AI is changing how people buy. Marketing needs to catch up

Picture a runner preparing for her first 10km. She asks an AI assistant for a training plan and shoes suited to her running style and an old knee problem. Within seconds, it weighs hundreds of reviews and suggests options tailored to her needs.

As she moves closer to buying, the assistant pulls useful advice from videos and online discussions. It screens out a generic email from a sports brand. When she is ready to purchase, it flags one retailer’s poor returns policy and chooses another offering next-day delivery. The brand’s carefully timed campaign barely enters the journey.

This experience is already possible. AI is starting to change not just how consumers find products, but how they decide what to buy.

The traditional campaign model grew up around a calendar. Weeks of work built towards launch day, after which the campaign ran its course. Consumers increasingly follow a different rhythm.

Over the past decade, as online channels have multiplied, shoppers have steadily increased the number of ways they research or buy products; in fact, they now use twice as many channels as they did 10 years ago. AI is rapidly changing the game. More than half of consumers already use AI-based search during the purchase journey, and a single conversation with an AI assistant can compress hours of searching and comparing into seconds. According to McKinsey survey data, 73% of AI-search users leverage it early on to learn about product categories, while over 60% rely on it later in the journey to compare specific products or digest reviews

AI now sits between a brand and the consumer it wants to reach. It can interpret a consumer’s needs, evaluate competing offers, and influence which brands make it into consideration at all. A campaign message can arrive at exactly the time the brand planned and still be filtered out before the customer sees it. The campaign has run on schedule, while the purchase has moved elsewhere.

Many marketing departments still work in bursts. A campaign is developed and approved, goes live, then results are reviewed to inform the next one. The pauses are built into the model.

AI can make parts of that process much faster. Campaign cycles can shrink from six to ten weeks to same-day execution. Content that once took days can now be created within hours or minutes.

That gives marketers the chance to use new information sooner. A marketer can use a synthetic customer persona to test an idea without waiting for a traditional round of focus groups. Performance data can then shape what happens next, so each interaction improves the following one.

Marketers have embraced AI tools quickly. Nearly 60% now use AI several times a week, yet fewer than 10% say AI is creating value across their end-to-end marketing workflows. Using AI to produce a first draft or analyze a set of results faster does nothing by itself to make the rest of the work move any more quickly. Any time gained is soon lost if the output enters the same approval queues, handoffs and decision-making processes.

Most companies remain at this stage. Only 28% of marketers say their companies are pursuing a fundamental rewiring of marketing teams and workflows. The rest are largely adding AI to existing processes, and little of that activity is yet showing up on the bottom line.

Suppose AI reduces a day’s work to a morning. A marketer could spend the freed afternoon understanding why customers are behaving a certain way, giving a promising idea more attention, or responding to new information while it is still useful. Multiply that across a marketing team and the opportunity goes well beyond a few hours saved.

Turning that extra capacity into growth depends on how the work itself is redesigned. Marketing leaders need to look at how work moves from a customer signal to a decision, then decide where AI can act and where human judgement is essential. As AI agents take on more of the execution work and respond to data in near real time, marketers will need the skills to work differently. That means directing AI, interpreting what it produces and deciding what deserves attention. They will also have to judge whether the resulting work is good enough to put in front of a customer.

We can see what that looks like at one global food company, which rebuilt the way it produced marketing content using AI. Early stages of this initiative cut campaign costs by around 80% and made production roughly 90% faster. AI helped create different versions of content for local markets, allowing assets to be adapted faster and more accurately. The process had been redesigned from beginning to end, carrying those gains across the content supply chain.

That way of working also changes who does what and who answers for the result. Marketing will need more people who can guide work shared between humans and AI, bringing the creativity and judgement needed to shape the result and using AI to amplify both. Reskilling has a key part to play.

Campaigns will remain an important part of marketing, but they can no longer be the clock that marketing runs on. To turn AI into a source of growth, marketing organizations must shift to an always-on operating model – bringing insight, creativity and execution much closer together.