Home CTV How Programmatic – And AI – Took Over The 2026 Winter Olympics

How Programmatic – And AI – Took Over The 2026 Winter Olympics

SHARE:
Digitally drawn image of a winter landscape with a skier on skiing slope, and with a forest and mountains in the background.

The action in Milano Cortina 2026 wasn’t limited to the slopes.

NBCU used this year’s Olympic Winter Games to lean further into programmatic buying for live sports at scale.

At the Paris Olympics in 2024, which was almost like a practice run, programmatic buying was only available through The Trade Desk. This year, NBCU worked with seven buy-side platforms: TTD again, Google’s DV360, FreeWheel’s Buyer Cloud (formerly Beeswax), Amazon DSP, Yahoo DSP, StackAdapt and Viant.

NBCU also made Olympics inventory available through Comcast’s Universal Ads self-serve platform for the first time and used AI for creative review and campaign optimization in near real time.

These upgrades helped create a better livestreaming experience for both advertisers and viewers, Ryan McConville, NBCU’s chief product officer and EVP of ad products and solutions, told AdExchanger.

As a result, live sports is becoming “more digitized than ever before,” he added.

According to NBCU, 200 net-new brands that had never advertised during the Olympics before bought programmatically during this year’s games.

Are we exclusive?

But a surge of demand often brings its own set of challenges.

Livestreaming – especially live sports streaming – is its own beast with specific nuances. For example, brands are willing to pay millions for category exclusivity.

Large brands can be especially strict about their expectations of exclusivity, especially during major cultural moments like the Olympics. If an airline, for instance, is a title sponsor of the Games, separation from other airline ads may not be enough. That sponsor might even demand separation from any commercials that simply include an airplane in the creative, McConville said, regardless of whether the brand has anything to do with travel.

Connecting programmatic pipes to Olympics inventory meant that “we needed to put up guardrails to look at bids and make sure [ads] actually qualify,” McConville said, including a combination of AI and human oversight.

During last month’s Games, NBCU used AI to scan and preview roughly 6,000 ad creatives to determine whether they met brand safety, exclusivity and category protection expectations. The AI would assign a “confidence score” based on how well a given creative met all requirements, McConville said.

If an ad scored 100%, it would automatically be earmarked for delivery. Ads that didn’t get a perfect score, but also weren’t automatically rejected by the AI, would be classified as “questionable” and set aside for manual human review on an hourly basis throughout the Games.

Pace yourself

NBCU also used AI for pacing and optimization.

Pacing is a long-standing challenge for livestreaming. Viewership can fluctuate drastically during broadcasts, especially for a sporting event that takes place over the course of 17 days.

For Milano Cortina, NBCU used what it calls a “yield agent” to manage reach and frequency during traffic spikes to ensure that brands didn’t miss the chance to get in front of viewers – but without bombarding viewers either.

NBCU’s yield agent optimized as many as 10,000 ad placements any given time, a task that used to require an entire operations team, McConville said. In addition to pacing ads evenly, “it saved thousands of hours,” he said.

AI was also handy for quickly turning around measurement reporting.

During the Winter Games, “we were able to see campaign performance within an hour or two,” said John Alleva, EVP of platform monetization, advertising and partnerships at NBCU. It used to take multiple hours or even days to get reporting back for certain digital environments depending on the amount of data involved.

With quicker reporting, NBCU could better optimize campaigns based on reach and frequency throughout the Games.

Ad optimization for live sports streaming is something “we had not really perfected before we started doing it in Paris,” McConville said. “But we were a well-oiled machine this time.”

During the Milan Cortina Games, NBCU processed the largest volume of data in Olympics history, according to McConville, “and we did it two times faster than we did in Paris.”

That may be a brag, but it’s also a testament to the rapid digitization of live TV – and especially sports.

Must Read

Kickbacks Takes An Outsider’s View While Bringing Ads To AI Agents

Andrew McCalip is a founding engineer at Varda Space Industries, where he oversees the manufacturing of things like hypersonic reentry vehicles and satellite buses.

CTV Buyers Are Getting The Show-Level Performance Optimization They’ve Always Wanted

A collaboration between InterMedia Advertising, Peer39 and Pontiac Intelligence provided show-level cost-per-acquisition data for 94% of CTV ad impressions.

Advertisers Await Programmatic Pause Ads

The IAB Tech Lab is working on standardizing programmatic signals for new streaming TV ad formats, including pause ads. Meanwhile, many brands are eager to add pause ads to their repertoire.

Privacy! Commerce! Connected TV! Read all about it. Subscribe to AdExchanger Newsletters

Why Media Mergers And Spin-Offs Don’t Always Keep Their Promises

With media megamergers, acquisitions and spin-offs left and right, the media landscape is changing at a pace that is difficult to keep up with.

TransUnion is partnering with Blockgraph so that advertisers can use its identity data to target, reach and measure TV households across channels.

How This Disaster Relief Nonprofit Tapped First-Party Data To Reach Donors Year-Round

Staying top of mind for potential donors is an ongoing challenge for Direct Relief. Nexxen’s audience curation helped it spread and sustain awareness.

Why Major UK Publishers Are Finally Joining Forces To Curate Ad Inventory

Atria’s collective approach is a response to growing monetization challenges and the need to protect the value of human journalism in the AI era.