Imagine you... could talk to a brand instead of browsing its website. This is conversational commerce.

Imagine you want to buy a new coffee machine and you open a manufacturer's website, where you would normally first browse products, categories and filters, click through several models, compare technical data and eventually try to find out why machine A should actually suit you better than machine B.

This time, instead, simply write:

"I mainly drink espresso, in the mornings I need something quick, my kitchen is small and I want to spend a maximum of 700 euros. What would you recommend?"

The brand responds with two models, explains why these two are suitable, and you ask further which one is quieter, which is easier to clean, and whether you can also make a good cappuccino for friends with it on the weekend.

A few minutes later you made your decision.

You didn't search through a product category, work your way through ten tabs, or compare technical data sheets; you simply described what you needed, and the brand advised you.

This is precisely where a development lies that initially looks like a new form of website UX, but is actually much more fundamental because it changes how people use digital services in general.

Websites have traditionally been built around information, but people think in situations.

The classic website follows a relatively clear structure.

Companies own information, products, and content and organize them in such a way that people can find their way around them as easily as possible.

Home.

Products.

Categories.

Subcategories.

Filter.

Product detail pages.

FAQ.

Contact.

We have learned to use these structures because for many years they have been the most sensible way to make large amounts of information accessible.

However, people rarely think that way.

When buying a dining table, we don't usually think first about width, depth, material, color and price range, but rather: We need a table for six people, the room is relatively small, we like wood, but definitely don't want anything rustic and it shouldn't cost more than 2,000 euros if possible.

The same applies to travel, fashion, beauty, insurance, software and many other purchasing decisions.

We think in terms of needs, situations, problems, preferences and often also feelings, while digital systems have so far forced us to translate these thoughts ourselves into search terms, categories and filters.

Conversational Commerce reverses this principle.

Humans need to understand the structure of the system less and less because the system is beginning to understand what humans actually want.

And that's precisely why this issue is much bigger than just the next chatbot.

This development is already taking place.

In March 2026, OpenAI significantly expanded product search capabilities in ChatGPT. Users can now describe what they are looking for, modify requirements during the conversation, upload images for inspiration, visually compare products, and refine their selection step by step. Simultaneously, the Agentic Commerce Protocol is being developed to provide merchants with up-to-date and structured product information for such systems.

Google is also consistently expanding this development. With the Universal Commerce Protocol, the Universal Cart, and other agent-based shopping features, a system is emerging in which AI agents can retrieve prices and availability, add products to shopping carts, and handle parts of the purchasing process. Google has also introduced a Business Agent, which allows people to communicate directly with brands within search results, similar to using a digital sales assistant.

Shopify is also clearly positioning itself in this direction and now refers to Agentic Storefronts, which allow merchants to make their products available on various AI commerce platforms. Shopify merchants can be found via ChatGPT, among other methods, while further integrations with Google and Microsoft are being developed.

The technology has also long since arrived on the company's own website. The Shopify App Store now has a dedicated category for AI shopping assistants, including Zipchat, Chizy, Manifest AI, and other systems that combine advice, product recommendations, and customer service.

From our perspective, the crucial question is therefore no longer whether companies can use such a technology.

The more interesting question is what changes when conversation actually works better than navigation.

Filters sort products, understand conversations and intentions.

The difference becomes very apparent, especially in e-commerce.

A classic fashion shop would ask for size, color, price, brand, material, or category.

A really good saleswoman would probably start quite differently and want to know what you need the dress for, what you usually like to wear, what you feel comfortable in, what you definitely don't want, and maybe also what shoes you already have to go with it.

Filters primarily work with product properties.

Consulting works with intentions.

A skincare shop might start by asking what's bothering you about your skin right now.

A furniture retailer might ask you to upload a photo of your living room and then explain which solutions would suit the space, light, and style.

A hotel could find out whether you want peace and quiet in the morning or prefer to wake up in the middle of life, whether you want to be out and about a lot or are traveling precisely because you want to organize as little as possible for three days.

A B2B provider might want to first understand which process is currently wasting time in a company, instead of immediately listing its individual software modules.

This changes the order in the sale.

The starting point is less the product and much more often the problem, the situation or the desire behind it.

That's precisely where conversational commerce starts to become strategically interesting for us.

A shopping assistant is far from being a strategy.

Many companies will initially respond to this development with a technical approach and integrate an AI assistant into their website that politely asks how it can help.

However, this solves very little.

A good conversational commerce approach needs much more than a language model.

The system must understand the product range, take prices and availability into account, be able to explain differences between variants, know sizes and delivery times, understand relationships between products and ideally also know which recommendations are useful in which context.

In addition, there is a level that, from our point of view, is even more important.

The system needs to understand how this brand advises.

A luxury brand should choose differently than a discount store, an independent bookstore differently than Amazon, a boutique hotel differently than an international hotel chain, and an avant-garde fashion brand can make recommendations that are bolder and perhaps even a little more unconventional.

