * **Content strategy**: Understand what video formats, durations, and topics drive engagement with Gen Z audiences. * **Geographic insights**: Analyze regional content preferences and discover location-based viewing trends. * **Creator & brand analysis**: Track channel performance and identify rising creators popular with Gen Z viewers. * **Market research**: Gain insights into Gen Z interests, preferences, and cultural moments as they happen. * **Advertising intelligence**: Optimize ad placements by understanding which content categories and channels capture Gen Z attention. * **Alternative data**: Use real-time behavioral signals for investment research, competitive intelligence, and consumer sentiment analysis. *** ### FAQ **How is this different from YouTube Analytics or the YouTube API?** YouTube Analytics shows creators their own channel data. The YouTube API provides public metadata. Neither reveals *who is watching what, when, and where* at a demographic level. This dataset captures actual viewing sessions from a consented Gen Z panel, providing demand-side behavioral data that doesn't exist in any YouTube-provided product. **Can I identify individual users?** No. The dataset contains no user-level identifiers. All geographic data is at the city/region level for aggregate analysis only. **How large is the panel?** Panel size is proprietary and changes over time. Contact Verb AI for current panel metrics. **What does "near real-time" mean?** Viewing sessions typically appear in the Snowflake table within minutes of occurring on the panelist's device. **Can I get historical backfill?** Contact Verb AI to discuss historical data availability. *** ### Support & Contact For questions about the dataset, partnership inquiries, or support: * **Documentation**: [https://docs.generationlab.org](https://docs.generationlab.org/) * **Email**: [moshe@generationlab.org](mailto:moshe@generationlab.co) * **Website**: [https://verbai.generationlab.org](https://verbai.generationlab.org/) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.generationlab.org/marketplace/youtube-genz-mobile-viewing-dataset.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.