Stability, not raw subscriber acquisition, has become the priority for pay-TV operators and streaming businesses. Deloitte data shows churn rates are climbing as high as 41%. Cancellation behavior is fluid. Consumer expectations around value and flexibility are higher than ever. Retention is top of the agenda.
At the same time, operators across pay TV, cable and telco are under pressure to become more than distributors. The vision to evolve into digital lifestyle providers – capturing a greater share of consumer spend – remains compelling. But the old levers are losing effectiveness. Bundles without a real value-add don’t win new subscribers. Blanket price increases won’t land in a market shaped by subscription fatigue and rising household costs.
NAB comes at a big moment for an industry fighting for profitability and revenue diversification. This year, one theme that will dominate is agentic AI.
Here are three predictions for what will spark the conversation.
1. Agentic AI becomes the orchestration layer
Agentic AI is moving from isolated use cases into something far more structural – an orchestration layer across systems, rather than just within them. Subscriber lifecycle management is an area where agentic AI is starting to deliver real gains.
In practical terms: a churn agent that detects early signs of disengagement and triggers a timely, personalized intervention. A pricing agent that continuously tests and refines offers based on real-time behavior. Or a support agent that resolves issues instantly, without requiring the customer to repeat information across channels. Revenue leakage detection that runs in real time rather than surfacing in a quarterly audit after the damage is already done.
This matters because analytics and personalization alone do not move the needle. AI models can now predict churn with over 94% accuracy, but the real value – and efficiency – comes from connecting that insight directly to autonomous action.
2. Entering the intelligence economy
The industry is now entering a new phase that can be described as the intelligence economy.
For decades, advantage was built in layers. First, owning the pipes: investing in infrastructure and distribution. Then, owning the experience: delivering compelling content and personalized interfaces that kept audiences engaged. Today, that is no longer enough. The next layer is intelligence – the ability to understand subscriber behavior in real time and act on it continuously.
This is where many AI investments will be tested. Success won’t come from deploying generic AI tools that lack domain expertise. It’ll come from embedding intelligence that understands the nuances of subscription businesses: pricing complexity, billing relationships, entitlement structures and regional regulations.
One constraint the industry needs to understand here is that agentic AI is only as good as the data it operates on. Operators who deploy AI on top of fragmented, inconsistent, poorly governed data will get wrong answers, miscalculated recommendations – and a quick path to failure. Agentic tools need to work from clean subscriber data, clear system boundaries, and well-defined entitlement logic.
We’ve been talking about AI at NAB for years. This year, expect a sharper focus on AI that is built with domain-specific expertise. Data alone has little value without the context needed to translate it into meaningful decisions.
3. Revenue velocity will win
Legacy systems continue to drain resources, stifle innovation and limit the ability to respond to changing market conditions. Reducing operational complexity through transformation remains a top priority. But at the same time, programmes measured in years and system replacements are the wrong instrument for a competitive environment that moves in quarters.
The metric that matters is revenue velocity: how quickly can an operator take a new service concept from decision to live, revenue-generating offer? For most today, that timeline runs to many months at minimum, once partner negotiations, billing integration, and go-to-market execution are all factored in. Operators that have invested in modular, API-driven monetization infrastructure are running that same cycle in weeks.
Bundling is a perfect example. Parks Associates data shows streaming video is now the most common add-on to broadband subscriptions globally. But the operators capturing that opportunity are not the ones with the most sophisticated architecture. They are those that can deploy a new streaming partnership, configure the offers, and get a personalized package in front of the right subscriber within weeks, not months. It’s all about speed and precision.
Ongoing M&A in media and entertainment makes this need all the more pressing. Large organizations that are navigating consolidated tech stacks need to know they can quickly pivot their billing, entitlement and customer management systems to activate new revenue streams while keeping subscriber experiences effortless.
The bottom line
The AI conversation will move from hype to a focus on practical, measurable results. Priorities will recalibrate toward stability, efficiency and smarter subscriber engagement. Agentic AI will be central to that shift – but its impact will depend on how effectively it is applied, and the data it is trained on.
We’ll be talking to streaming companies and operators about what it takes to really understand users, what’s next in pricing and personalization – and better, faster ways to monetize new services. The common thread? Treating subscriber management as a core competency rather than a back-end function.
