The Marketing Dashboard Is Entering Its AI Phase

By Mark Zagorski Chief Executive Officer at DoubleVerify

To paraphrase an oft-used quote: “Reports on the death of dashboards have been greatly exaggerated.” The incessant drumbeat that AI is eating everything, including how marketers engage with data and insights, is relentless and dramatic, and will undoubtedly be a theme at this year’s Advertising Week. The sentiment is not without merit, as it is clear that we need to rethink not just what dashboards can do (see today’s overstuffed UIs), but what they should do and how AI can enable that.  

For years, marketers have relied on dashboards to answer fairly straightforward questions: What happened? Where did my money go? Who saw my ads? Did they work? As ad spend proliferated across more channels, platforms and media types, leveraging a growing list of data sources, our solution has largely been to “stuff more stuff” into the dashboard. More charts, more tables, more variables, more filters. More of everything but clarity. 

Conversational AI offers a different way to approach the problem, which flips the usual engagement with data on its head. Instead of a “push,” it fosters a “pull,” where a marketer can simply ask: Why did performance fall yesterday? Where are we losing efficiency? What changed this week that I should care about?

This is the obvious direction that nearly all in the industry believe we are heading. However, I’m not convinced that just asking better questions is where this evolution ends.

The Blank Prompt Problem

There’s a familiar moment (of panic?) when you open a generative AI tool and see an empty prompt box: Where do you start? We risk bringing that same problem into marketing platforms by replacing dozens of meticulously constructed dashboard charts and controls with a blinking cursor in a blank box and expecting users to know exactly what to ask.

Other industries are already seeing this. Starling Bank has described “writer’s block” as a barrier to AI adoption, so they introduced tools to help customers ask the right questions while also enabling AI to respond to queries and take actions on their behalf. Marketers may face a similar challenge: how do you recognize that performance has changed without immediately knowing where to look for the cause?

That’s why I think the more interesting evolution is toward guided intelligence. The best systems won’t just wait for the perfect question. They’ll help users understand what’s worth asking in the first place.

From Reporting to Guided Intelligence

Imagine opening a marketing platform in the morning and, instead of scanning a dozen charts, you see something like this: Three things changed yesterday. Two are within normal ranges. The third could meaningfully affect campaign performance. These prompts would be based on a set of criteria that you told the system were important to you. No dazzling bar charts to dig through, no pivot tables to analyze. Just the good stuff. From there, the system could help answer a few basic questions: What actually requires my attention? Why is it happening? And what can I do about it? Spotting anomalies isn’t necessarily the difficult part. We’ve been doing versions of that for years. What’s harder is understanding which changes will truly impact what I am  trying to accomplish.

A CPM increase, for example, could be a problem. Or it could be completely acceptable if it comes with better quality, stronger performance or more valuable inventory. Making that distinction requires context, reliable signals and an understanding of the advertiser’s objectives. AI can move this process along and is starting to help close the loop.

And Then, Action

This is where the shift toward agentic AI starts to matter. Once insights have been delivered and questions answered, what’s next?  AI is moving beyond explanation and recommendation and into the realm of action. Delegating some decisions to intelligent systems is the natural progression of the evolution of the dashless dashboard.

This shift marks a fundamental change in the definition of a “dashboard”. It is becoming less of a place to decide what to do next and more a place to oversee what AI is doing on a marketer’s behalf. IAB’s 2026 Outlook Study found that two-thirds of buyers are focused on agentic AI for ad buying and campaign execution, but greater delegation will depend on trust: marketers need confidence that systems understand the objective, rely on trusted signals and operate within clear boundaries.

The question is no longer whether AI can or should act, but how much responsibility marketers are willing to give it.

The Dashboard Is Changing Jobs

Although the days of the dazzling dashboard are rapidly coming to an end, it doesn’t mean that marketers stop needing a place to engage with and act on data. We’ll still need shared views of performance to monitor trends, compare periods, align teams and dig deeper when something doesn’t look right. How we receive those insights and the role the dashboard plays in what happens next will be substantially different. For decades, we’ve largely asked our dashboards to show us what happened. Increasingly, I think we’ll expect them to tell us what deserves our attention, help us understand why and figure out what to do next. In some cases, they may eventually take that next step for us.

The dashboard isn’t disappearing. It’s becoming the place where marketers and intelligent systems work together. And that may ultimately be a much more useful job.