Appendix A: The Common Workflows
Determining which workflow is most suited to cloud migration:
For our purposes, the media supply chain starts with receiving a piece of content from a producer. That piece of content may have existed before this point—as a record, some metadata or rights—but the media supply chain begins where you take it on board, and just accepting content from producers can be a painfully manual task in a traditional media supply chain.
Once content is received it needs to be checked for validity. Does it meet expectations and technical specifications? Is the metadata complete? Does it include all the component parts needed such as languages, subtitles and so forth? Does it pass technical quality tests? Assuming it conforms to all these required conditions, it may still have to be transcoded to a house format, depending on the incoming formats you’re willing to accept.
There are a lot of steps involved in just receiving, acknowledging and accepting content into a facility. Depending on the scale of the operation and how many independent producers contribute content, this could be a huge, labor intensive exercise. How efficient is it today? How many people are involved? What are the failure rates of different providers? Do you even know? How long does it take to accept each item? For the vast majority of people, these are often unanswerable questions because there’s no real way to measure or get analytic insight in to the processes.
If we go back to the left-to-right model described in Step 1. of “Cloud Migration in Five Steps,” content receipt is a logical place to start. If your current mode of operation includes manual QC and having people watch content in real time, for example, starting the migration with content receipt can significantly increase opportunities to automate as well as eliminating many of the the complexities around it—including format, quality and metadata validation. Remember the evaluation guide for “Identifying the Processes for Automation” in Appendix B.
It’s also a logical starting point if the objective from the get-go is a completely cloud-native media supply chain. A migration that begins with content receipt can be achieved using an online upload portal to the cloud of choice, where files can be inspected, corrected, accepted or rejected automatically according to specification. Content suppliers can be onboarded over time while the next logical step is being executed.
All of the typical complications and obstacles are eliminated in a cloud-native content-receipt environment. Not only are manual processes automated, but automated in a way in which they only incur cost when active. Plus there is full access to analytical data, any way you want to slice it.
1st Place: With most media supply chains, content receipt is the most logical place to start migrating to a cloud-native environment.
Many media organizations have a preferred house format for their master media—something that’s undergone serious deliberation around quality, efficiency and interoperability with other systems. It’s unlikely, however, that all content arrives in the preferred house format. Most media operations receive content in several formats from many different places. Even with strictly enforced content delivery specifications in place, non-compliant content will still be submitted, and when push comes to shove, it will be accepted if deadlines are tight and there is no other option.
All this leads to the need to “normalize” content to the preferred house format—to make everything the same and restore order. In a traditional media supply chain, this means selecting a transcode vendor and making sure their technology can deal with all the incoming and outgoing formats—those you deal with now and possibly in the near future. Yet another new producer is right around the corner. If the existing transcoding technology can’t handle it, another is added. And another. Before you know it, you have one of each, all underutilized.
Within a cloud-native environment, transcoding and other tools can be selected from any of several vendors on a per-usage basis. Welcome to the world of utility transcoding.
If you already receive your content in the cloud, this is a great place to start your media supply chain migration. No more picking transcode vendors one at a time and being limited by scale. Just pick the one you want to use for each content source and automatically route content through it, making everything the same. Normalizing in the cloud also creates the beginning of a cloud-based archive.
3rd Place: We consider normalization the third most practical place to begin a cloud migration.
The primary goal of content logging is to enrich the content metadata to make it more searchable, more organized, and easier to monetize at any point in the future. This could include production metadata; process metadata delivered by the tools that have operated on the content; descriptive metadata added by operators; or AI data, including things like object tags, face detection, transcription etc.
While content logging is inherently easier to perform in a cloud environment because of the broad availability of different tools, it’s not often one would start a media supply chain migration here unless the goal was to create a rich cloud-based archive.
Tie for 5th Place: Not really a logical starting place in that you’ll be moving content into and out of the cloud, though it may be useful with widely and diversely distributed content. However, once your content is cloud resident, this is an easy add-on to provide value to your content and processes.
