Defining Metadata Schemas for Media Workflows

Managing content at scale is an ongoing challenge for media organisations. Across every genre, every platform, and every region, the appetite for entertainment has exploded. But whether that content is the latest movie release, or a classic series from decades ago, it all needs to be secure and accessible – both remotely and on-prem.

Working with large volumes of media comes with a unique set of challenges. Teams need to be able to archive seamlessly and offload content from different storage tiers to maximise resources. Operators need to playback and repurpose media using a customised UI. Companies need intuitive solutions that can handle content scale, different user rights, and a host of post-production workflows.

So, what’s the key to optimising all these different elements?

‍
Gold-standard Metadata

Metadata helps organisations get the most out of their media. It’s the window into content workflows, and it means you can utilise relevant information to speed up post-production timelines. Metadata helps companies to identify and categorise not just newly created media assets, but also older archived content.

To ensure that workflows are as efficient as possible, media operators must be able to review content, then mark clips and caption audio descriptions for tagging. Without an advanced search management solution, that has annotation and indexing processes built in, media can easily become cluttered and confused.

Metadata is nothing without context. To be leveraged successfully, it must be searchable, accurate, and consistent.

Common Metadata Pitfalls

1.  What’s the purpose? – For metadata to be useful it must have a clearly defined reason for being there. Organisations need to efficiently index, categorise, and process large volumes of assets, without content becoming siloed. Technical metadata such as; duration, framerate resolution, aspect ratio, codec and bitrate, is always recorded. But if you expand this to include descriptive metadata such as; title, season, episode, genre, when it was published, content description, creation date, updated date, language, file status, and individual scene descriptions – then things get complicated.

If unnecessary metadata is logged, it can end up obscuring the useful information teams need. This can leave media operators searching for a needle in a haystack. The creation of metadata is an investment of both time and resources, so it should always have a clearly defined use-case.

‍

2. What’s in a name? – AI tools can be used to support the identification and tagging of metadata. But skilled team members are needed to add nuance and context to the information generated. One of the key challenges for large media organisations, when processing a lot of assets, is a lack of consistency. If one operator has a completely different way of describing the elements in a scene to another operator, that can create some serious challenges.

For example, a common use case in content compliance is the identification of unsuitable scenes that might contain nudity or violence. But if a scene that needs to be removed contains an inconsistent description of a gun / rifle / weapon / firearm – then it could be missed altogether. Ensuring that naming conventions and terminology are tagged consistently, helps prevent any mishaps.

‍

3.  What’s the consensus? – There is a level of unnecessary aggregation of metadata that happens at nearly every media organisation. Different teams have different use cases, and metadata can be split into various categories. These categories might include; technical metadata, descriptive metadata, custom metadata, supply chain metadata, and customer specific metadata. To manage all of this effectively, there needs to be some kind of cross content, cross divisional guide, that is agreed upon and followed by different teams.

However, establishing a metadata rulebook for cross team adoption is easier said than done. There is also (unsurprisingly) a lack of consensus across the industry about how metadata should be recorded. There’s a lot of time pressure on people working in indexing, ad marketing, and file request roles. People in those roles may need to search through thousands of assets per day. So, they need to have confidence that the metadata supports efficient search functionality. A clear set of rules can provide that confidence.

‍

Well-defined frameworks for metadata tagging, makes searching for content much more efficient downstream. To manage content at scale, companies need a customisable, metadata-driven MAM solution, that is integrated into their infrastructure.

Cantemo is a hybrid-cloud, media asset management system that supports efficient, remote production workflows. It allows operators to run elastic searches, connect with preferred post-production workflow tools, and set advanced archiving rules. Find out more about solutions to manage content at scale here.