78% of e-commerce brands are invisible to AI recommenders
Four stories this week that all point the same direction: the machines are doing the shopping, setting the budgets, and picking the winners. Most brands aren’t in the running.
Amex requires proof of intent before agents can spend
American Express built an Intent Intelligence Framework that logs a customer’s instructions before any AI agent touches their card. Through a Create Intent API, the shopper specifies the merchant, a maximum budget, and product preferences, then authenticates and approves the transaction scope.
From that, Amex issues an Intent ID and a Proof of Intent Token. The agent gets a digital payment credential and never sees the real card number. Amex reports it has not yet seen a rise in agentic disputes.
In an Amex survey, 66% of US consumers said they are likely to use AI agents to research prices within 12 months, and 61% said the same for product recommendations. The operator implication is direct: your offer has to clear a shopper-set budget and a product spec before any marketing persuasion even enters the picture.
78% of e-commerce brands never show up in AI recommendations
CommerceV3 ran more than 1 million US shopping prompts across ChatGPT, Gemini, and Perplexity between March and July 2026. The same small group of brands filled 75% of recommendation slots in every category tested.
Brand absence broke down by platform: 81% on Gemini, 76% on ChatGPT, and 74% on Perplexity. Problem-solving prompts were the worst case, with an 84% absence rate. As for what drives citations, editorial and review sites accounted for 34%, Reddit and forums 22%. Brands’ own websites supplied just 14%.
If you’re betting on your product pages to get you into AI answers, that 14% figure is the reality check. The work is earning mentions on third-party review sites and forums, then checking prompt by prompt whether you appear at all.
Google tests campaign setup from a plain-language goal
Google Ads is experimenting with a new campaign creation flow. Instead of selecting from a menu of campaign objectives, advertisers type a goal into a text box. Google’s AI then suggests goals and creative assets based on that input.
A manual creation option remains. Google Ads Liaison Ginny Marvin confirmed the test is an experiment and asked for feedback. Google has not specified which campaign types are included, when a broader rollout might happen, or how much control advertisers retain over the AI’s suggestions.
If objective-led setup disappears, the brief your team writes becomes the campaign. Getting precise about goals, audiences, and limits before that change lands is worth doing now.
️ Shopify’s Shop MCP opens its catalogue to every AI agent
Shopify launched Shop MCP, a single connection that gives any AI agent access to more than one billion products, cart building, and Shop Pay checkout. Product search requires no API key.
Signed-in shoppers can use saved addresses and Shop Pay. The agent completes the purchase only after the shopper confirms the total and payment method. Orders appear in the merchant’s Shopify admin. There are no webhooks at launch, so developers need to poll for order status.
For merchants, this makes product data the new storefront. Clean titles, accurate specs, and up-to-date availability now carry the same weight as creative on your product pages.
Quick hits
- ChatGPT now cites primary sources more often. Reddit’s share of citations fell from 19.1% to 2.3%. PubMed jumped to fourth place.
- Semrush finds fewer than half of brands that rank highly for AI mentions are also highly cited. Visibility scores may be masking a real gap.
- Kenvue’s Aveeno cut a 30-day brief to three days using synthetic personas, part of a broader agency move toward synthetic audience testing.
- Publicis CEO Arthur Sadoun says bypassing CMOs and pitching AI directly to CEOs helped win PepsiCo without a traditional pitch process.
- Bing Webmaster Tools added a beta Topics Portfolio report that groups AI citation queries by topic. No click data included at launch.
- A Raptive survey found consumer trust in AI answers rises from 42% to 77% when a trusted publisher is cited as the source.
- Google is testing more prominent inline links in AI Mode and AI Overviews, including official-site placement tests.
- A Product.ai survey found 89.9% of AI shoppers verify AI advice elsewhere before buying, and only 21.6% have let an AI assistant complete a purchase for them.
I’m founder of BizStack at Cagri Sarigoz LLC and a passionate advocate for entrepreneurs.
With over 14 years in tech, marketing, and AI, including my role as Head of SEO at CitizenShipper and co-founder of TaleBot at Intale AI, I’m dedicated to sharing genuine, useful product insights and tips.
At BizStack, I aim to cut through the digital noise to provide clear, actionable advice.
And more than all else, I’m a father to a (always) little girl and a husband.
Contact me at [email protected] for assistance.
