In retail the address
decides. Online there
is no address.

A shop on Bond Street is judged before anyone walks in. The street has already answered the questions that matter: who else is here, what this costs, whether the house is serious. Online none of that exists. No street, no neighbour, no doorway that took a hundred years to earn. There is a line of address, and whatever loads after it.

What the address was actually doing

Physical retail never sold only the product. It sold the pre-qualification that came with the location. A house on Place Vendôme borrows the credibility of every house around it. A ground-floor space inside an old department store borrows the floor it stands on. The rent was never for the square metres. It was for the sentence the address says before the brand opens its mouth.

Online that sentence has not grown quieter. It is gone. A visitor arriving from a search result or a link lands in a space with no surroundings. Whatever pre-qualification is going to happen has to happen inside the page, in the first seconds, before anyone has looked at a product.

We measured what the purchase moment actually looks like

To see how much room for differentiation is left, we looked at what luxury shops are built on. The sample: the top thirty organic results in the US for three queries, “luxury online store”, “luxury fashion online shop” and “designer online boutique”, deduplicated to 48 shop domains. For each one we read the homepage and identified the commerce platform from its page signatures.

Thirty-four were readable. Thirty of them run on one of two platforms: nineteen on Shopify, eleven on Magento. Four use a system of their own. That is eighty-eight percent on rented architecture. Same cart, same checkout, same product-page grammar, same payment row. The moment of the decision is literally the same software.

The second finding is the sharper one. Fourteen of the forty-eight could not be read at all, and they are precisely the names you would expect: Net-a-Porter, Mr Porter, Farfetch, Bergdorf Goodman, Saks Fifth Avenue, SSENSE, 24S, Vestiaire Collective. Houses that build their own infrastructure also defend it. The measurable middle is the part running on something it rents.

The four houses running their own system share something that cannot be expressed in software: a purchase path timed differently from the standard. Longer routes, fewer steps per screen, no urging. That is expensive, and it is the only place where anything can still be differentiated inside the purchase moment. For everyone else the moment belongs to the platform, not to the house.

Measured on 14 September 2026 and reproducible: sample drawn from the search results, platform detection from page signatures. Blocked domains are excluded from the percentage rather than guessed at.

Why platform sameness is not a technical problem

None of this argues against Shopify. It is excellent software, and for most houses building your own checkout is a six-figure way of reproducing something that already works.

The problem is what follows. If the purchase moment is identical across the category, then the purchase moment cannot be where you differ. Every argument that lives inside the cart, free returns, a trust badge, a payment icon, a countdown, is an argument your competitor makes with the same words in the same position in the same template.

Which means the decision has already been made by the time someone arrives there. Not inside the funnel. Before it. What happens in the cart is confirmation, not persuasion, and treating it as persuasion is how houses end up optimising the one part of the journey they do not own.

The three reflexes that get tried there anyway

When a house notices it is not cutting through at the purchase moment, it almost always reaches for the same three devices. All three work against its own position.

The first is the trust signal. Seals, review stars, payment icons, notes about free returns. Each one is harmless, and together they answer a question a confident house would never have raised. Reassure a lot and you have conceded there is cause for unease.

The second is scarcity. Only two left, offer ends in four hours. That works where price is the main argument and works against any house whose price is justified by composure. A brand that pushes has left the rank it claims.

The third is more copy. Where differentiation is missing, description grows. What ends up on the page is a product that explains everything, and that is the admission: what is being shown does not speak for itself.

What has to carry the pre-qualification instead

In physical retail the address did three things. Online each one has to be built.

It said who else belongs here. The street did that through adjacency. Online it comes from who is visibly standing around the brand: which publications write about it, which houses are named alongside it, whose name is attached to it. Not testimonials. Third parties.

It said what this costs, before anything was picked up. A doorway on Place Vendôme shows no price and communicates the price level perfectly. Online the equivalent is register: typography, pace, restraint, how much white the page can afford, how little it explains. A house that reassures too much has already said it is nervous about the number.

And it said that things are serious here. The street proved permanence simply by having been there. Online the substitute is depth: real work shown in full, named people, positions stated rather than implied, a view somebody had to think about. Thin sites read as temporary because temporary sites are thin.

The side-by-side test for online stores

Take your product page. Remove the logo, remove the product, and leave the layout, the typography, the spacing and the register of the microcopy. Place it beside two competitors treated the same way.

If a stranger cannot tell which is yours, the brand is not doing the work the address used to do. That is not a design problem to be fixed at the end. It is a positioning problem that shows up at the end. The same question is asked systematically in the Signature Brand Audit, there for the whole outside perception rather than a single page.

The order in which this can be fixed

Houses that recognise the problem almost always start at the wrong end, with the design of the page. That is understandable, because it is the most visible part, and it is the most expensive order to work in.

Position comes first: who this house is for, and who it is explicitly not for. As long as that answer is missing, design has no standard to measure itself against, and every decision becomes a matter of taste.

Register follows from the position. How much is explained, how fast the pace is, how much surface is allowed to stay empty. This is the part competitors find hardest to copy, because it requires consistency across many small decisions.

Then the proof: who writes about the house, which work is shown in full, which names are visible. And only then the shop itself, which by that point has nothing left to do but execute what was already decided.

What this means for a house that sells online

The conclusion is uncomfortable for anyone who treats the online shop as a sales channel and the brand as decoration on top of it. In physical retail pre-qualification could be bought with rent. Online there is nothing to rent.

The brand is not the layer above the shop. It is the only part of the shop that is yours. Everything else, and we measured it, is the same software as everybody else.

The encouraging part is the reverse side of that. If technology no longer provides differentiation, it is also no longer an advantage a larger competitor can buy. A house making sixty pieces a year with a clear position stands in the same cart as a group making sixty thousand. What decides who is remembered first no longer sits in the budget.