Why major events expose the limits of traditional news production

Peak news events quickly show how resilient your crew and your technology stack actually are. And if they’re not resilient—and able to rise to the moment—the quality of your news broadcast and viewer trust can quickly be impacted. But that’s not to say that you need to spend an exorbitant amount of time, money, and energy creating the perfect newsroom just to cater to a single news event. That’s not fiscally responsible and puts a great deal of strain on your operations.

Let’s take NBC News as an example. The company’s 2024 election night operation spanned 15 studios and 13 control rooms across NBC, MSNBC, CNBC, Telemundo, and NBC News Now. More than 100 cameras, with over 100 channels of graphic rendering. To make this possible, the engineering team built a custom web application from scratch because nothing off-the-shelf could synchronize that many production environments at once.

Once election night was over, the news operation changed significantly. Even at NBC’s scale, building custom systems for short-lived peak events adds cost and complexity. For smaller news teams, that model is even harder to sustain.

Every news organization faces some version of this problem. A peak coverage event arrives: an election, the Olympics, a major breaking story. And demand for production capacity spikes far beyond what the infrastructure was built to handle.

The only option is to stand up temporary control rooms, rent trucks and flypacks, and repurpose studios; and pull staff from their regular assignments and hope the integrations hold.

It works because it has to. But it’s getting harder to sustain over time because the infrastructure these surges are built on top of is stretched to its limits.

The defining characteristic of a peak coverage event, whether it’s an election, an Olympics, or a fast-moving national story, is that production demand doesn’t increase incrementally. It surges. The EBU’s Dynamic Media Facility reference architecture and SVG Europe’s operational benchmarks put the figure at three to four times normal capacity for major events.

The 2024 election cycle illustrated what that multiple looks like in practice.

• At 30 Rock, NBC’s build required eight control rooms, six studios, over 50 cameras, and nearly 200 render engines—just for the core operation. That’s before accounting for MSNBC, CNBC, and Telemundo, which brought the group-wide total to 13 control rooms and 15 studios.
• CBS converted its CBS Mornings studio in Times Square into the “America Decides” Election Headquarters with 39 dedicated render engines. ABC built its election set inside Studio TV2 using rented modular frames, LED walls, and outside equipment layered on top of permanent infrastructure.
• On the other side of the Atlantic, the BBC faced a different kind of scaling challenge for the 2024 UK general election. The team needed to ingest 369 simultaneous live feeds from constituency counts nationwide. Rather than replicating hardware, the BBC’s engineering team used a cloud platform to scale the feeds and deliver outputs into the existing broadcast chain. As BBC News’ Geraint Thomas put it, the cloud approach “allowed us to scale up a vast number of feeds and handle peak traffic seamlessly without investing in additional hardware.”

Each of these examples is a different response to the same underlying constraints of modern newsrooms. Peak events required major infrastructure upgrades, but their permanent infrastructure wasn’t built to handle those additions effectively.

The reason peak events require so much improvisation comes down to how production switching infrastructure is designed. A traditional production switcher is a fixed-configuration system. A 3 ME switcher is always a 3 ME switcher.  A traditional production switcher is typically sized for the most demanding show it is expected to support. If election night requires more capacity than the system was designed to provide, the newsroom may need to add, install, and integrate additional hardware. The existing switcher cannot simply expand temporarily to meet the increase in demand.

In a multi-control-room news operation, each room is typically equipped and sized for the most demanding production it may need to support. When one control room is operating below capacity, however, its unused switching and processing resources generally cannot be reassigned to another room or production.

Thomas Gunkel, market director for broadcast at Skyline Communications, highlighted the scale of this utilization challenge in a 2025 analysis of broadcast infrastructure. Looking specifically at ingest systems, he found that actual utilization can fall below 10 percent of theoretical operating capacity.

While that figure does not measure production switching directly, it illustrates a broader issue across broadcast facilities: infrastructure is often provisioned for occasional peaks, leaving significant capacity underused during normal operations. In fixed production environments, that unused capacity also cannot easily be reassigned to another control room or production when demand shifts.

This stands in stark contrast to how IT infrastructure has evolved over the past two decades. Computing moved from dedicated servers per application to virtualized, pooled resources that scale with demand. No IT team would provision a dedicated physical server for an application that runs at full load two nights a year. But that’s essentially what broadcast production infrastructure still requires.

Each control room is its own island, sized for its own peak, with no mechanism to share or borrow resources across the operation.

In practice, that means M/E banks, keyers, and processing capacity that can be created, shared, and reassigned across control rooms from a common pool. And this is possible today. Hyperconverged platforms pursue a similar goal to the MXL project: making production resources more flexible, shareable, and responsive to changing demand. They take a different technical approach, combining routing, switching, and processing on a unified hardware platform and using software to configure, license, and dynamically allocate those resources.

Broadcasters, sports networks, and enterprise production teams are adopting them. But the transition is early, and it’s important to be honest about where things stand. Most news control rooms still rely heavily on SDI, particularly in North America, although hybrid SDI and IP environments are becoming more common.

The credible near-term path isn’t necessarily a full cloud-native or IP transition. It’s pooling on-premises production resources, using dynamic licensing to flex for peak events, and introducing IP where it delivers a clear operational benefit.