.webp)
.webp)
Agentic commerce: When the customer is an AI — is your business visible at the moment of purchase?
McKinsey estimates that AI agents will handle between $3 and $5 trillion in global trade by 2030. Already today, 73 percent of consumers use AI as part of their buying journey, from product discovery to price comparison and ordering. In the United States, AI driven commerce has already taken off at a pace most Norwegian companies have not been paying attention to. The question is not whether it will get here. That's how well prepared you are when a whole new type of customer shows up at the checkout: an artificial intelligence acting on behalf of a human.
Agentic commerce is about AI agents like ChatGPT, Microsoft Copilot, Google Gemini and Perplexity not only answering questions, but actually executing the purchase. A consumer can ask the assistant to “find the best hiking jacket under NOK 3000 that can be delivered by Friday,” and the agent searches through catalogues, compares options, assesses inventory and completes the transaction. The human being may approve the final order, but the actual research and decision takes place without the user having visited a single website.
This is a fundamental change. SEO was all about ranking on Google. AEO, or Answer Engine Optimization, was about being quoted in the answers provided by AI. Agentic commerce goes one step further. Here, the goal is not to be read by the customer, but to be picked by an algorithm that acts in the customer's place. The machine assesses your product data based on price, availability, delivery time, reviews and how well structured your information actually is. Homepages with beautiful designs and little metadata, or product pages with lots of images or PDFs, become invisible.
Shopify launched Agentic Storefronts in the Winter '26 issue, and as of March 2026, it has been rolled out to millions of stores. Together with Google, they have developed the Universal Commerce Protocol, or UCP, an open standard that allows AI agents to trade directly. Microsoft has launched Copilot Checkout. ChatGPT has integrated trading directly into the chat.
Important to be aware: This currently only applies to Shopify stores that sell to US customers. ChatGPT and Microsoft Copilot are live for stores that have the United States as a market, regardless of where the store itself is registered. Google AI Mode and Gemini are available to select US brands in a wider rollout phase. For the Norwegian and European markets, the timeline is still open. Experience-wise, the European rollout often follows 6 to 18 months behind the American one, and privacy, VAT handling and local payment solutions are complicating factors that require time.
This does not mean that Norwegian companies can wait. It means the opposite.
Preparation is not about pressing a button on the day the feature launches here. It deals with the groundwork in product data, schema markup, service descriptions and brand authority. These are things that take months, not days, to get in place for most businesses.
Norwegian companies that sell into the United States, either through their own online commerce or through international scaling, can also leverage the U.S. rollout directly now. For many B2C players in clothing, cosmetics, home and interiors, this is a real opportunity to test agentic commerce in a market where the infrastructure exists, before competitors in Norway have understood what is happening.
It's easy to think that agentic commerce is just about e-commerce, but Forrester estimates that AI agents will have at least as much impact on B2B purchases. Buyers are already using AI to compare suppliers and assess contracts. When someone asks the assistant to “find three relevant CRM vendors for a team of 50 employees with integration against HubSpot,” it's your product data, case studies, and pricing information that are considered. Not your sales team. This means that service descriptions must be machine-readable, case studies should be structured with clear results, and price indications should be available wherever possible. AI agents often filter out options that require a demo just to get an estimate.
In order for your business to be visible when AI agents trade, whether it's through U.S. channels today or Norwegian channels tomorrow, there are a few basic things that need to be in place:
When the customer never visits the website, there are no clicks to track. As AI trading also hits Norway you need to start measuring new things. How often is your brand mentioned in the responses of AI agents? What categories are you proposed in? GA4 has started reporting AI referrals, but a lot of the job is about setting up your own tracking regimes between your product catalog, AI platforms and your customer data platform. The difference between those who succeed and those who fall behind will lie in how early they start building this.
We at WeAssist help Norwegian and Nordic companies prepare for what is to come. It involves clearing product data, optimizing structured data for AI agents, making content machine-readable and building measurement systems that can capture the new traffic as it arrives. Wondering how well prepared your company is for agentic commerce? The first step is a mapping of product data and content. Get in touch, then we'll figure out where you stand and what should be prioritized first.
.webp)
