Broadcasting in the Cloud – Deloitte On Cloud Blog | Deloitte US has been saved
Limited functionality available
Cloud has radically transformed the operating models of organizations across industries, providing the flexibility, agility, and scalability that’s needed to quickly adapt to changing demands, innovate, and bring new offerings to market faster. In the broadcast industry, cloud has also enabled organizations to embrace AI and machine learning and operate with the elasticity and scalability that was not previously possible.
So why, despite years of cloud aspirations, has the broadcast industry only recently started migrating the core of their content distribution infrastructure into cloud platforms? The answer to this question rests with the immense bandwidth and storage requirements and extremely low latency needed by highly specialized software ecosystems. In recent years, cloud service providers have addressed many of the barriers preventing broadcaster adoption: Latency, bandwidth, and vendor ecosystem support to name a few. Indeed, the time is ripe for media companies to leverage cloud to operate more efficiently and cost effectively, and to sate viewers’ appetite for more content, delivered in new ways.
Why cloud, why now?
As customer demands for content delivery rise, broadcasters are struggling to keep pace. Consumers are looking for more content—where and when they want it—and media companies are eager to make that possible. The gap between broadcasts’ current capabilities and growing expectations points to cloud as a promising solution to elevate the customer experience. To understand the value of a cloud broadcast migration, it’s worthwhile to scope the discussion. Broadcast is the highly complex and technical series of systems, processes, and activities facilitating the flow of picture, sounds, graphics, and metadata from a live content source into a central Master Control operation responsible for the final “packaging” of the broadcast stream for distribution to cable companies, OTT providers, and broadcast transmissions.
Cloud can enable media companies to transform these highly complex broadcast operating models to function with greater resiliency and elasticity. And for those companies considering adoption, cloud can provide substantial benefits. Some advantages include helping broadcasters manage workloads more effectivity, innovate faster and get new content to market quicker, as well as allow for improved content flow and more content delivery options.
Meeting the challenges: How cloud helps transform content delivery
To meet rising customer demands, media companies need to address fundamental challenges in four key areas: Infrastructure, data and insights, talent, and cyber security. Adopting cloud can help broadcasters tackle these challenges head on and outpace competition.
The first challenge is the need for improved cost control and operational agility. Broadcasters increasingly need to deliver more content, more efficiently across multiple platforms with increased scale and resiliency. However, they’re often hampered by legacy workflow inefficiencies, expensive legacy technologies, and a lack of integration between people and technology.
With a move to cloud, broadcasters can worry less about the underlying core infrastructure scale and maintainability and focus more on the business of broadcast. Elasticity and resiliency represent a significant portion of a broadcaster’s infrastructure costs. When scaling a broadcast operation to its peak-utilization point (special events, sports seasons, etc.) and factoring instant-failover for resiliency, the elasticity and in-built resiliency of cloud becomes very attractive. Because cloud vendors provide infrastructure capabilities as-a-service, content providers can pay for only those services they need, when they need them, and gain cost-effective means to store, access, and deliver content with very low latency, even in the event of a service provider’s datacenter going down.
Data and insights
There’s also a critical need for better insights into business performance. Yet, media companies often have difficulty converting massive amounts of raw data into actionable insights because data is often inaccessible, there’s limited visibility into viewer habits, and their current analytics applications aren’t optimized for customer personalization.
With the cost-effective data storage that cloud provides—coupled with as-a-service analytics and AI/ML—media companies can turn the mountains of data they collect on viewer behavior and market trends into actionable insights. Those insights can lead to the development of more effective content-delivery models that drive better performance and increased revenue.
Evolving viewer habits require new content-delivery methods, but talent is an issue. There is a large skill gap from legacy broadcast to modern cloud-native broadcast technology architecture that needs to be addressed. A large portion of the highly-skilled, specialized broadcast workforce doesn’t have digital broadcasting skills—they’re used to plugging in cables, not APIs. Broadcasters that want to digitally transform often find it difficult to backfill specialized roles and acquire new, cloud-skilled workers to a work environment that may be considered outdated.
Migrating to cloud is a great opportunity to re-evaluate and improve talent models. Companies can help workers reskill, or upskill, and they can use their digital transformation to develop a new talent strategy and operating model to attract and retain the talent they need to operate in cloud.
Finally, media companies want to expand their offerings and channels, but expanded delivery can increase the risk of piracy and copyright infringement. There are also increasing strains on resources that monitor content rights and IP infringement violations. Further, there’s still uncertainty among some organizations about security in cloud as it relates to resiliency versus on-premise control.
Cloud security is paramount for cloud providers. Because of the resources cloud providers deploy to keep data sets and infrastructures secure, data and applications stored in cloud are inherently more secure than those stored on-premises at most companies. Thus, digital rights management and IP protection are more easily managed, and the threat of security breaches is lessened.
Wrapping it up
Cloud has been nothing less than transformative for those organizations that have adopted it. The flexibility, scalability, and innovation it affords organizations lead to improved performance and deeper relationships with customers. While other industries have enjoyed a couple decades of cloud availability for their business, broadcast cloud is new and media companies that embrace it now will inevitably gain an advantage in the industry. Savvy media broadcasting companies are beginning to reap these benefits for themselves. As cloud adoption grows, the benefits will only increase. Those media companies who are early adopters will accelerate time-to-value on those benefits, and they’ll get the chance to leapfrog their competition and unlock new opportunities.
Mike is the cloud leader for Deloitte Consulting LLP’s Technology, Media, & Telecom industry practice, he works with global organizations to help them transform their organizations through cloud-based solutions. Mike has over twenty years of experience managing enterprise technology systems, including on premise and cloud infrastructure for leading companies in media and entertainment like NBC Universal, Time Warner, and Warner Bros. He has been involved in many large transformation projects including SAP ERP implementations. His industry experience also includes work with public utilities. Mike earned a bachelor’s degree from Kansas State University. He has three children and lives with his wife in Castaic, California. He and his wife enjoy organizing the annual Paseo Club Triathlon in his community to support A Light of Hope Charity.
Considering building a data ingestion and preprocessing pipeline to train a machine learning model? Did you know that there are specific design considerations that we need to think about when we are building a data pipeline to train a Machine Learning model?