Every US media leader I meet can tell me precisely how many subscribers they have. Far fewer can tell me what those same people watched, read, skipped or bought across all of their platforms last week.
That is not a failure of effort, and it is certainly not a failure of investment. It is the predictable result of how this industry grew up. A US media company today is really several businesses sharing a logo: national broadcast, local stations, a streaming service, a digital newsroom, an advertising operation, and increasingly a commerce and creator business. Each was built to win on its own terms. Each has its own systems, its own identity graph, its own working definition of an audience.
So, the signal is there. It is just scattered.
Last month at IBC, my colleague Silvia Candiani set out how we think about this. Orchestrating media intelligence means moving from isolated, single-purpose assistance to agent-led workflows that connect people, trusted data and specialized systems across the media lifecycle. Kevin Shatzkamer followed with the architecture underneath it.
Kevin Shatzkamer followed with the architecture underneath it. They introduced three frontier use cases. I want to spend this post on the first of them, increasing cross-platform audience and content intelligence, because in the US market it is where I see the shortest path from pilot to P&L.
Increasing cross-platform audience and content intelligence
I think we have been focused on the wrong problem.
US media companies are not short of audience data. Between broadcast, streaming, digital, social and commerce, most of the organizations I work with generate more signal in a week than they generated in a year a decade ago. What they don’t have is a governed way to let any of it touch anything else. The content team’s metadata can’t practically meet the ad team’s identity graph. The rights system doesn’t know what the recommendation engine is doing. Every one of those walls was built deliberately, and for good reason. Every one of them is now a tax on the business.
My colleague Simon Crownshaw, who leads media and entertainment for Microsoft worldwide, puts it more plainly:
“In media, your gold is where three things meet: content metadata, what the audience is doing, and your rights around all of it. In most places they live in three separate systems that never speak to each other.”
He’s right. In the US, those signals are often spread across even more systems, leaving each team to make decisions from only part of what the company knows.
So, the question I bring to customers is never ‘Do you have the data?’ It’s ‘What would you do differently tomorrow if your content, advertising, commerce and rights signals were allowed to inform each other, under a policy you actually trust?’
That reframing changes who needs to be in the room for strategic planning. Your “what’s next” meetings stop being a data engineering project and become a decision about governance, accountability, and where the business is genuinely willing to move.
The impact of connected data
Creative Artists Agency shows how connected data can change a high-stakes business decision.
For me, data comes down to three things: bringing all of our sources together—audience data, consumption trends, project performance, talent metadata, and more—making that information easy for the business to access and use, and turning it into actionable insights. With Microsoft Fabric and Azure Databricks, we’ve been able to do exactly that. And when I say our agents go into every negotiation prepared, with no information gap, I mean our actual agents, the human ones. They walk in with a clear understanding of a client’s value and the insights they need to focus on what matters most: creating the best opportunities and securing the best deals for our clients.
André Vargas, Chief Data Officer at Creative Artists Agency
Kantar rebuilt LINK AI on Azure to help marketers improve creative, not just score it. The new platform made batch scoring 20 times faster, cut operating costs fourfold, and made single-video scoring six times faster. Heath Greenfield, EVP of Solutions and Platforms at Kantar, says the work will help the company “grow our creative solutions portfolio and drive incremental revenue.”
The NBA extends the pattern across an entire sports and media ecosystem. The league unified data in Azure across broadcast, operations, development, its app and direct-to-consumer products. As Keith Cafferty, the NBA’s Associate Vice President of DTC Product, Technology & Operations, explains, “Our broadcast engineers, operations teams, and developers rely on data to make informed decisions. Unifying this data to deliver innovation is the foundation to our success.” That foundation now supports personalized experiences for fans in 60 languages across more than 200 countries.
These gains also show up at the industry level. Forrester’s Total Economic Impact study of AI for media and entertainment projects a 118 to 251 percent return.
Governance can be an accelerator
None of this works without the governance underneath it. Connecting signals raises real questions: consent and privacy, rights and licensing, measurement integrity, and who is accountable when a model informs a commercial decision. Those are not obstacles to the work. They are the work.
In my experience, the organizations moving fastest are the ones that settled their governance model first and then connected aggressively inside it. Permission, granted deliberately, turns out to be an accelerator. If you want the technical view of how that is built, Kevin’s companion post goes a level deeper.
Industry and enterprise conversations
If you are working through these questions, I would welcome the chance to continue the conversation at one of two upcoming events.
My team will be at NAB Show New York at the Javits Center on October 21 and 22, 2026, then I will be speaking at the Microsoft Frontier Tour in Philadelphia, PA on October 22, 2026. My team will be across the show floor and in sessions in New York, alongside partners including NVIDIA and Adobe.
Start with one decision your teams cannot make today because the signals sit in different places. Then work backward: what would need to connect, and what guardrails would you need to trust the answer? That is the conversation worth having.
The companies that benefit most over the next two years will not be the ones that bought the most of it. They will be the ones that connected their own data into something they can actually reason over: their own media intelligence. The data is the starting point. Everything else follows from it.
