AI and Advertising ROI

Getty Images - Narumon Bowonkitwanchai

As AI accelerates execution, reshapes discovery, and fragments customer journeys across channels, marketers are redefining success around measurable outcomes, making performance the central currency of modern advertising.

For years advertising was measured by inputs: time spent, campaigns launched, and budgets deployed. Success was often inferred through metrics like reach, impressions, or brand lift, with true business impact assessed only after the fact.

That model no longer holds. Today, marketing is expected to deliver measurable outcomes in real time, with the same precision and accountability as any other revenue-driving function. Performance, defined by tangible business results, has become the standard by which success is judged.

This shift is being driven by artificial intelligence (AI), which is compressing timelines, increasing precision, and raising expectations for measurable impact across every stage of marketing. As AI becomes embedded across planning, execution, and optimization, it is fundamentally changing how marketing operates.

“Performance has become a common language between marketing and the rest of the business,” said Todd Parsons, Chief Product Officer and President of Performance Media at Criteo. “When you can tie activity directly to outcomes, marketing operates less like a cost center and more like a growth engine.”

At the same time, the environments where performance is created are becoming more complex. According to a recent survey from my company, Prosper Insights & Analytics, 54% of generative AI users now use it for search, introducing a new layer to how products are discovered and evaluated. Retailers are already seeing that when discovery moves into AI assistants, performance moves with it: Albertsons’s, for example, has reported that shoppers who use its “Ask AI” shopping assistant are already delivering a 10% increase in basket size for customers who have started to use it. For marketers, this expands the number of signals shaping any given outcome and makes it harder to manage within any single channel.

But this shift is not only about transactions. It is also about where discovery begins. As generative AI becomes a more common entry point, consumers are increasingly encountering brands earlier in the journey, often before they have expressed clear intent.

“AI assistants can be high-intent environments, but they’re not only that,” Parsons said. “Much of the interaction is still exploratory, with people working through what they think and feel before they’re ready to do—and that’s reshaping how true discovery happens.”

Together, these dynamics are redefining how marketing operates and reinforcing why performance has become its central currency.

The Compression of Campaign Lifecycles

AI-driven automation is transforming performance marketing from a linear process into a continuous system of iteration. In this environment, time becomes a constraint. Campaigns that once unfolded over weeks must now evolve in real time as performance signals emerge and shift. This changes the speed and the operating cadence of marketing itself.

Traditionally, creative development was a hurdle that marketers had to overcome. Iteration required coordination across teams, layered approvals, and manual execution, making it difficult to adjust messaging at the pace of the market. But now, AI is removing much of that friction by compressing production cycles and enabling parallel experimentation at scale. As a result, creative shifts from a fixed input in a campaign structure to a dynamic variable that adapts in response to performance data.

“Automation is one thing. The real change is timing,” Parsons explained. “Decisions that used to happen after a campaign are now happening during it. That fundamentally alters what optimization means.”

Performance, in this context, is no longer something measured at the end of a campaign cycle. It is continuously shaped while the campaign is live. But as execution accelerates, a new constraint emerges that speed alone cannot solve visibility.

Fragmentation as a Structural Challenge

If AI defines the acceleration of modern marketing, fragmentation defines its limitation. Without full cross-channel visibility, faster optimization can just as easily amplify inefficiency as improving performance.

The contemporary consumer journey is inherently non-linear, spanning retail media networks, social platforms, connected TV, search, and marketplaces. Increasingly, AI-assisted commerce is further fragmenting this journey by adding a new discovery layer, reshaping how consumers evaluate and engage with products through recommendations, assistants, and generative interfaces. As a result, performance is no longer focused within a single channel, but across a sequence of interactions, each contributing differently to the final outcome and requiring marketers to understand and optimize them together.

“Fragmentation forces you to rethink how you define performance in the first place,” Parsons noted. “You have to think beyond what happens in one channel and instead focus on how different touchpoints contribute to an outcome.”

Meanwhile, the platforms that underpin this ecosystem are designed to optimize independently. Closed, highly automated environments such as major social platforms and search marketplaces are extremely effective within their own systems, but those optimizations are inherently local. Each operates as a self-contained system, limiting visibility into how campaigns perform across the broader landscape.

“These systems still work,” Parsons said. “But the challenge is that they work independently. Without a way to connect those signals, you’re making decisions with an incomplete view.”

This creates a structural mismatch. Organizations are often aligned around channels, with separate strategies, budgets, and metrics, while consumers move fluidly across them. The result is a compounding problem: fragmentation reduces visibility across platforms, and limited visibility undermines optimization. Gains within individual channels are often offset by inefficiencies across the system as a whole.

Bridging this gap requires more than faster optimization. It requires coordinated visibility across the entire ecosystem.

The Emergence of Coordinated Performance Systems

To address this, the industry is moving toward systems designed to coordinate performance across channels rather than optimize them in isolation. These platforms unify data, decisioning, and execution into a single layer, enabling marketers to manage performance holistically. The objective is not consolidation for its own sake, but coherence that aligns optimization with how performance is created.

Criteo GO, the company’s AI-powered performance platform, reflects this broader shift. Built on two decades of commerce intelligence, Criteo GO is a self-service tool for SMBs and growth-stage brands, enabling campaign setup in minutes while unifying display, video, native, and social - and soon its first AI platform with ChatGPT - in a single environment. It also leverages generative AI to adapt creative formats and optimize spend in real time.

This level of coordination is becoming essential as discovery itself evolves. Prosper Insights & Analytics data shows that roughly 30% of users already use generative AI to shop for products and services, which is expanding where and how purchase decisions are made. For marketers, this expands the number of decision points they must influence and makes performance harder to manage within any single channel.

“As more of the journey moves into AI-driven environments, signals are getting more distributed and harder to act on in isolation,” Parsons explained. “What’s changed is the ability to move budgets and activate strategy fluidly across channels. You’re no longer locked into decisions made upfront. Systems can adjust based on where performance is actually happening.”

Performance as an Operating Model

As these capabilities mature, performance is evolving from a measurement framework into an operating model. Campaigns are becoming more modular and iterative, teams are reorganizing around cross-channel coordination rather than channel ownership, and performance is increasingly defined by outcomes rather than activity.

At the same time, real-time measurability is tightening marketing’s connection to revenue and embedding it more directly into broader business decision-making.

“Performance is no longer something marketing reports on. Instead, it’s becoming the system through which marketing operates,” Parsons said. “It creates alignment by giving different parts of the organization a shared understanding of what success looks like and how to achieve it.”

That alignment is also what raises the bar. As performance becomes the central currency of advertising, it brings both opportunity and pressure. It enables greater precision, faster iteration, and clearer accountability, but leaves less room for ambiguity. Campaigns must deliver, and do so visibly, because performance is no longer optional. It is the system marketing is built on and the standard it will be held to.

Disclosure: The consumer sentiment study referenced above was conducted by my company, Prosper Insights & Analytics. This is the same dataset used by the National Retail Federation, and available from Amazon Web Services, Bloomberg, and the London Stock Exchange Group for economic benchmarking.