AI agents don't care about your brand — and 47% of execs know it

Mastercard expects 1 in 10 shoppers to use AI agents by 2030, yet only 20% of execs think they can influence AI recommendations. Brands are racing to become machine-readable.
By Grace Kim
4 min readAt a glance
- Mastercard forecasts 1 in 10 people will routinely use AI agents to shop and pay by 2030
- 52% of U.K. shoppers used AI for a shopping task in the past three months; 80% in China
- Only 20% of consumer products executives believe they can influence AI recommendations today, per EY
- Only 38% of URLs cited by AI rank in Google's top 10; 31% rank outside the top 100
- 41% of consumers would feel comfortable automating household product purchases, per Mintel
Meta's shopping-capable AI agent Muse is expected to launch in the U.K. "within weeks," and Mastercard forecasts one in 10 people will routinely use AI agents to shop and pay by 2030. Those two data points frame why boardrooms are suddenly rethinking who — or what — their customer actually is.
In the U.K., 52% of shoppers have used AI for at least one shopping task in the past three months, according to McKinsey research. In China, where adoption is more advanced, that figure hits 80%. Today most consumers use AI for research and make the final call themselves. That balance is shifting toward delegation.
What does this change for marketers?
The uncomfortable part: the usual advantages don't transfer. Fredrik Kridahl, founder of AI search consultancy Hyper Agency, points to research showing only 38% of URLs cited by AI rank in the top 10 on regular search engines — and 31% rank outside the top 100. "AI doesn't care about how big your company is," he says. "It cares about the facts and giving the right advice."
Query behavior is different too. "A normal Google search is 3.2 words on average," Kridahl says. "An average query on an average prompt is 91 words." That means brands must answer long, conversational questions, not chase keywords. Add the "black box" problem — nobody fully knows how agent algorithms pick recommendations — and the confidence gap becomes predictable. EY found 47% of consumer products executives believe influencing AI recommendations will be essential to stay competitive over the next five years. Only 20% believe they can deliver that today.
The stakes aren't marginal. EY notes just 10% of consumers need to shift behavior to materially affect revenue and profits. Low-emotion, high-frequency categories go first: Mintel data shows 41% of consumers would feel comfortable automating household product purchases. Mintel retail principal strategist Nick Carroll says Unilever's beauty, wellbeing and home care products are prime candidates for delegation.
The unglamorous work of marketing to bots
Retailers are already rebuilding infrastructure. At the start of the year, JD Sports launched tech letting U.S. shoppers buy through Microsoft Copilot, Google's Gemini and ChatGPT in a single click. John Lewis announced "a significant investment" in AI-powered shopping in March to surface its products on those same platforms.
Very Group's digital director Sarah Hanna says AI search currently drives less than 2% of its site traffic — but she's not waiting. "Getting in front of AI agents is set to become just as important as traditional SEO," she says. That means product information that is "accurate, detailed and structured so AI can understand, recommend and compare" effectively.
JD Sports' chief technology and transformation officer Jetan Chowk frames it as making the offer "visible, accessible and accurately represented" — up-to-date product data, pricing, availability, all machine-readable. EY Americas commercial excellence leader Steven Bailey calls this the "unglamorous" side: encode price, promotion, product, persona and differentiation so machines can parse them.
Reviews matter too, because a well-rated product gives an agent confidence to recommend it. Unilever homecare global CMO Tati Lindenberg describes the tactic plainly: "We optimize metadata, product pages, and community platforms such as Reddit, so our brands surface correctly when people ask natural questions." Her example: someone asks "I need a quick clean up, what detergent should I use?" — and Persil Wonder Wash should appear in the answer, "for the right reasons." Her larger framing: "The AI answer is becoming a new kind of shelf."
Does emotional marketing still matter?
Here's the caveat worth holding onto. An agent won't be moved by Unilever's period-stigma campaign or JD Sports' Christmas ad. But Bailey argues trust and humanness now count for more, not less, because they differentiate — and shoppers are far more likely to let AI automate purchases from brands they already trust.
Accenture's Kathryn Gramling says product descriptions must "make the emotional and experiential dimensions more understandable to AI," especially in fashion, beauty, luxury and travel. Expedia's CMO Jochen Koedijk puts it concretely: "A hotel that feels right to people but is described poorly in its data can get skipped by an agent." His team's B2A work turns atmosphere into evidence — guest language, amenity signals for a quiet weekend or family trip.
The risk cuts both ways. "The danger is optimizing so hard for algorithms that we lose the creativity and distinctiveness that make a brand worth choosing in the first place," Lindenberg says. Her bet: as AI makes content easier to produce, taste, originality and human imagination become greater competitive advantages. Muse's U.K. launch will give that theory its first broad retail test.
Source: fortune.com
Grace Kim
Show full bio
Senior reporter covering companies, deals and strategy in Social platform news and policy at Reach Brief.
Related articles
AI Agents Could Steer Up to $385 Billion of US E-Commerce by 2030
ChatGPT Ads Convert at 0.37% — Six Times Worse Than Organic Answers
Similarweb: Over 90% of AI-Driven Sales Get Miscredited
Cloudflare Says AI Bots Scrape Thousands of Pages Per Reader Sent
AI Search Now Decides Which Creators Get Discovered