This suddenly makes brand strategy an integral part of the technical infrastructure.

Because in this context, brand means much more than tone.

It's about which questions a system asks first, which criteria are important for a recommendation, which alternatives are suggested, when a product is discouraged, and what attitude is visible behind a selection.

A truly good salesperson possesses something that classic e-commerce systems have so far been unable to replicate: sound judgment.

He understands that the most expensive product is sometimes the wrong choice, that someone might ask for a black dress but is actually looking for a specific effect, and that good advice often consists of sensibly reducing the selection.

This type of knowledge could become scalable for the first time through conversational commerce.

The brand's voice becomes the interface

Until now, brand strategy has been visible primarily where companies have designed and communicated, in campaigns, websites, images, language, social media, packaging and physical spaces.

When a larger part of the customer journey takes place through conversations, this role shifts.

Brand strategy must also define how a system questions, thinks, selects, and recommends.

This is a fairly new task.

Because as soon as a large number of companies use the same models and similar technical systems, a digital world quickly emerges in which every brand sounds friendly, efficient, helpful, and surprisingly similar.

The more interesting question, therefore, is how a brand stance can be translated into decision-making logic.

What would this brand recommend?

What criteria are important to her?

Which products would she consciously not recommend?

How much does she explain?

How directly does she speak?

When does she disagree?

When does she tell a customer that while the product they are looking for is basically suitable, another one would probably be a better choice?

That is precisely where branding evolves from a visual and linguistic surface to something that actually influences decisions.

Perhaps the website itself will also change as a result.

Websites will likely remain visual for a long time, because people like to browse, see pictures, find inspiration, and sometimes consciously explore without a clear goal.

Especially in fashion, interior design, travel, or food, this form of exploration remains important.

However, the weighting could shift significantly.

Today, navigation is usually the central structure of a website, while search, chat, or recommendations represent additional features.

In the future, conversation could become the most important entry point, and navigation could take on more of the role of browsing.

This also changes the central question in the development of digital offerings.

Companies today are dealing with questions such as:

How do we organize our content?

What categories do we need?

How do we build our navigation system?

Which filters help?

In the future, the more important question could be:

What questions do people have before they decide to choose us?

And this question is much more interesting for a truly good customer experience.

Product categories are losing some of their power

Categories seem so natural today that it's easy to forget they are ultimately just a system of order.

A dress might be categorized as "Dresses" in the shop, then as "Midi", "Summer" or "Evening Wear".

For one person, the same dress can simultaneously be the right outfit for a wedding, something for a dinner in Italy, a good solution for someone who doesn't like patterns, a dress that works with flat shoes, or something that looks elegant without appearing particularly formal.

An intelligent system can reclassify this product each time, depending on the context.

This turns taxonomy into context.

This is relevant for brands because in the future, products will be found less and less by what they are , but rather by the situation in which they provide a good answer.

That's precisely why product data is suddenly becoming strategic.

For a system to make such decisions, it needs far more information than many companies currently provide.

“White cotton blouse, 149 euros” provides relatively little information.

Ideally, a system would need to know how transparent the blouse is, how it drapes, whether it looks more formal or casual, whether it fits under a blazer, how much it wrinkles, what temperatures it is suitable for, and what specifically distinguishes it from other white blouses of the same brand.

This gives strategic importance to an area that has long sounded like back office or IT.

Product data becomes part of marketing and brand communication.

Because the more AI systems are expected to understand and recommend products, the more important it becomes to be able to explain to them why a product is relevant and in what context it is a good choice.

And that also changes SEO.

From our perspective, this is where the development becomes particularly interesting.

SEO was long based on a relatively clear relationship.

A person is looking for something, Google shows results, the person clicks on a website and their question should be answered there as well as possible.

Conversational Search changes this sequence.

Humans can express what they are looking for much more naturally and in greater detail to an AI system.

Instead of:

"high-quality black blazer for women"

A possible request could read:

"I am looking for a high-quality black blazer that looks professional in the office but not too formal in the evening, has a close-fitting cut, and costs a maximum of 400 euros."

The system can derive various search intentions from this, combine information from multiple sources, evaluate products, and then present a small selection.

This means that brands are no longer competing solely to rank highly for a specific search term, but increasingly to become part of a relevant answer.

Google itself explicitly emphasizes that classic SEO fundamentals remain relevant for AI Overviews and AI Mode because these systems build upon existing search and ranking systems. At the same time, Google recommends high-quality, original content and now explicitly explains how content should be prepared for generative search experiences.

However, in our view, the consequences of this are greater than the question of whether we will be talking about SEO, GEO or AEO in the future.

The next generation of SEO optimizes less for words and more for meaning.

Keywords remain relevant, as do technical SEO, structured data, crawlability, internal linking, and a clear information architecture.

However, this creates an additional layer.

Companies need to be able to explain things much better:

Who is this product suitable for?

In what context is it relevant?

