Defining workflow

The broadcast industry has never been one to shy away from buzzwords. The term “workflow” topped the list in 2011, capturing the attention of virtually every broadcaster — and every manufacturer working hard to adapt their digital platforms to quickly evolving requirements.

But workflow, perhaps more than any preceding industry buzzword, causes a great deal of confusion due to its many different meanings. A quick survey of station managers and technical engineers that inquires to the definition of workflow will prompt such responses as “asset viability,” “efficiency,” “monetization,” “program normalization,” “platform-agnostic” and “standards-based,” among others.

Most would agree that workflow is a process that encompasses many small, definable tasks that link together to achieve a desired result. Workflow also can be defined as the progress performed. Automation is more than just controlling broadcast devices.

The future: “Smart workflow”

Workflow in itself is just a process. Although careful definition and construction of the process can provide some efficiency, at what point does it move from being purely mechanical to being a “smart workflow,” adding real value and effectiveness? That gain in efficiency comes when workflows are automated.

Automation encompasses much of what people describe as efficient workflow. It is more than just controlling broadcast devices. It requires that a system have knowledge not only of tasks but also the procedural steps, organizations and people involved, required I/O information, and tools needed for each stage in the business process. Automation not only manages the process but also monitors information about status, potential bottlenecks and workloads, providing information that is crucial for effective day-to-day decision making. Automation and knowledge can turn a workflow into a smart workflow.

The explosion of metadata

As processing power has multiplied and bandwidth has grown, broadcasters must consider the many new ways their content can be delivered. This proliferation of platforms has also caused an explosion of content metadata — to the point where metadata often is scattered among several systems and people. This creates islands of metadata that are sometimes redundant, sometimes siloed and often completely wasted.

Solving the metadata challenge

Smart workflows need information, and sharing data is crucial to the process. A smart workflow uses this data to make decisions and route content appropriately. From this point, a rules-based engine can initiate and manage events, actions, resources, tasks and their relationships along many paths.

The first step to establishing this interoperability is to evaluate the content workflow. The ability to gain a detailed understanding of workflow — including production, programming, sales, traffic, ingest and playout components — will help clarify how elements relate to each other within the environment. Build a diagram that captures each step of the process, from the trigger that starts the workflow through to the metadata required to complete the steps — and the manpower commitments along the way, including tasks performed. It is critical to pay close attention to what metadata is originated and how much metadata is captured manually.

Step two is an analysis of the workflow, starting with the metadata. Look for data that is captured multiple times and for data that doesn't seem to fall into the flow.

Communication and usage of scheduling data is often a workflow bottleneck. Schedule data such as format requirements, playout rights, and time and date windows often are communicated in inefficient ways. Schedules requiring multiple versions with different playout formats should automatically trigger workflow steps to create those versions at the right points.

Also consider the need for information that might not show up on the standard workflow while analyzing the metadata. The facility's standards and practices approver will not have to make a phone call to find out the schedule windows for questionable ad copy if the captured metadata was automatically inserted into the workflow. And the metadata that signals content has arrived and is available for viewing makes life simpler for those who must find and view pieces of content.

Validation of the process is ideal once the metadata flow and capture has been analyzed. This will clarify which steps in the overall process are necessary. Automated tools can find and detect many problems that in the past required manual intervention.

Loudness is one example. A smart workflow can instruct a system to create a new version of the content with the loudness problem corrected.

A second example is a content length issue. Say that a piece of content scheduling to air in the morning is identified as 30 seconds long, but it seems to be two seconds. An alert message is immediately sent to the operator's mobile phone detailing the problem — a problem that can now be solved by the proper decision maker.

The idea is to find a comfortable balance of automated and manual process steps that do not ignore the savings that come with task automation. This is particularly true when evaluating how you prepare and move material for nonlinear playout.

Consider the idea of and efficiencies found with a “just-in-time” production process. The just-in-time process allows the workflow to use scheduling metadata and destination requirements to construct specialized versions for each platform. In doing so, it may choose to swap infrastructure or cloud components based upon availability and costs to maximize device capabilities across the content chain. Any format to any destination at the lowest cost is a goal for the smart workflow.

The process of analyzing and redefining workflow requirements often identifies inefficiencies that can be resolved through a simple process change. It is surprising how many historical process steps are no longer needed.

Automation is key

Consider overlaying workflows with automation tools once needs are defined. This may include the management of media, tasks or people resources. Unnecessary processes often can be eliminated, streamlining workflows and freeing up valuable time and resources.

Automated workflows must provide management by exception to be efficient. They must also offer clear visual fields that present necessary data without moving among multiple displays or data screens.

As distribution mechanisms evolve, operators need to easily visualize the processes of file-based workflows and the distribution of the content to the proliferation of platforms, while at the same time being able to monitor content preparation and playout. The result is higher control of the automation of devices, but more importantly, actually automating the production of that content.

Intelligent content workflows can deliver real financial and resource benefits when implemented across multiple content distribution platforms. By automating processes and communicating the appropriate information and metadata to the best decision makers at the optimal points in the process is the way to fulfill the true promise of the workflow concept.

Mark Darlow is senior product line manager, Automation and Digital Asset Management, for Harris.