How AI is Reshaping Marketing Effectiveness: 5 takeaways from Tom Davenport

AI is transforming marketing fast. Generative AI is scaling content production. Analytical AI is improving decision-making. And agentic AI may soon change how consumers search, compare and buy.

But for marketing leaders, the real question is not whether AI will change marketing. It already has. The question is: what does AI actually change for marketing effectiveness?

To explore this, we sat down with AI expert and MIT professor Tom Davenport*. His perspective confirms what we see every day with our clients: AI may change the tools, channels and pace of marketing, but the need for robust measurement has never been more important.

Here are five takeaways from the conversation.

Influence changes. Effectiveness doesn’t.

The core questions behind marketing effectiveness remain largely unchanged.

Marketers still care about the same signals they have tracked for years. As Tom Davenport puts it: “Did people look at what I’m offering? How long did they spend looking at it? Do they come back? Do they convert?”

Those outcome metrics have been stable for decades. What AI changes is how marketers influence those outcomes.

As AI assistants become more embedded in consumer journeys, brands may need to optimize not only for human attention, but also for AI-assisted decisions. In the age of agentic commerce, marketers will need to understand the logic behind those systems. Brands will increasingly need visibility into “what algorithms are driving the choices that an agentic system makes”, and find ways to position themselves accordingly.

Marketing effectiveness is not disappearing. But the systems that influence it are becoming increasingly algorithmic — which makes robust measurement more critical, not less.

AI scales content. Measurement guides strategy.

AI is already changing how marketing assets are produced.

Across advertising, social, CRM and product content, generative AI is helping teams create faster and at greater scale. As Tom Davenport says, “the content will be generated in most cases by AI across all those different channels.”

But more content does not automatically mean better marketing.

In Davenport’s view, the future marketing stack will rely on two complementary capabilities.

Generative AI can scale execution. It cannot decide where investment should go, which channels are truly incremental, or how budgets should shift. That is where analytical AI, econometrics and advanced measurement frameworks like Marketing Mix Modeling remains essential.

More content. More complexity. Less clarity.

AI makes it easier to produce more assets and more variations but it also creates a new challenge: measurement complexity.

Marketers will still rely on familiar signals such as clicks, engagement and conversion. Yet, as Tom Davenport explains, “it’s just going to be harder to get them given the vast amount of content that will be available.”

Three structural shifts are driving this complexity:

More content

AI has made it possible to produce creative variations at unprecedented scale.

More channels

Brands now need to manage content across retail media, marketplaces, CTV, social, and beyond.

Less clarity

With so many assets and channels interacting at once, pinpointing what drives results is increasingly difficult.

‍

Marketing teams can no longer rely on fragmented channel metrics. They need a system-level view of what is driving performance. This is where advanced measurement frameworks such as Marketing Mix Modeling help provide that perspective, showing how channels, assets, and investments work together.

Not all AI drives better decisions.

One of the biggest risks is treating “AI” as a single capability.

Generative AI and analytical AI play very different roles. As Tom Davenport notes, “Generative AI is predictive… but it’s predicting the next word.”

That capability is powerful for creating text, images and campaign assets. But it is not the same as predicting customer behavior or business and marketing outcomes. Those decisions require what Davenport describes as “econometrics, analytical AI, classical machine learning.”

Confusing a content engine with a decision engine is not a nuance: it's a strategic error that leads to misallocated budgets and missed opportunities. The leading organizations connect both: analytics to guide strategic decisions, and generative systems to execute at scale.

AI agents may gatekeep demand.

The most disruptive shift Davenport anticipates is the rise of AI acting as a buyer agent for consumers.

We observe that customers are increasingly using AI “almost like a buyer agent.”

Instead of browsing websites or comparing options manually, people may ask AI assistants to recommend the best product, provider or service for their needs. Over time, these systems may become embedded into everyday decision-making.

“Have your bot talk to my bot.”

In that world, marketing effectiveness will depend not only on how consumers evaluate brands, but on how AI systems interpret, rank and recommend them.

For marketers, the challenge becomes clear: understand not only what consumers think, but how algorithms decide. Davenport suggests that algorithm designers may become the real gatekeepers. The people building recommendation systems, he notes, may ultimately play a larger role in determining which brands consumers see.

The bigger shift: Marketing is entering an algorithmic era

These five shifts are not incremental. They are structural.

Marketing is being rewired by AI. It now shapes how content is created, decisions are made, and consumers buy. Organizations that do not adapt their measurement and investment frameworks risk being outpaced by those that do.

The discipline is not disappearing. But it is becoming increasingly algorithmic.

For marketing leaders, three capabilities will matter most:

• Analytical intelligence to understand what drives growth.
• Generative scale to create and adapt assets efficiently.
• Algorithmic literacy to understand how AI systems evaluate brands.

For brand activation, this matters now. AI will not only change how content and campaigns are produced. It will change how they are measured, optimized and experienced. Because the next phase of marketing will not only be a competition for human attention. It may also be a competition for algorithmic preference.

‍

Ready to turn AI and analytics into decisions that drive real business outcomes?

Ekimetrics helps brands move from data to decisions, faster and more confidently. Let's make your next move your brightest decision.

‍

‍

_{*Thomas H. Davenport is the President's Distinguished Professor of Information Technology and faculty director of the Metropoulos Institute for Technology and Entrepreneurship at Babson College, a visiting scholar at the MIT Initiative on the Digital Economy, and a senior adviser to Deloitte's Chief Data and Analytics Officer Program.}

‍