Skip to main content
All CollectionsAbout global demand dataFAQ
How do you select which content you track?
How do you select which content you track?

Understanding the Content Discovery Engine

Product Education avatar
Written by Product Education
Updated over a week ago

The process by which our Demand Measurement system selects and tracks TV, movie, and talent content on a global scale utilizes a system unique to Parrot Analytics. The following three sections explain the sequence of events that occurs for the selection, collection and reporting of global demand data on a title-by-title basis:

1. Content Discovery Engine

The process begins with a highly-intelligent "Content Discovery Engine" within our platform: In summary, this engine enriches each title's metadata, and in doing so, identifies a title across all of the demand expression data streams we capture.

Our Content Discovery Engine plugs into all leading metadata providers about titles. This data includes title-specific metadata like episode data and demand expression platform tags like social media handles, hashtags, video streaming channels and tags, pages (etc.) for each title.

In addition, the platform also populates Parrot Analytics’ internal Content Genome™, which tags every title with additional layers of metadata at a granularity level not provided by off-the-shelf metadata providers. Combined, this metadata is crucial for enabling us to identify and capture the correct data for each title across all of our data streams.

Please note that because our platform is language-agnostic, non-English titles can certainly be tracked on the platform: In fact, many non-English titles are already being tracked. The Content Discovery Engine is currently being expanded to plug into wider non-English metadata providers to accelerate local language title on-boarding. Our title on-boarding queue includes many non-English titles that will be on-boarded over the coming months. In addition, please note that on-boarding of local titles can be accelerated by submitting a manual on-boarding request.

In summary, our Content Discovery Engine identifies, on-board and enriches titles in part by interrogating metadata providers and applies multiple layers of metadata enrichment via our Content Genome™ system, and then feeds this metadata into the downstream “Data Collection Engine” (refer to next section). This process happens for content in all languages originating from over 50 countries.

2. Data Collection Engine

Following on from content discovery, our Data Collection Engine checks every live data stream (from social networks to downloading/streaming sites and video streaming sites, etc), and then begins tracking the global demand for that title.

3. Demand Measurement Platform

Once a title has gone through the validation steps, its global demand data will be available to be consumed via our DEMAND360 platform and demand API.

Did this answer your question?