UK fashion sales recorded a seventh consecutive week of growth, with like-for-like sales up 0.85% according to the BDO High Street Sales Tracker. The composite figure is unremarkable. The split underneath it is not.
In-store fashion sales rose 6.72%. Online fashion sales fell 1.53%, the third consecutive week of decline. The entire category is being carried by physical retail while the channel that has driven fashion's growth for two decades goes backwards.
That is genuinely under-covered, and it matters to an argument this publication makes constantly. But it requires an honest reading before it can be used at all, because the comparatives make the raw figures considerably less dramatic than they appear.
The base effect, stated before anything else
Like-for-like figures compare a week against the same week a year earlier, which means the previous year's performance determines how impressive this year's looks. Those bases are wildly different here.
In-store growth of 6.72% is measured against a base that itself grew 5.26%. Online decline of 1.53% is measured against a base that grew 22.78%.
That changes the picture substantially. Online is not collapsing; it is failing to grow against an exceptionally strong comparative, which is a very different statement. A channel that expanded nearly 23% last year and is roughly flat this year has still added an enormous amount of volume over two years. In absolute terms, online may well remain far larger and further ahead than the percentages suggest.
Anyone presenting these numbers as evidence that people are abandoning online shopping is misreading them, and this publication would be doing the same thing it criticises elsewhere if it presented them that way.
There is a second caveat that cuts the same direction. Reporting from the same period notes that July's heatwave suppressed high-street footfall, with shoppers staying indoors. That should have depressed in-store performance rather than flattering it, which arguably makes the 6.72% more interesting, but it also indicates that weather is exerting real influence on these figures in a period when weather has been unusual. Weather-affected weeks revert.
What survives the caveats
Having discounted the figures properly, something does remain, and it is worth stating precisely.
Over seven weeks, growth in UK fashion has come from physical stores rather than from screens. Whatever the base effects, the direction of travel within this period is that the in-store channel is contributing and the online channel is not. After two decades in which the assumed trajectory ran entirely the other way, a sustained period where that reverses is worth noticing even if it proves temporary.
The interesting question is not whether stores are winning, which the data does not establish. It is why online growth has stalled at all, in a channel that was until recently assumed to have unlimited runway.
Three explanations, and the one that matters to a reader
The first explanation is saturation. A channel that grew 22.78% in a year has captured a great deal of the demand available to it, and growth rates fall as penetration rises. This is ordinary and predicts nothing about quality.
The second is returns. Online clothing carries return rates that are enormously higher than in-store, and retailers have spent the past few years actively discouraging them: charging for returns, tightening windows, deactivating serial returners. That improves profitability and reduces recorded sales, so some of the online decline may be retailers deliberately suppressing the least profitable transactions rather than customers leaving.
The third is the one that connects to everything this publication publishes. Buying clothing online requires accepting, in advance, that you cannot assess the thing you are buying. You cannot feel the cloth, test the shoulder, examine the lining, work a zip or judge how a garment hangs. You are relying entirely on photographs and description, which is to say on the seller's account of its own product.
Every quality framework published here depends on physical contact with the garment. The weight test requires lifting it. The shoulder test requires reaching inside. The recovery test requires stretching the ribbing. None of that is available through a screen, and no amount of photography substitutes for it. If a meaningful number of shoppers have concluded that buying clothes unseen produces disappointing results, they have arrived at this publication's position by direct experience rather than by reading about it.
Why the returns figure is the real evidence
The clearest indication that online clothes-buying has a quality problem is not any sales statistic. It is the return rate.
Fashion has return rates far above almost every other retail category, and the reasons are specific rather than general: garments do not fit, or the fabric is thinner than it appeared, or the colour is wrong, or the thing simply looks and feels cheaper in the hand than it did in the photograph. That is a category-wide, structural admission that the online experience routinely fails to convey what a garment actually is.
Consider what a return actually represents. It is a purchase where the buyer's expectation, formed entirely from images, collided with the object and lost. Fashion has institutionalised that mismatch as a cost of doing business, built reverse logistics around it, and priced it in. But it remains a system in which a large proportion of transactions are wrong, and the wrongness is almost always discovered at the moment the garment is first touched.
That is the argument for handling clothes before buying them, made in the industry's own operational data rather than by an editor.
The environmental point nobody makes
There is a consequence worth naming, because the standard assumption runs the other way.
Online shopping is frequently presented as environmentally preferable, on the reasoning that one delivery van replaces many individual car journeys. That holds when the purchase is right. It fails badly when the purchase is wrong, because a returned garment travels twice, requires processing and inspection, and is sometimes not resold at all.
A garment bought in person after being examined, and kept, involves one journey and no reverse logistics. A garment ordered in three sizes, of which two are returned, involves several. Buying well the first time is an environmental act as much as an economic one, and physical inspection is what makes buying well the first time possible.
