As a developer, you are generally prohibited from creating metrics that replace or modify the data returned by the YouTube API Services. However, subject to your acceptance of the amendment to the Developer Policies, to support advanced analytics and creator tools, YouTube permits the calculation of specific additional metrics provided they adhere to the following guidelines subject to your acceptance of the amendment to the Developer Policies (select "Section 5: Use Cases, API Integration, and Feature Implementation" then "Analytics & Reporting" as your use case to accept these terms).
Acceptable metrics
Acceptable metrics are those that use YouTube API Data for analytics purposes and created in compliance with the amendment to the Developer Policies.
List of Example Metrics:
- Custom Channel Scores and Ratios
- Financial Performance Projections
- Content Categorization and Tagging
- Viewer Sentiment Analysis
- Gamification and Leaderboards
- Brand Suitability & Safety Scoring
Your API Service must reflect an analytics use case on YouTube.
1. Custom Channel Scores and Ratios
What this means: You may calculate and assign custom "scores" to channels based on independently calculated averages, sums, or ratios of API Data. These metrics must add unique value through analysis and must not simply redisplay raw API data under a new name.
Examples
- Allowed:
- Creating a "Creator Influence Score" based on a weighted formula of average views, subscriber growth rate, and engagement ratios.
- Ranking channels on a leaderboard based on a "Content Quality Score" derived from comment-to-view ratios.
- Do not:
- Misrepresent API Data's definition or provenance (e.g. displaying "subscriber count" as a "score").
2. Financial Performance Projections
What this means: You may infer or project the financial performance of a YouTube channel (e.g. estimated revenue), provided these figures are clearly presented as third-party estimates and not as YouTube approved or published data.
Examples
- Allowed:
- A website offering an "Estimated Channel Revenue Calculator" that uses API Data (views, subscribers) as inputs to estimate earnings.
- Using API Data to advise brands on how much to pay a channel for a sponsorship based on your revenue estimates.
- Do not:
- Present financial data or projections (e.g. Revenue, CPM) as YouTube published or approved financial data or projections (implying private access) and prominently display a disclaimer that the financial metrics are not approved by Google.
3. Content Categorization and Tagging
What this means: You may use analysis to assign descriptive sub-genres or
tags to videos and channels. These must be additive and distinct from YouTube's
video categories, such as videoCategories.snippet.title.
Examples
- Allowed:
- Augmenting API Data with descriptive sub-tags (e.g. labeling a video in the "Gaming" category with specific tags such as "Speedrun" or "Minecraft").
- Creating your own tagging system that includes similar terms already found in the API Data, such as "Gaming", so long as it is clearly disclosed to the user that they are your tags.
- Do not:
- Replace or override a published YouTube category (e.g. displaying a custom genre as if it were a published YouTube category).
4. Viewer Sentiment Analysis
What this means: You may estimate viewer satisfaction or dissatisfaction (sentiment) using aggregate data, such as like/dislike ratios or comment analysis. You cannot use this data to profile users based on protected attributes, such as their age, race, religious affiliation, political leaning, sexual orientation, or health status.
Examples
- Allowed:
- Creating a tool that assigns a "Viewer Sentiment Score" to channels based on engagement ratios.
- Using Natural Language Processing (NLP) on comments fetched via the Data API to publish a "Channel Happiness Report" or satisfaction rating.
- Do not:
- Infer or estimate sensitive protected attributes of the audience or creator (e.g. age, race, religious affiliation, political leaning, sexual orientation, or health status).
5. Gamification and Leaderboards
What this means: You may rank channel performance or track views between different channels for the purpose of comparison or historical analysis.
Examples
- Allowed:
- A website displaying a daily or weekly "Top YouTuber by View Growth" leaderboard.
- A "History of YouTube" line graph showing the 36-month subscriber growth curves of different creators.
- Do not:
- Present data in a way that fosters harassment or brigading (e.g. framing comparisons as a "war," "battle," or real-time rivalry intended to incite derogatory behavior).
6. Brand Suitability & Safety Scoring
What this means: We are introducing a distinction between Planning (pre-campaign) and Measuring (post-campaign) to determine allowed use cases.
- Planning (Creator Insights): You may use API Data metadata to assess channels and videos for both "Brand Suitability" (brand-specific fit) and "Brand Safety" for the purpose of influencer marketing, vetting, and matchmaking.
- Measuring (Reporting): You may use Google-supplied post-ad campaign reports augmented with API Data sourced metadata to calculate "Brand Suitability" scores (e.g., performance or topic alignment). However, you must not use these reports augmented with API Data sourced metadata to make "Brand Safety" claims regarding the safety of specific ad impressions or platform-wide verification.
Examples
- Allowed:
- Creator Insights (Planning): "Brand Safety" or "Brand Suitability" scores for a list of potential influencers to help an advertiser plan an ad campaign.
- Suitability Reporting (Measurement): Generating a report that identifies which channels an ad campaign ran on and alignment with an advertiser's suitability preferences.
- Do not:
- Safety Reporting (Measurement): Generate a report containing claims about the safety of specific ad impressions delivered in a campaign, or a "Post-Campaign Brand Safety Report". Note: Validated safety reporting requires impression-level data restricted to Brand Safety Reporting Partners.
- Generalizing Delivery: Generate a report containing claims about the Brand Safety of a specific video or channel by utilizing Google-supplied post-ad campaign reports in any way, including via augmenting with Data API-sourced metadata.
Definitions
"Brand Safety" means (i) the foundational standards and protocols designed to ensure that advertising does not appear adjacent to content that is universally deemed inappropriate or harmful; (ii) the categorical exclusion of content defined by the industry-recognized "Brand Safety Floor," which includes, but is not limited to, illegal acts, incitement of violence, hate speech, and explicit or graphic material; and (iii) the maintenance of a digital environment where the presence of a brand does not imply association with content that violates fundamental community or legal standards. Detailed definitions found here: https://www.brandsafetyinstitute.com/resources/frameworks/brand-safety-floor-suitability.
"Brand Suitability" means (i) the framework used to align the context of content with an advertiser's specific brand identity, values, and risk tolerance; (ii) the assessment and categorization of content, which while meeting the Brand Safety Floor, may contain sensitive themes such as debated social issues, suggestive language, or news-related violence; and (iii) the application of enhanced advertiser controls to determine the appropriateness of such content for customized placement based on brand-specific criteria. Detailed definitions found here: https://www.brandsafetyinstitute.com/resources/frameworks/brand-safety-floor-suitability.
"Planning" means (i) the use of Data API Services to identify and define target audiences, keywords, or media inventory; (ii) the establishment of campaign objectives, budget allocations, and bidding strategies; and (iii) the generation of performance forecasts, reach estimates, or media plans prior to the execution or purchase of advertising.
"Measurement" means (i) the process of tracking, quantifying, and analyzing user interactions with advertising content, including but not limited to impressions, clicks, views, and conversions; (ii) the attribution of these interactions to specific campaigns, creatives, or placements using Google-provided tracking technologies; and (iii) the evaluation of advertising effectiveness through the retrieval of performance metrics and reporting data to inform optimization.