This guide is designed to help you through the process of integrating metadata effectively into a streaming app. You’ll learn how to identify your content catalog and feed details, build and test your feed, and deliver your feed in a way that supports a seamless and engaging user experience. Effective metadata integration drives user engagement, streamlines operations, and opens up monetization opportunities.
Prerequisites
Before integrating, you will need to confirm if metadata integration is appropriate for your app on our platform. If you’re unsure if your app is eligible, reach out to your partner manager to confirm.
Identify your content catalog
Ahead of your meeting with our metadata team, you will need to answer the following questions:
What size is your catalog?
Is the catalog content VOD, LiVOD, or both?
What kind of Program Types (movies, episodes, short-form) does your catalog contain?
Are there any special entitlements, such as ads or no ads?
What language(s) does your catalog support?
Will your program metadata be in the language being spoken?
Do you have a relationship with Gracenote?
Choose your feed delivery model
You must choose a method of delivery to use when building your feed. Metadata can be provided via a push or pull model.
Push model is preferred
Pushing your metadata feed to our systems is the preferred method of integration that gives you full control of how and when your feed is updated.
With a push model, you actively send (push) the metadata to a server or destination without waiting for a request from our platform.
The push model consists of the following:
You call the API.
You can send updates ad hoc.
You are in full control of when you add and remove content and speeds up the overall ingest process.
You will be issued a SAT user that can be used to generate an authentication token.
You might use different tokens for QA and production environments.
You will be able to test your feed more quickly.
You have more fine-grained control over what you send to us.
With each push feed, we will need to know:
If the feed is just VOD or Live (i.e., Live VOD or Linear)
If you have multiple subscription models
With a pull model, our platform initiates the request, at a predefined cadence, to receive metadata from you. Our platform will retrieve the content from an agreed-upon location and ingest the metadata into our ecosystem. The Pull model gives you less control, and you have to coordinate updates with us.
The pull model consists of the following:
We call the feed location (Static URL, FTP, S3, etc.).
You can tell us how often to pull the feed (e.g., every six hours, daily).
You need to tell us if the cadence changes with as much lead time as possible—a 48-hour minimum. Lead time may be affected by moratoriums or other technical delays.
You’ll map with standardized content tags that we provide.
With each pull feed, we will need to know:
Feed Location (e.g,, Static, URL, FTP, S3)
Feed Auth - if a token is needed to call the API
Data specification
How often to pull the data
Mechanism for updating
If the feed is a delta or full feed (full is preferred)
If they are sparse or full objects (if there is a delta)
How deletes are handled
If content is just VOD or Live (Live VOD / Linear)
Subscription models (single or multiple)
If the content is linked to availability or if the two provided separately
Build and test your feed
After you’ve chosen your feed model, you are ready to start building and testing your feed. To build your feed, you will need to:
Configure your feed according to our Metadata Specifications.
Ensure your feed is formatted correctly by validating your metadata syntax.
Talk to your partner manager and metadata contact once you are done building and validating your feed.
Coordinate with our technical teams to integrate with our Metadata Endpoints and correct any feed issues before deployment.
Work with your metadata contact to test in the QA and production environments before final deployment in production.