I recently joined Mollie’s EMEA agency meetup at its Amsterdam headquarters, alongside agency leaders from across its partner network. It was a chance to catch up, share experiences and hear more about Mollie’s plans for the payments industry. Alongside the social side, we had talks and round table discussions about where ecommerce and payments are heading.
One conversation that stayed with me was agentic commerce and what it could mean for both industries. Everyone had a view, but there was not always agreement about what the term actually covered. I found that uncertainty interesting because it showed how early we still are in understanding the shift. It also brought the conversation back to something more useful: what does agentic commerce actually mean for retailers and their customers?
For me, the simplest way to understand agentic commerce is to look at what AI can already do for shoppers. An AI assistant can help somebody find products, compare alternatives and narrow down a decision before visiting a retailer. Over time, those services may also become more involved in completing purchases on a customer’s behalf. That changes more than the technology behind ecommerce, because it changes where important buying decisions may happen.
Is This the Death of the Website?
I do not think the website is disappearing, but I do think its role deserves a serious rethink. Traditional ecommerce assumes the website is where customers discover products, compare options and eventually complete a purchase. Most marketing activity has supported that model by bringing people onto the site and guiding them towards a purchase. Agentic commerce challenges that assumption because more discovery and comparison can now happen before a customer arrives.
Imagine somebody asking an AI assistant to recommend a product that meets a very specific need and budget. The assistant could compare several options, explain the differences and reduce a long list to a few suitable choices. By the time that customer reaches a retailer, much of the research may already have happened elsewhere. In some cases, the customer may eventually complete more of the purchase without visiting the retailer at all.
That does not make the website irrelevant, but it does mean the website has to earn its place. I think websites may do less basic transacting and more work around building trust, authority and customer confidence. Customers still need to understand who they are buying from and why that business deserves their attention. Useful content, strong service and a distinctive experience can become more important when simple product discovery happens elsewhere.
For agencies, this also changes how we should approach new website projects and conversations with clients. We may see fewer projects driven mainly by redesign cycles and more work focused on purpose and performance. The question becomes less about what the new website should look like and more about what it should achieve. That is a healthier conversation because it starts with the customer and the commercial role of the site.
Data Is Still the Most Important Thing
One point from the Mollie conversations felt especially clear: product data still matters more than ever. Whatever the channel, customers and AI services both depend on accurate information about what a retailer actually sells. If product details are incomplete, inconsistent or outdated, that weakness can follow the product into other shopping environments. In an agentic world, poor product information can become a visibility problem as well as a website problem.
A human customer can sometimes work around unclear information by checking another page or contacting the retailer directly. An AI assistant may have less tolerance for ambiguity because it needs reliable information before making a recommendation. Old descriptions, inconsistent attributes or outdated prices can therefore have a much bigger commercial impact. They may affect whether a product appears in the list of products considered or whether it is understood correctly.
This is why I see clean product data as a basic business requirement rather than an internal housekeeping exercise. Businesses need clear descriptions, dependable pricing, accurate availability and consistent product details across every relevant channel. The same applies to content that has been left online long after it stopped reflecting the business. If an AI service finds old information, the retailer may not get a chance to explain that the position changed.
Williams Commerce has explored this wider shift in our article on how AI changes discovery before customers reach a website. Clear information helps customers and AI systems understand your offer. That does not require businesses to chase every new AI trend or rebuild everything around a new channel. It does require them to take the quality of their product information much more seriously.
Agentic Commerce Is Not One Single Thing
Part of the confusion comes from using agentic commerce to describe several different experiences. Those experiences need separate thinking. I find it useful to divide the conversation between what happens on your own website and what happens elsewhere. Both involve AI helping customers, but the retailer has a very different level of control in each situation. The commercial questions are also different, so treating everything as one trend can hide important choices.
The first area is agentic commerce on your own website, where an AI assistant helps customers navigate, discover and decide. A shopper could explain what they need instead of working through filters, categories and several product pages. This type of experience is already practical for many retailers, although it still needs careful implementation. Businesses also need to consider transparency, privacy and regulation when AI interacts directly with customers.
Brands operating across European markets should treat AI transparency as a planning requirement, not something to retrofit later. The important point is to consider these responsibilities while the experience is being designed. That helps businesses avoid adding customer safeguards only after a service has already launched. It also keeps trust at the centre of the customer experience rather than treating it as an afterthought.
The second area is product discovery through AI services that sit outside the retailer’s own website. A customer may ask for a recommendation without naming a brand or visiting a store first. Whether a product appears can depend on the quality, consistency and clarity of the information available about it. Retailers have less control over these environments, which makes the information they do control even more important.
The bigger question is whether these services remain a discovery layer or become somewhere the transaction also happens. Those are very different commercial models, even if both can produce a sale for the retailer. Sending a customer back to your website preserves more control over the experience and the relationship. Completing the purchase elsewhere may offer convenience, but it can also mean giving up useful customer information and control.
Who Owns the Customer Relationship?
I think this is one of the most important questions retailers should ask before embracing every new buying channel. Ecommerce value does not only come from the transaction, because the customer relationship can create value long afterwards. Direct relationships help retailers understand behaviour, improve service and keep more customers over time. If another platform owns more of the journey, the retailer needs to understand what that changes.