What to actually do with this
One. Handle anything you intend to keep for years. Coats, tailoring, knitwear, shoes and bags reward physical inspection more than anything else, because construction is what determines whether they survive, and construction is invisible in photographs. These are the purchases to make in person wherever you can.
Two. Use online for the known quantity. Repeat purchases of something you have owned before, from a maker whose sizing and cloth you already understand, carry almost none of the risk described here. That is what online is genuinely good for.
Three. Treat product photography as advertising. Images are lit, styled, pinned and edited to present a garment favourably. They are not evidence about cloth weight or construction, and they are not intended to be.
Four. Ask for the specifications when buying unseen. A maker who will tell you the fabric weight, the composition, the lining and the construction is giving you the information physical inspection would have provided. A seller who cannot or will not is asking you to buy on trust alone.
Five. Count your returns honestly. If a substantial proportion of what you order goes back, the problem is not bad luck. It is a purchasing method that does not give you enough information, and the fix is to buy fewer things with more contact rather than more things with less.
Where this lands across the four channels
One. The vintage and estate market. The strongest beneficiary of the argument. Vintage buying is overwhelmingly physical, unique items cannot be reordered, and the whole practice depends on inspecting an object that has already proven durable. It is the purest form of buying with your hands.
Two. Small independent designers and craft workshops. The tier most willing to supply specifications when you cannot handle the garment, which is the closest available substitute for physical inspection.
Three. The accessible-luxury tier. Worth it where a maker publishes real detail on cloth and construction, and worth handling in person for anything substantial.
Four. Selective use of mainstream luxury. Where the price makes physical inspection non-negotiable. Nobody should spend at that level on the basis of photographs.
And the universal skip: the mid-tier mass market. Which depends most heavily on photography doing work the garment cannot, and which is where the gap between the image and the object is widest.
The honest caveats
Beyond the base effects already covered, physical shopping is simply not available to everyone. Rural readers, disabled readers, people without time or transport, and anyone whose local high street has lost its good shops do not have the option of handling things first. Recommending in-person inspection as though it were universally accessible is exactly the kind of thoughtlessness a publication in this position should avoid.
Online retail also delivers genuine benefits, particularly access. A small independent maker in another country is reachable only online for most people, and that reach is one of the better developments in fashion over the past decade. Nothing here is an argument against buying online; it is an argument about which purchases suit which method.
And seven weeks is a short series. Weather has been unusual, comparatives are distorted, and this could reverse entirely by autumn. The structural point about physical inspection stands on its own logic rather than on these figures, which is fortunate, because these figures are not strong enough to carry it alone.
The honest takeaway
The numbers are more modest than they first appear, and anyone using them to declare a return to the high street is overreading a distorted comparative. What they do offer is an occasion to notice something that was true before this data and will be true after it: clothing is a physical product, and almost everything that determines whether it is worth owning can only be assessed by touching it.
The industry knows this, which is why fashion carries return rates that would be considered catastrophic in any other category. Those returns are the sound of people discovering, on first contact, that the object is not what the image promised. You can avoid a great deal of that by buying fewer things and getting your hands on them first, and where that genuinely is not possible, by demanding the specifications that would have told you the same thing. The photograph is the seller's account. The cloth is the fact. The next move is yours.
Frequently Asked Questions
What did the latest UK fashion sales data show? Like-for-like fashion sales rose 0.85%, a seventh consecutive week of growth, with in-store sales up 6.72% and online sales down 1.53%, the third consecutive week of online decline. The composite growth in the category was therefore carried entirely by physical retail rather than by online channels.
Does this mean online shopping is declining? Not in the way the figures suggest. Online's 1.53% decline is measured against a base that grew 22.78% the previous year, while in-store growth of 6.72% is against a base that grew 5.26%. Online is failing to grow against an exceptionally strong comparative rather than collapsing, and over two years it has still added enormous volume.
Why might online fashion growth have stalled? Three plausible reasons: saturation, since a channel growing nearly 23% has captured much of its available demand; deliberate suppression of returns, as retailers charge for them and tighten policies, which reduces recorded sales; and the underlying difficulty that buying clothing online means accepting you cannot assess the garment before purchase.
Why does handling a garment matter so much? Because the properties that determine whether clothing is worth owning are physical: cloth weight, shoulder construction, lining quality, hardware, how a garment hangs. None can be assessed from photographs, which are lit, styled and edited to present the item favourably. Fashion's very high return rates are essentially a category-wide record of expectations formed from images colliding with the object.
Is buying in person better for the environment? Frequently yes, though not automatically. Online delivery can replace multiple car journeys when the purchase is right, but a returned garment travels twice, requires processing and is sometimes not resold. Buying well the first time avoids reverse logistics entirely, and physical inspection is what makes buying well the first time possible.
