When AI Agents Become Shoppers: How Small Businesses Can Stay on the Shortlist
This week, US shoppers gained the ability to buy Newegg products without leaving Google's Gemini app or AI Mode in Search. They describe what they need, compare options in the conversation, and check out with Google Pay on the spot. Newegg remains the merchant of record (Business Wire). Google described it as "a new milestone as we scale agentic commerce."
The same day, Simon-Kucher published its annual holiday shopping study. Of 1,501 US consumers surveyed, 55% plan to use AI for holiday shopping. Among those AI users, 55% will use it to find prices and deals, and 16% would let AI make the purchase for them (PPC Land). Only 22% of shoppers said advertised discounts fully reflect true savings, down from 46% a year earlier.
Put together, the direction is clear. A growing share of buying journeys now start, and sometimes finish, inside an AI assistant. The assistant does the research, builds the shortlist, and increasingly completes the order.
What changes for small businesses
The shortlist is built by a machine. When a buyer asks an assistant for the best option, it does not show ten blue links. It names two or three. If you are not one of them, you were never considered.
Assistants read data, not design. A beautiful website does not help if the assistant cannot work out your prices, stock, delivery times, or what makes you different. Product data, structured information, and clear pages matter more than visuals.
Trust signals are checked, not felt. Shoppers trust discounts less every year. Assistants compare claims against reviews, return policies, and other sources. Vague claims get filtered out.
It is not only for retail. The same pattern is spreading to services. A buyer asking an assistant to "find three accounting firms in Pune that work with startups" is running an agentic shortlist, even if a human still makes the call.
Five things to fix now
1. Make your basic facts machine-readable. Name, location, services or products, prices or price ranges, hours, delivery or service areas. Put them in plain text on your site and in structured data. Our free schema markup generator creates the code without any programming.
2. Keep product and price information accurate everywhere. If you sell products, keep your feed in Google Merchant Center complete and in sync with your stock. An assistant that finds a wrong price or an out-of-stock item will move on, and may not come back.
3. Say clearly why someone should choose you. Assistants summarise. Give them a sentence worth repeating: who you serve, what you are best at, and the proof. Our positioning statement builder helps you write it.
4. Collect reviews and publish real proof. Specific reviews, case studies, and results give an assistant something to cite. "Great service" is weak. "Delivered our office fit-out two weeks early" is strong. If clients are under NDA, our case study builder has an anonymous mode.
5. Check what the assistants say about you today. Ask ChatGPT, Gemini, and Perplexity the questions your buyers ask. Note whether you appear, how you are described, and which competitors appear instead. Our free visibility audit for Google and AI walks you through it and gives you a fix list.
A word of caution
Agentic commerce is early. Checkout inside assistants is live only with some merchants and in some countries, and the 16% of AI users willing to let AI buy for them is still a minority. You do not need to rebuild your business for agents this quarter.
But the foundations that help agents (clear facts, accurate data, real proof) also help human buyers and search engines today. There is no wasted effort in fixing them now.
If you want help, our AI search visibility service makes your business easier for assistants to find and recommend. For a broader view of what to fix first, start with a fixed-price growth audit.