What are the best practices for preserving and optimizing video assets in DAM?

I'd like to first present some basics on video strategy, regardless of how you intend to use your DAMS to support that that strategy. I would expect if you were investing in a DAMS, you would already have a video strategy in place, in production, and resourced. Video delivery throughout an Enterprise and out to customers, vendors, agencies etc. is not an easy task. On-premise origination or cloud? SSO requirements, CDN, hybrid, on and on and on. In many cases the codec(s) used for delivery will be determined by company security policies, network bandwidth, inside versus outside firewall etc. Outside the scope of this article. You do need to have a video platform and delivery strategy in place that you will SUPPORT with your DAMS.

If you find yourself at this point you likely already have a workflow that includes decisions you have made about platforms, storage, formats and codecs and how you wish to work. Therefore I would urge you to decide to think about the incoming and outgoing content - what are your consumers watching on or downloading/viewing on? And how do your editors ingest the content onto the platform? Do you wish for them to work within the platform? It's important to assess individuals workflows and analyse how they fit together before deciding what format to output your content in.

I'm going to follow here the broadly supported architecture where the DAMS contains final finished goods. Most DAMS that support video assets usually have TWO main asset formats. An uncompressed final deliverable such as a .mov at approx. 20Mb+/sec, and what we called a mezzanine file in .MP4 at approx. 5Mb/sec. Raw video and edit clips etc. remain in MAM or other storage. I've always utilized a third party delivery service that includes Player management for actually streaming the content to the end viewer. Companies such as Brightcove, uStudio and others provides capabilities that you would spend a fortune on trying to do it yourself. You need to stream. Serving from web or network storage is will download the asset, not stream it.

Maintaining the original quality of video assets is crucial for ensuring the longevity and usability of video files across various platforms. By preserving the master file in its original format, you not only safeguard against quality degradation but also maintain the flexibility to adapt the content for future technological advancements. Leveraging lossless compression and avoiding unnecessary edits can significantly enhance the asset's value, ensuring it remains sharp and professional.

Since I have always used a 3rd party service to provide rendition and bandwidth management based on the end device, I focus on requirements such as thumbnails, custom player mods to support video carousels, thumbnails, chapter support, accessability, captioning and foreign language support. I've always used DAMS to provide metadata and formats that specific social applications require (usually custom sizing and aspect ratios). The smart players provided by these video platforms take care of responsive design options, hot links etc. Use the DAMS as intended, as a DAMS. Utilize a video delivery platform to handle the delivery environment.

Adaptive bitrate streaming and responsive design are essential for optimising video assets in a Digital Asset Management (DAM) system, ensuring seamless delivery across various devices and network conditions. Leveraging these technologies not only enhances user experience but also maximises reach and engagement. Additionally, incorporating thumbnails, captions, and transcripts not only improves accessibility but can also assist with SEO, making your video content more discoverable and user-friendly.

Organising and tagging video assets in a DAM system is crucial for efficient retrieval and reuse. Implementing a robust metadata strategy, including detailed descriptions, keywords, and technical specifications, enhances searchability and discoverability. Leveraging AI and machine learning can further automate and refine the tagging process, ensuring consistency and saving time. This approach not only streamlines asset management but also maximises the value and utility of your video content.

Use the DAMS as a metadata driven engine. Forget, I repeat, forget about folder structures. That is what you are trying to get away from. Yes, some DAMS (such as AEM) use a folder format that is used primarily for workflow automation, security policy and Group/Role permission enhancements. They can also be used for ingest profile creation where any asset that goes into a specific folder has a profile associated against assets going into that folder. Successful DAMS implementations then use elegant User Interfaces to search based on metadata, categories, groups etc. You never want see folders again. The real work is defining taxonomy and metadata field usage. You will learn to capture way more metadata than just basic name, desc etc.

No matter how you deliver your assets. You want to have EVERYTHING backed up. Even if it's in a cloud, you want YOUR stuff backed up on your stuff. If you have large amounts of video, tape can be a pain. We always had all video (raw, final etc.) available on fast, network-based storage locally. Network storage was RAID with multiple hot spares in each tray. Hey, storage is cheap today, there is no reason to not have local, fast storage that covers this need. Anything less is just more work, and more risk.

Many 3rd party video platforms today support some connection to some DAMS. Look into it. If the vendor has a plugin, you are halfway to awesome things. At two of my previous employers, we had an in-house studio video-DAM that managed all RAW video and was tied to our edit suites and network storage. This DAMS was less in cost since it was built solely for just VIDEO and was built to be a robust back-end. I put in place a workflow where we pushed the final renditions to the regular DAMS direct from the Studio Video DAMS. The Enterprise DAMS would then replicate to the 3rd party delivery platform, passing along both assets and metadata to be ingested into the delivery platform. Use each piece of the architecture as it was intended.