Anyone distributing content globally is familiar with the localization problem—sometimes referred to as “compliance.” Every region that content goes to likely has its own regulatory restrictions, meaning it must be modified multiple times for each destination. Adding languages and captions is the small part of the process, but what about removing nudity, violence, objectionable language or other material not permitted for some locations?
Public cloud-based artificial intelligence services already offer a lot of value in this area, as do specialized AI tools developed for the media industry. Having computers detect and identify these areas of concern and then guide operators through the compliance process can save a great deal of manual processing time.
Plus, if all your content is already in the cloud, taking advantage of these services is as simple as adding an option to your media supply chain. We’ve even worked with customers who have specifically moved their content to the cloud to analyze it using AI services in order to increase the reach and marketability of their archive.
Tie for 5th Place: Not really a logical starting place in that you’ll be moving content into and out of the cloud, but may be useful with widely and diversely distributed content. Here again, once your content is there, adding the capability is both simple and valuable.
Other content modification processes may be necessary before distribution. Whether that’s removing ad breaks for VOD platforms, adding ad breaks for linear distribution, trimming durations for specific purposes or even creating promos specific to each platform—content will often need to be modified.
In a cloud-based environment, avoiding egress is avoiding additional cost. If you need to modify content that’s already in the cloud, there are ways to do in without taking it out of the cloud. This may involve leveraging technology such as PC-over-IP, remote or proxy-based editing or other cloud-native tools.
It’s important to make sure these steps are tracked within the automated media supply chain, particularly manual interventions or modifications. While modification is perhaps not the ideal place to start your cloud journey, it’s a critical and simple addition once your media is in the cloud.
4th Place: Not really a logical starting place in that you’ll be moving content into and out of the cloud, but may be useful with widely and diversely distributed content.
Distribution used to mean getting program and ad content ready for a playout automation system feeding linear channels. Now, it can mean supporting the distribution of content to as many as 200 disparate platforms that may be linear, on-demand, electronic sell-through, over-the-top, mobile, international or even 4K with a combination of languages and closed captioning and audio formats. It’s a lost revenue opportunity when a media distribution system can’t modify content sufficiently to meet the requirements of a new platform.
Traditional on-premise workflow systems often fall short here because of their fixed nature. Accommodating new and emerging platforms can be an exercise in retrofitting the workflow again and again. Even so, these systems still can’t scale quickly enough to meet the inherent fluctuations in supply and demand.
Cloud-based media supply chains can offer significant advantages for distribution—particularly multiplatform distribution—in terms of being able to scale the output up and down as you need it, the elasticity to try new configurations and the cost-predictability to do so. Distributing media from the cloud allows you to try out individual tools and services on demand, providing unprecedented flexibility to adapt as new distribution platforms emerge.
2nd Place: We rank distribution as the second most practical place to start, after content receipt. It has the biggest payoff, and you could do everything up to this point on-prem, move master content to the cloud prior to distribution, and publish from there.
The cloud is a universal access point for retrieval. Once content is in the cloud—and once it is analyzed, logged and stored—retrieving stored content for future opportunities becomes as simple as running a supply chain on it.
Once you are working with content in the cloud, creating a cloud archive is a practical and relatively simple move. With deep cloud storage starting from as little as $1 per terabyte per month, its impossible to beat this price with an on-premise archive.
If you are undertaking any of the previous workflows in the cloud, adding an archive step is a simple and painless thing to do. And if you want to start with the archive, all you have to do is work out how to get the content from its existing location to your chosen cloud vendor, and there are multiple solutions to help with that.
Honorable Mention: Starting a media supply chain cloud migration with archiving may be preferable when most of the content is long-tail and has an afterlife on multiple platforms, or where the existing on-premise archive solution is in a replacement cycle. Even when the migration is started in another area of the supply chain, archiving into the cloud will become more and more practical as the migration progresses.