What’s Working

Broadsign sits across a genuinely broad slice of the global DOOH ecosystem publishers, platforms, and programmatic pipes so we get a pretty honest view of where AI is actually moving the needle.

The two areas where we see real, consistent value today are yield optimization and operational reliability. On yield, AI-driven forecasting is helping media owners move past static rate cards toward models that factor in demand patterns and day-part dynamics in a way that just wasn’t possible manually. On operations, AI is doing some of its most useful work quietly in the background, monitoring screen health, flagging scheduling issues, and verifying proof of play. It’s less exciting to talk about, but the compounding effect over time is significant. Publishers who have got this right are spending far less time firefighting and more time focused on growth.

‍

Best-Practice Integration

What separates genuine AI integration from a “surface layer” is whether the AI is connected to a decision or just a report. Surface-layer AI is just a dashboard you look at; best-practice integration is a tool you use to take action.

‍

We see the most success when AI is used for Prescriptive Analytics and Simulation. This means the system doesn’t just show you that revenue is down; it identifies the specific risk, like a fill rate drop in a specific market and provides an immediate recommendation to fix it, such as adjusting a programmatic floor price. Even better is the “What-If” simulation. Best-practice platforms allow publishers to simulate different channel allocations (Direct vs. Programmatic) and see the projected revenue impact in real-time. When AI helps you visualize the outcome of a decision before you make it, it becomes an essential part of the workflow.

‍

The Data and Infrastructure Foundation

If there’s one thing we’d say to any publisher thinking seriously about AI, it’s this: don’t rush the foundation. AI applied on top of fragmented, inconsistent data doesn’t give you better outcomes, it gives you the same mistakes, faster and at a greater scale.

‍

The publishers who are genuinely AI-ready are those who have done the hard work of standardizing their taxonomy and ensuring their CMS and SSP are perfectly synced. We need to be honest about this variance; layering AI on top of shaky data isn’t going to accelerate the journey; it’s going to create new problems. Get your “data house” in order first, standardize your impression methodology and clean up your inventory records and then let the AI drive the growth.

‍

What’s Not Working

There is a fair amount of “AI-washing” happening right now, and the industry needs to be direct about where the promise and reality are misaligned.

- The Hyper-Personalization Gap: The biggest gap is the push for “one-to-one” AI messaging. While this is often promised, the reality of OOH is its inherent strength as a one-to-many medium. AI is far better used for driving contextual relevance (weather, traffic, mood) than trying to chase individual identities, which often hit privacy or technical roadblocks.

- Generative Creative Overkill: There is an oversell on Gen-AI for creative. While it’s great for versioning, using it without strict human-in-the-loop guardrails can lead to “AI Slop”, content that doesn’t fit the premium, high-impact nature of DOOH. We need to ensure AI supports the medium’s premium status rather than diluting it.

‍

Looking Ahead

In the next 12 to 18 months, I expect AI to take us toward the “Total Yield” era. We are moving past the siloed management of static and digital inventory; AI will act as the bridge, allowing publishers to unify audience data across all assets to provide a “Total Network” view that hasn’t been possible manually.

‍

We also expect to see the rise of Generative Business Intelligence. Instead of digging through complex reports to understand performance, publishers will be able to use natural language to ask their platforms, “Why is my Dubai fill rate lower than historical averages?” or “How can I adjust my allocation to stay on target?”.

The future isn’t about AI replacing the human element of OOH; it’s about AI removing the friction from data management so we can focus on having more impactful and meaningful conversations with agencies and media buyers.