Agentic Commerce Grows Up Fast: $1.44B Now, $2.17T Forecast by 2030

Agent-led U.S. e-commerce hits $1.44B this year, per eMarketer, as OpenAI ships ChatGPT wallet and Walmart adds spending caps — but only 31% of shoppers trust agents at checkout.

Wire notes

  • End-to-end agentic commerce will generate $1.44B in U.S. e-commerce sales this year, rising to $2.17T by 2030, eMarketer forecasts
  • 61% of senior marketing leaders have adapted content and SEO/AEO for machine interpretability, but only 30% have built agent transaction workflows, Gartner found
  • Only 31% of consumers would let an AI agent complete a purchase without review, while 62% of early adopters would, per Okta
  • AI-referred traffic converted 90% better than other channels in September, Adobe Digital Insights reported
  • Amazon Q3 2026: $68.5B online stores, $46.5B third-party seller services, $20B ad services, per Enders Analysis
In Graphic Detail: Welcome to the agentic commerce quarter
PhotoIn Graphic Detail: Welcome to the agentic commerce quarter — Firmbee / pixabay

U.S. retail e-commerce sales completed through end-to-end agentic commerce will reach $1.44 billion this year, according to eMarketer — and the forecast climbs to $2.17 trillion by 2030. Holiday 2026 is shaping up as the first season where the shopping agent actually pays, which solidifies Q4 as the agentic commerce quarter.

The past week made the shift concrete. OpenAI launched ChatGPT wallet to push agent purchases through retailer checkouts, alongside a commerce partner playbook to help businesses prepare for more agent-led buying. Google began rolling its UCP-powered checkout in Canada through AI Mode and Gemini. Walmart now lets customers cap an agent's spending at $50, $100 or $200. Amazon, by contrast, blocked Meta's Muse AI agent from its retail site. And Bain reckons roughly one in four online shoppers plan to start their holiday shopping journeys on an AI platform.

Put together, agentic commerce is having a growth spurt even if it is still a while from growing up.

Where is the money going?

The commercial numbers point one direction. Beyond this year's $1.44 billion, eMarketer expects agentic commerce sales to increase 10.6% to $1.59 billion in 2027, then 11% to $1.76 billion in 2028, before the steep climb toward $2.17 trillion by 2030. Platforms and retailers are already positioning themselves for a new way of shopping rather than treating agents as a hypothetical.

What are marketers actually doing about it?

The honest answer: discoverability first, checkout later. Gartner's 2026 Marketing Strategy and Engagement Survey found 61% of senior marketing leaders have adapted content and SEO/AEO to improve machine interpretability. The picture thins out at the transaction layer:

  • 42% monitor and correct misinformation about their brands in AI-generated outputs
  • 39% have tailored product information, pricing or offers for machine and agent decision-making
  • Only 30% have designed interfaces or workflows for agent transactions
  • 32% manage machines and agents as a distinct audience alongside human customers

The gap between interpretability work and transaction readiness is the practical story here. Most marketers are optimizing to be found by agents, not to be bought from by them.

Are shoppers ready to let agents spend their money?

Mostly, no. Okta's Customer Identity Trends Report 2026 found only 31% of consumers said they would be very or rather likely to let an AI agent complete a purchase without reviewing it first. Two thirds (67%) said they were unlikely to do so, and 41% said they would not allow it at all.

Early adopters break the pattern. Among them, nearly two thirds (62%) said they would let an AI agent complete a purchase without checking, and 42% said they would be very likely to. The gap suggests the hard part of agentic commerce may not be getting the right tech, but building the right amount of trust.

Where does agent traffic already pay off?

Shoppers may not hand over the checkout, but AI tools already influence where they buy — and the visitors those tools send spend more. Adobe Digital Insights found that in September, shoppers arriving at U.S. retail e-commerce sites from AI tools converted 90% better than those from other channels. That is the highest level Adobe has recorded, and a sharp turnaround from August 2025, when AI traffic converted 9% worse than other channels.

What does this change for Amazon?

Agentic commerce could hit Amazon's economics beyond product sales. Enders Analysis puts Amazon's Q3 2026 net sales at $68.5 billion from online stores, $46.5 billion from third-party seller services and $20 billion from ad services. If shoppers increasingly rely on agents to identify and recommend products rather than browse sponsored listings, Amazon's advertising business could face pressure from fewer encounters with paid placements — while its fulfilment infrastructure could benefit by supporting transactions initiated elsewhere.

For practitioners, the sequence is now clear: machine interpretability is table stakes, transaction readiness is the differentiator, and the September conversion data shows agent-referred traffic already outperforms. The brands that build trust and agent-ready checkout paths before the 2030 curve steepens will set the terms of the agentic commerce quarter.

via Digiday (Source)

Filed under

Share this article:

More from Priya Raman

Priya Raman

Show full bio

Correspondent covering people, launches and day-to-day practice in marketing, advertising, marketing technology, media, SEO and search at Marketing Herald.

359 articles

More on the wire

« Previous articleNext article »