What problem does it solve?

What distinguishes it from alternatives?

What specific characteristics demonstrate these differences?

What questions do people ask before making a purchase decision?

What expertise does the brand possess on this topic?

In parallel, Google has announced new Merchant Center features designed to help merchants understand how their brands and products appear in AI-driven search experiences, while new product attributes and UCP features are intended to make commerce data more understandable and up-to-date for AI systems.

As a result, SEO is increasingly evolving from a discipline of visibility to a discipline of being understood.

In the future, brands must be understandable to both humans and machines.

This is precisely where we see one of the biggest strategic breaks.

In many companies, SEO, brand, e-commerce, product data, PR, customer service and technology are still largely treated separately.

The SEO team deals with rankings.

The e-commerce team maintains products.

The brand develops campaigns and positioning.

PR builds reputation.

Customer service understands the real questions customers have.

IT is responsible for data structure and systems.

In a conversational world, these areas are becoming increasingly intertwined.

Customer Service is where the questions people actually ask come from.

Brand strategy determines the attitude with which a brand responds.

Products, prices, and availability come from e-commerce.

SEO and content provide the context that makes information findable and understandable.

Authority is created through PR and external sources.

Technology provides structured data, interfaces, and systems through which information can be processed.

From our perspective, conversational commerce is therefore much less a chatbot project than an organizational and strategic project.

How ready is a brand for conversational commerce?

We would therefore consider this question on five levels.

1. Findability

Do search engines and AI systems even understand who the brand is, what it stands for, and what questions it might be relevant to?

This is where classic SEO, content, PR and increasingly AI search optimization meet.

2. Product Intelligence

Is product information so detailed, structured, and up-to-date that humans and machines can recognize real differences and derive recommendations from them?

This includes technical product data as well as application situations, comparison information, availability and context.

3. Brand Intelligence

Is it clearly defined how the brand advises, what criteria it uses to select clients, what stance it takes, and which recommendations actually align with its values?

A brand voice alone is hardly enough.

What's needed is brand logic.

4. Conversation Design

What questions should a truly good digital consultant ask to understand a client's needs?

What information does he need?

When should he ask?

When do we recommend it?

When to reduce?

When to object.

This is where UX, service design, and sales psychology meet.

5. Conversion Infrastructure

Can a good recommendation actually lead to action?

Is the system connected to availability, shopping cart, CRM, checkout, or other relevant systems?

This is precisely the point at which it is decided whether an interesting conversation will actually turn into commerce.

Which tools should you look at today?

Companies don't need to rebuild their entire website today, because there are now a number of technologies that allow them to test individual elements.

For Shopify stores, tools like Zipchat, Chizy or Manifest AI already offer conversational product discovery and advice directly in the store.

Platforms like Shopify are simultaneously expanding the infrastructure to make products available outside their own shop via Agentic Storefronts.

Google is further developing the commerce infrastructure at the search and platform level with UCP, Business Agent and Universal Cart, while OpenAI is establishing its own form of conversational product search with ChatGPT and the Agentic Commerce Protocol.

This results in an interesting strategic sequence for companies.

In our view, the first question should not be:

Which tool should we install?

Rather:

Which decision made by our customers would we like to improve?

Because technology can scale consulting, but it can only build on the knowledge that a company actually possesses and makes available in a structured way.

What does this mean for companies today?

Therefore, the most sensible starting point lies in one's own customer journey.

Where do customers today have to search, compare, or understand unnecessarily?

What questions come up repeatedly in customer service?

What differences between products are difficult to discern on the website?

What information do salespeople or consultants who are barely visible online have access to?

What questions does a truly good employee ask before making a recommendation?

Which products are frequently bought together or compared?

Where do people give up?

And especially:

What questions do we want to be one of the best answers to in the future?

We find this last question much more interesting than the classic question of which keywords a company wants to rank for.

Because as people increasingly interact with systems that search for, combine, and evaluate information, visibility alone becomes less and less sufficient.

Brands need to be understood.

Their products need to be understood.

Their expertise needs to be understood.

And their stance must be so clear that it remains recognizable even when an AI system suddenly stands between the brand and the person.

And what remains of the website?

Perhaps the website won't disappear at all.

Perhaps she's just changing her role.

A place we click our way through could increasingly become a place we talk to.

Categories become situations.

Filters become questions.

Search becomes consultation.

SEO is becoming increasingly about comprehensibility in addition to visibility.

And a customer journey that was previously largely dictated by the company is emerging as a journey that is increasingly based on what a person actually wants to achieve at that moment.

Therefore, the most exciting development for us is not the chatbot.

It will be interesting to see what happens when companies suddenly have to define how their brand thinks, advises, selects and judges, while simultaneously ensuring that this knowledge is equally understandable to people, search engines and AI systems.

Because as soon as people start talking to brands, it's no longer enough to just look good.

A brand needs to know who it is, what it knows, and why it might be the right answer.