There may be very good reasons to allow more purchases to happen through AI-led channels. Additional reach and a simpler buying experience can be valuable, especially where customers already use those services. However, businesses should know what customer information they retain and what control they still have after purchase. Ceding the transaction and the customer data around it should be a deliberate commercial decision.
For some retailers, the trade will make sense because the additional sales opportunity outweighs the loss of control. For others, maintaining a direct relationship may remain central to how the business grows and keeps customers. There is no universal answer, and that is precisely why this needs a strategic conversation. The technology may make something possible, but that does not automatically make it the right commercial choice.
Where Are the Platforms Today?
My view is that Shopify is currently ahead in thinking about this shift and building around it. Its investment in AI shopping tools and systems for retailers puts it in a strong position as the market develops. Other platforms are moving too, and I expect the differences between them to change quickly. Retailers should watch that progress without assuming they need to change platform simply because one moves first.
Payment providers are working through their own version of the same challenge as purchases started by AI software develop. A payment started by software still needs the security and trust customers expect from a normal online purchase. Verification will need to keep pace, especially when the customer is not manually completing every step. The providers that solve this well will have an important role in making agentic commerce feel safe and dependable.
Regulation will also continue to catch up with the way AI is being used in commerce. Businesses already need to consider how AI interactions are presented, especially when customers engage with automated services directly. Rules around transparency, responsibility and customer protection will continue to develop as these experiences become more common. Retailers should build with that in mind rather than treating legal requirements as something to add later.
I expect connections between ecommerce platforms, payment providers and AI services to mature over the next two or three years. Nobody can say exactly how each part of the market will develop during that time. That uncertainty is not a reason to ignore the subject or rush into every new service. It is a reason to strengthen the parts of the business that will matter whatever happens next.
What Should Retailers Do Now?
The practical advice is not to wait for every question around agentic commerce to be answered. It is also not to chase every new AI service because the market is talking about it. Start with the foundations that will matter whichever tools and channels become important over the next few years. Product data, content quality and a clear understanding of the customer journey are sensible places to begin.
Get your product information in order and remove content that can no longer be relied upon. Then be honest about what your website is actually there to do for customers and the business. If you are planning something new, build around authority, experience and customer value rather than another redesign cycle. Those priorities make sense today and become even more important if discovery moves beyond the website.
For agencies like Williams Commerce, this should lead to more consultative conversations with clients about where ecommerce is heading. Our Strategy and Consulting work starts with the commercial problem and the customer journey before deciding what should be built. That approach matters because businesses may need different answers as discovery, transactions and customer relationships happen across more places. The website remains important, but it has to work as part of a wider commercial picture.
Nobody can say exactly how quickly customers will adopt every new form of AI-assisted shopping. However, businesses can prepare without making speculative bets or abandoning what already works. The brands that start thinking about these questions now will be better positioned to respond as customer behaviour changes. Better data, clearer content and stronger customer experiences are useful investments regardless of which services become dominant.
The website is not dead, and I do not think that is the most useful way to frame this shift. The bigger change is that a website may no longer carry the entire burden of discovery, comparison and purchase. Its job may increasingly be to create trust, deepen the relationship and give customers reasons to engage directly. The website that only sells things probably has a shorter runway than we used to think.
Talk to Tanya and the Williams Commerce Team.
If agentic commerce raises questions about your website, data or future strategy, we can help. Speak to Tanya and the Williams Commerce team.
📞 Call us on 0116 326 1116
✉️ Email us at [email protected]
Frequently Asked Questions (FAQ)
Agentic commerce describes shopping journeys where AI takes a more active role in helping customers discover, compare or buy products. The level of involvement can vary, from recommending products to supporting more of the transaction. For retailers, the important change is that parts of the buying journey can happen outside the traditional website. That affects product visibility, customer relationships and the role of ecommerce platforms.
No, but the role of the website may change as customers use more ways to discover and compare products. Websites can still build trust, explain products and provide service that other channels cannot easily reproduce. They also remain important for customers who want a direct relationship with the retailer. The opportunity is to make the website more valuable, rather than assuming every journey must start there.
AI services depend on clear, accurate information when they compare products or make recommendations for customers. Inconsistent pricing, outdated descriptions or missing attributes can make products harder to understand and recommend correctly. Better product data also improves the experience across the retailer’s existing website and other sales channels. This makes data quality a practical priority, regardless of how quickly agentic commerce develops.
Start by reviewing product information, older content and the purpose of the current ecommerce website. Make sure customers can access accurate information and remove content that no longer reflects the business. Then consider where AI could genuinely make shopping easier without weakening the customer relationship. Retailers do not need to adopt every new tool to start preparing for this change.
Williams Commerce can help retailers assess their current ecommerce experience and decide where change will create genuine commercial value. That can include website strategy, product data, platform decisions and planning for new customer journeys. The aim is not to introduce technology for its own sake, but to solve the right customer problems. A clear strategy also helps businesses invest in the areas most likely to support long-term growth.
