BrainBlog by Jason English
No other industry experiences seismic shifts as quickly or as frequently as the media and entertainment industry.
Whether the next disruption comes in the form of a technological leap in capabilities, or a change in how customers prefer to consume media on different devices, keeping a business ahead of competition in this arena requires both creativity and discipline.
To improve a business, improve its processes. Sounds simple enough, but there are thousands of possible ways to go about process transformation. Fortunately, there are some proven patterns to follow.
If virtually any software-and-data defined business process could be better automated, where do the firms that are most successful in process orchestration start? We gathered four winning process orchestration plays that have already yielded productive results for media and entertainment enterprises.
1. Cultivating massive server fleets
The first use case for scalable computing in media happened at the turn of the millennium. Maintaining the compute and storage power to render visual effects either required several high-end SiliconGraphics workstations, or renting a grid from a handful of service bureaus who could render frame-by-frame graphics on their dedicated servers with a disk array attached.
By comparison, the online media we create today may be far higher resolution, but the quest for scale is about much more than rendering clearer visuals. Media properties are far more omnichannel, and the lines have blurred between social media, ecommerce, and streaming video, meaning companies must consume a lot more digital infrastructure and data capacity to support the business.
The cost of compute and storage has come down drastically, so media companies can now lean on hybrid on-prem and cloud infrastructure to elastically scale up workloads and temporary data storage when needed, then shut them down when no longer needed.
The play: Akhil Ahjua, SRE at LinkedIn, discussed how they have as many as 340K queries per second with as many as 200K servers pushing well over two petabytes of data daily through its document store. At this scale, many system nodes, and the CPUs and hard drives that support them, are bound to fail over time under production loads.
They ran continuous monitoring, feeding offline events into Kafka, and creating tickets in JIRA for tracking setups and issue resolutions as engineers wrote custom code to manage and validate the many process steps and handoffs involved in replacing hardware and respawning clusters.
In order to make fleet management a more automated process, they scaled up Camunda as an orchestration platform to maintain the state and progress each job from detection to patching, builds and redeployments. Using this approach, within 6 months the team was able to eliminate a full month of maintenance labor per year for an SRE professional.
2. Customer onboarding and offloading
We used to consume almost 100% of our content through an aggregator—likely a conventional cable company that carried all of our channels. Now there’s a host of new ways to host, deliver, and sign up for content that increases our expectations of the process.
Because customers and businesses have so many options for consuming and distributing media now, it’s vitally important to make sure the engagement process is easy and seamless. Any hitch in the flow for signing up or adding services on any platform will cause a direct dropoff in business.
The play: Operative, a NYC-based creative SaaS platform for managing revenue and operations for linear and non-linear media and advertising companies, supports more than 25,000 users at more than 300 different media companies.
With so many unique businesses riding on their multi-tenant platform, onboarding new customers and managing account changes for existing ones was becoming a full-time task. Each new tenant might bring in their own media library, as well as their own integrations and data sources.
Operative called upon Camunda to load configuration templates for each tenant’s provisioning, billing rules, roles and task access privileges, based on a GraphQL back-end. Each client tenant can identify their own user profiles, custom configurations, task lists and data stores, so they can manage their own teams of users and share what works with others within their own organization without any possible interference from other teams.
3. Collaborative production review
For entertainment and media companies with high standards, there is a very detailed production pipeline for each and every piece of content that goes into the channel.
This is why most commonly used script and A/V editing tools already include some form of online client feedback capability. While it’s nice to be able to take in comments, the review data ultimately needed for measuring improvement is often tied up in these siloed applications as unstructured commentary.
There are dozens of steps that must happen over the lifecycle of any high quality piece of media, from initial planning to writing, production and editing, then determining exactly which delivery channels to post to, all the way to measuring audience sentiment and performance.
The play: A production company started by using intelligent document processing and conversational analysis to ingest a wealth of structured and unstructured data about each production in process. They applied Camunda’s process orchestration for gathering this inbound event data and managing alerts, handoffs, and approvals for each production in process. Now the media outlet can not only deliver new media to market faster, but they can perform decision modeling atop a common set of metadata in order to measure success and chart the path forward for new programming.
4. Managing rights and distribution
Back in the days of the VCR, licenses were unsuccessfully managed by copyright symbols and the FBI anti-piracy logo that ubiquitously appeared at the front of every video tape. Napster changed everything (before it was shut down) by sharing mp3 files over the Internet and waking up the industry to a new digital sharing threat.
Suddenly, every media company realized that digital rights management (DRM) was a critical feature needed for self-protection. Early forms of DRM used things like hardware dongles and scramblers, and were replaced by one-time passcodes and single-use tokens in media players.
Such arbitrarily restrictive countermeasures only turned customers against the industry, accelerating the proliferation of further copy piracy and anonymous download sites such as BitTorrent. Chasing down unauthorized views became a game of virtual whack-a-mole.
Fast forward our tape to the present, where new pressures include the widespread use of VPNs to get around regional access rules, and account sharing, which can make it hard to properly assess how successful any given piece of content is actually performing.
The play: A major content aggregator distributes media through subscription, metered (pay-per-view), and free ad-supported networks. By encouraging a standards-based data ontology and well-documented API libraries that feed their process automation, they were better able to collaborate with content providers, networks and advertisers. so that each party can monitor consumption and get predictable payments routed, based on performance.
The Intellyx Take
With new forms of media production and consumption arriving every year, the modern media and entertainment enterprise must constantly transform itself through process orchestration to compete for attention against millions of other options.
Any firm that fails to respond to customer desires or collaborative partner needs in a timely fashion can expect to become the next forgotten footnote in the mediaverse, ala Blockbuster Video or Quibi (remember that boondoggle?).
Fortunately, executives and development teams can deploy a proven set of process orchestration practices that when tied to real-time usage data and integrated with partner services, can provide business resiliency, even when the current models are disrupted again.
©2023 Intellyx LLC. Intellyx retains editorial control over this content. At the time of writing, Camunda is an Intellyx customer. No AI chatbots were used to write this article. Image source: Adobe Image Create, Dark media jungle with vines.
