While companies race to figure out AI and agentic commerce, fashion marketers are being left with an unprecedented problem to solve. Traditional funnels, first disrupted by Gen Z, have effectively split in two. One half belongs to machines. The other to human feeling.
We’re living in a kind of Matrix era of fashion marketing where algorithms decide what gets seen, but people still ultimately decide what matters. Brands now have to speak both languages fluently.
Interaction is not discovery
If there is any consumer behaviour that encapsulates where fashion marketing is heading, it is when a shopper has decided they want a specific bag without going through the usual discovery process. They know the brand, the style, and the rough price point.
What they don’t know is whether the price quoted is reasonable, if it’s genuine, or the item’s history. So, they open ChatGPT, Claude or Gemini and ask away. By this point, however, discovery has already happened, and what follows is validation.
The brand has won the consumer’s attention, and the AI is essentially a middleman that serves to close that gap between desire and confidence.
It’s here where marketers, and more broadly brands, must understand the crucial distinction between AI as a research tool and AI as a discovery channel.
McKinsey’s The State of Fashion 2026 report captures how embedded AI has become in the wider shopping journey:
- Shopping-related searches on generative AI platforms rose 4,700% between 2024 and 2025.
- More than half of US consumers who used AI for search in Q2 2025 also used it to help them shop.
- AI-driven revenue per visit on US retail sites grew by 84% in the first seven months of 2025.
These numbers underpin the channel’s shift from a novelty to an increasingly default consumer behaviour. But what they don’t describe is where those consumers were before they opened the chatbot. In many cases, AI is not the first interaction for the consumer.
Desire is typically curated elsewhere. The likes of ChatGPT, Claude and Gemini are where people go to interrogate, validate, or act upon their desire to purchase an item.
There is also the trust element. Some 41% of consumers say they trust generative AI search results more than traditional advertising. Whether that trust is well-placed is almost irrelevant because the perception exists anyway. These numbers have led some to assume AI is now the default discovery channel. It isn’t.
What is AI actually good at?
Large language models – the tech most people refer to as ‘AI’ – are incredibly good at synthesising information and responding to natural language queries. For instance, you could ask an LLM about a brand’s heritage, price-to-quality positioning, or its reputation, and it will pull together a relatively decent picture of what exists about the company across the web.
This makes LLMs powerful tools for evaluation and comparisons in the consumer journey, i.e., the exact phase where a shopper who is interested starts to ask harder questions.
To satisfy consumer queries and, more importantly, ensure your brand is surfaced in LLM results, fashion companies must start work on generative engine optimisation (GEO). Content needs to be easily read, contextualised, and considered credible by the model to appear.
Vague product descriptions, wafer-thin brand histories, and narratives buried in image captions are not good enough. LLMs won’t retrieve this kind of information because it isn’t scannable, credible, or even vaguely contextual in relation to the user’s initial query.
This is why early movers – like Estée Lauder, L’Oréal, and Mejuri – have already begun mapping out what semantically rich, API-accessible content looks like in practice. Given the more headline-grabbing capabilities of LLMs and AI agents, this is not sexy work by any measure.
But it is fundamental to long-term success.
Further out on the horizon, it’s likely we’ll see considerable developments in agentic AI and agentic commerce. Aside from OpenAI’s partnerships with Shopify and Etsy, Amazon has also launched “Buy for Me”, which lets shoppers buy from third-party sites without leaving the app.
These may be early experiments, but the architecture they represent is significant. When comparisons are algorithmic and instantaneous, a brand’s value proposition becomes the raw material an AI agent reads and ranks.
Delivery time, price, and assortment are all dimensions that an agent assesses with no emotional investment whatsoever. The storytelling that shapes demand does not translate cleanly into structured data, meaning brands that are all in on story are at a meaningful disadvantage.
However, the flip side is also true. Brands that frantically optimise for the machines and ignore what fashion is all about – the people – will equally be at a meaningful disadvantage.
This is because machine optimisation is only half the battle. The other half is less technical, less measurable, and considerably harder to resource at scale.
Discovery is still fundamentally human
AI is an important channel for evaluation, and sometimes discovery, but much of the latter still needs to take place between a person and a brand. Current data suggests people are moving away from algorithm-mediated scrolls toward something slower and more deliberate.
