Artificial intelligence agents are expected to facilitate more than $3.35 trillion in global consumer spending by 2030, transforming how products are discovered, compared and purchased, according to a new global study that argues marketers will increasingly need to persuade machines as well as people.
The report, From Abundance to Agents: How the delegation of choice is transforming marketing, by PHD in partnership with WARC, estimates that agent-facilitated consumer spending will rise from more than $944 billion in 2026 to $3.35 trillion by 2030, accounting for 3.8% of global consumer spending. The top 10 markets will generate nearly 68% of this spending, with the United States alone contributing $1.1 trillion.
The study says that while consumers will continue to make most purchasing decisions themselves over the next few years, AI agents will increasingly determine “what gets seen, shortlisted and bought” by taking over the repetitive and information-heavy parts of the buying journey.
“Marketing is operating in an age of abundance. More content, choice and complexity than consumers have the attention or means to manage,” the report says, adding that “agentic transactions may seem like buzzwords today, but for tomorrow’s global consumer, they will be fundamental tools for managing abundance.”
It defines agentic AI as systems capable of understanding goals, planning actions and acting autonomously with limited human supervision, while an agentic transaction is one in which an AI agent is authorised to make and execute purchasing decisions on a person’s behalf within preset rules.
Three sectors to undergo the biggest transformation
The report identifies telecoms and utilities, financial services, and travel and transport as the sectors that will experience the earliest and deepest shift towards agent-to-agent transactions, where AI systems interact directly with one another during the purchasing process.
Among them, telecoms and utilities is projected to witness the fastest growth. Agent-facilitated spending in the sector is expected to jump from $57.6 billion in 2026 to $410.3 billion by 2030, an increase of more than 600%.
“With information-dense decision processes, frequent billing and comparison-led contract switching, services in this category are ripe for agentic delegation,” the report says. It recommends that companies make pricing and product information easily accessible to AI systems because “AI can compare dozens of pricing plans simultaneously,” simplifying a traditionally cumbersome process.
Travel and transport will also undergo a profound shift. Agent-facilitated spending is forecast to rise from $78.1 billion to $275.6 billion by 2030.
The report notes that “search-heavy workflows for booking, itinerary planning and comparison make this category already well suited to agentic support,” and predicts that “the entire plan might be made without a consumer’s involvement” as AI agents increasingly handle travel discovery, planning and purchasing.
Financial services, meanwhile, are expected to grow from $71 billion to $237.9 billion in AI-facilitated spending by 2030.
Although consumers are unlikely to surrender complete control over major financial decisions, AI agents will increasingly determine “which savings account looks best, which mortgage options are worth comparing or which insurance policy is a better value,” according to the report. It adds that “trust signals will become ranking signals,” meaning reviews, regulatory ratings, customer service records and transparency will influence whether AI systems recommend one provider over another.
Everyday purchases increasingly delegated
Beyond these sectors, the report predicts growing AI involvement in food, soft drinks, media and publishing, retail and alcoholic beverages, where purchases are frequent and often habitual.
Food purchases represent “one of the clearest early use cases for agentic AI thanks to their very high frequency and low decision complexity,” the report says, forecasting AI-influenced spending to rise from $78.1 billion to $292.8 billion by 2030. It suggests brands compete to become part of AI-generated meal plans rather than merely grocery lists.
Media and publishing is projected to become the second-largest AI-facilitated category, with spending increasing from $73.3 billion to $367.8 billion. The report urges publishers to become “trusted sources for content” because AI systems are increasingly relying on established media organisations for information.
Retail spending facilitated by AI is expected to exceed $199 billion by 2030. The report warns retailers that the challenge will be “how to remain part of the consumer shopping journey as brands and avoid being relegated to becoming fulfilment providers.”
Human decisions will still matter
Despite the rapid growth of AI-assisted commerce, the report repeatedly stresses that humans will continue to make final purchasing decisions, particularly for expensive, emotionally driven or highly regulated products.
Sarah Sorrenson, Global Media Director at Diageo, says marketers need “the appropriate data to govern” AI recommendation systems and “the appropriate operating models and workflows to be able to deliver against this new future.”
At the same time, Tim Walther of Volkswagen Group cautions: “We are in a highly disruptive phase… while remembering that, ultimately, people—not AI agents—are buying our products.”
The report argues that the marketing profession will increasingly require brands to communicate with both people and machines.
Lucinda Barlow, International CMO at Uber, says “human and machine audiences are fundamentally different. Humans are driven by emotion, trust, habit, and social norms… Machines are driven by logic, structured data, and algorithms.” Marketing to AI therefore requires “comprehensive information” that is clearly documented and accessible.
Dr Oksana Koval of Audi adds that companies need to replace vague descriptions with precise, structured information.
“We can no longer describe a product as ‘great’ or ‘nice’—that is not meaningful for machines. Instead, we need to describe specific attributes, such as trunk size or maximum speed.”
Don’t abandon traditional marketing
Even as AI agents become more influential, the study warns brands against neglecting conventional marketing principles.
“Traditional marketing skills will remain really important,” says Sarah Sorrenson, arguing that brands must remain “distinctive, differentiated and chosen by AI.”
Meredith Kelly, CMO of Škoda Auto, says marketers should “experiment with technology” but remember that “humans will still make the choice. I still think the brands will win out.”
Lucinda Barlow offers a similar warning: “Do not sacrifice human-focused marketing. Do not forget about brand or human audiences… Brands should not over-optimise for machines at the expense of humans.”