Fashion has spent the better part of a decade rewiring the marketing engine around social media platforms. The logic was sound since algorithms reward speed, spectacle, and novelty. Microtrends emerged and collapsed in weeks, production cadences compressed, and influencer rosters expanded. Almost everything was optimised for velocity.
The problem is that the people the engine was built to serve have become tired of it.
It still has its place, but the same McKinsey research also shows consumers are gravitating toward longer-form content with narrative depth and brands that reflect their identities, not solely their feeds.
Almost nine in ten consumers say being part of a brand community strengthens their connection more than any other engagement tactic. Represent’s 24/7 line is a good example. It is marketed to fitness lovers, and the brand regularly hosts community events around it.
On the topic of wellness, this is now a top priority for many consumers (84% in the US, 94% in China). The values associated with it, however, are not what dopamine marketing was designed for.
Brands that built their growth engines around trend velocity now find themselves with the wrong infrastructure for the moment. The tools, KPIs, and content cadences were configured for reach and speed. Reconfiguring the engine for depth and belonging is not really a pivot either. It is essentially a rebuild.
Other brands that have navigated this transition well include Dior, with its branded spas; Lululemon, with its yoga hubs; and Missoma, with its London-based running clubs. These aren’t traditional marketing channels, though. They are the consequence of having a genuine point of view and building something around it to attract like-minded people.
These more experiential touchpoints also compound better than paid channels. Consumers who discover brands through community often arrive with more trust, identity alignment, and tolerance for price than those acquired through a promotional push.
(Just look at how successful On sales have been compared to Nike, especially at full price.)
But none of this is quick or cheap. Community programming cannot be A/B tested, and its returns will never be visible in a weekly dashboard, which is why many businesses ignore it.
Community programmes require full commitment from a brand and the nerve to hold it, which is why so few do it well, and why the ones that do find meaningful ground.
Of course, not every brand wants or needs to embark on such endeavours. Fast fashion, for example, will never really build a community like those described above and is better off leveraging microtrends and constantly pivoting with what people want in the here and now.
The point is more to show how nuanced the fashion marketing landscape has become and why neither GEO nor human connection alone can win in today’s market.
Fashion has a dual marketing imperative to fulfil
Marketing teams are now tasked with emotionally resonating with people as well as optimising their websites, product listings and articles for LLMs. Both require different skillsets.
Most organisations do not have the structure, budget, or internal consensus to run both in parallel. Nor do they necessarily understand that each imperative has a different function.
Product pages optimised for LLM discoverability have different aims and KPIs than a video ad or commercial designed to foster emotional connection with the end consumer.
Each one may be a critical touchpoint in the journey, but one is intended to capture demand while the other is needed to create it in the first place.
Treating them as interchangeable is how you end up with content that is technically complete but emotionally inert, or rich with feeling but invisible to LLMs.
Brands that understand this distinction will stop trying to solve both problems with a single content strategy and instead think carefully about what each piece of content needs to achieve.
For example, LLM-optimised content is not something that would ever justify an increase in price or change a consumer’s perception of a brand’s value.
Brands repositioning upmarket – value players retreating from their lowest price tiers or mid-market players chasing the white space left by luxury’s price inflation – can’t do it on price architecture alone. They need the story to carry the repositioning, like what Slazenger is doing.
The brand is undergoing a major creator-led rebrand spearheaded by 22-year-old TikTok creator Alexei Hamblin. He pitched a vision for reviving the brand on social media by showing an inside story of the rebrand. Slazenger subsequently gave him full creative and commercial control.
Whether or not the rebrand succeeds is yet to be seen, but, importantly, many people are now invested in Alexei’s attempt to rebuild a legacy sportswear player into a cool, modern streetwear and lifestyle brand.
Yes, TikTok is a short-form media content platform, but this isn’t a microtrend; it’s a full-blown rebrand that is being documented in real time and told over a longer horizon.
This isn’t your usual run-of-the-mill play. It’s bold, it’s interesting to younger consumers, and it’s the kind of storytelling-led repositioning that 47% of consumers cite as a key driver of high-end brand perception – something LLM-optimised content could never achieve.
If you are to take away anything from these examples, it’s that both GEO and emotional resonance are equally important for winning in this new era of fashion marketing. They both address different, but crucial, moments in a journey consumers now take by different routes.