Do British shops still ship to Germany? We tested 49 checkouts in 2026 (Part 3)

British brands. A German delivery address. What happens between the country selector and the postcode field. 

The goal was to find out how British shops handle EU delivery and returns in 2026 - five years after the rules changed, and long after the industry stopped writing about it.

We did not start with 49. That number is what was left after cutting a much longer list of British consumer brands - fashion, footwear, beauty, homeware, cycling - down to the ones an ordinary German shopper would realistically find and afford.

So we tried to buy from them. Between May and August 2026 we approached those 49 brands as an ordinary German shopper: one real, self-funded order per brand wherever the shop would accept it, delivered to a German consumer address, then returned through that shop's standard customer flow. No press contact, no research disclosure.

Before any of that could happen, the shop had to accept the order. This first part is about that step: how many of the 49 did, at what point in the checkout the others stopped us, which categories are affected, and whether the cross-border layer everyone worries about - duties, tracking, clearance - is the thing that actually breaks.

Parts 2 to 4 cover delivery times, return labels and the money.

Key findings at a glance

17 of 49 brands, or 35%, do not ship to Germany at all. Not restricted, not surcharged. In some cases the country is absent from the dropdown; in most, the country is selectable and the basket only refuses once a German postcode is entered.

An 18th brand priced itself out. Shipping cost more than the item in the basket. We closed the tab. Total attrition before purchase: 18 of 49, or 37%.

The blocked brands are not a random sample. British heritage footwear, British knitwear, country and outdoor clothing, high street womenswear. The categories Europe most associates with Britain are the categories least available in Europe.

The block is set per brand, not per platform. Three brands belonging to one high street group refused the German address while a fourth brand from the same group sold to us, shipped, and processed a refund. Nobody made a group-level decision. Nobody made any decision.

At least eight brands in our list run split UK and EU domains, and the split is where accountability disappears. We ordered from six of them. In one case we ordered from one company's website and the parcel arrived from a different company entirely — which is why 49 brands produced 50 trading entities in this test.

Where a shop did sell to us, the technical cross border layer worked. All 31 orders shipped DDP. Tracking worked from Britain to Germany in 31 cases out of 31. Customs stopped exactly one parcel.

Test scope: 49 British brands, one real self funded order per brand wherever the shop would accept it, delivered to a German consumer address, then returned through each shop's standard customer flow. 31 orders placed. Baskets roughly 15 to 70 euro.

How we ran the test

Between May and August 2026 we built a list of British brands selling online direct to consumers. No marketplaces, no resellers, no aggregators. The list was drawn from the brands a European shopper actually lands on when they search for British clothing, footwear, outdoor kit, bikes or beauty products, rather than a random sample of everything with a UK address.

Forty nine brands, deliberately mixed across five groups:

• British heritage footwear and knitwear, the long established makers with a story on the About page.
• Country, outdoor and cycling, including specialist retailers and manufacturers.
• High street and mid market womenswear, including several groups running multiple labels off one shared platform.
• Premium fashion houses and designer labels.
• Direct to consumer challengers in streetwear, activewear, sustainable clothing and beauty.

Every brand was approached as an ordinary German consumer. Where the shop would accept a German address we placed one real, self funded order, baskets of roughly 15 to 70 euro, and then sent it back through that shop's standard customer return flow. No press contact, no research disclosure, no special handling. Thirty one orders went through. Eighteen brands never let us get that far, which is the subject of this article.

One of those thirty one orders arrived from a company we had never bought from, so the test touched forty nine brands and fifty separate businesses. We will come back to that.

The forty nine were not picked for drama. They were picked to represent what a European customer actually encounters when they try to buy British online, which means the list contains as many quiet, competent operators as it does failures. Where there was a choice inside a category we took the brand sitting closest to the middle of it rather than the outlier.

Still, one order per brand is a snapshot, not a census. That is exactly why we do not name brands. A single incident says very little about a company, but forty nine incidents side by side say a great deal about a market. Every observation in this series is attributed to a category, never to a shop.

A postcode field is a business decision that nobody reviews

Thirty five percent is not a rounding error and it is not a technical limit. Every one of these companies could ship to Germany tomorrow. Several of them did, before 2021.

And here is the detail that makes it worse than a closed door. Our tester's note on the blocked shops reads that in these cases the error only appeared once she entered the German postcode. Germany was in the dropdown. The country page loaded. The basket held. The refusal came at the address form, after the browsing, after the product choice, after the decision to buy.

That is not a market you have declined to enter. That is a market you advertise into, pay to acquire from, and then reject on the last screen. Every euro of that traffic was bought.

What the pattern shows is where post Brexit friction actually landed. It did not land evenly across British e-commerce. It landed on the shops whose average basket was too small to carry the compliance overhead, and then it stayed there, quietly, inside a shipping rule that nobody reopens once it is written.

The categories are the uncomfortable part.

Start with heritage footwear. Long established British shoemakers, the kind with two centuries of history on the About page and a waiting list for a resole. Most of the ones we approached would not send a pair to a German address. The one that did charged the highest delivery fee in the entire test, sent the parcel into a customs hold, and never delivered it - which we will come to in Part 2. British knitwear tells a blunter story: a brand whose entire proposition is where the wool comes from and who spins it, unavailable in the largest market on its doorstep.

Country and outdoor clothing is the category that should be the easiest export on the list, produced by the country that invented most of it. It came back split down the middle. Four of the brands we approached would not ship at all. Two did, and both of them ended up among the better performers in the study. Same category, same border, same season, opposite answers - which tells you the block is not a category constraint. It is a per company habit.

And in high street womenswear, the most instructive case in this section. One group runs several labels off one shared back end. Three of them refused a German address. A fourth, sitting on the same infrastructure, took the order, shipped it, and refunded it. There was no portfolio decision. There was no platform limit. Three country lists were trimmed at some point and one was not, and nobody has looked since.

Compare this to the German test, where the equivalent figure was zero. Every German shop sold to a German customer, because of course it did. The point is not that domestic beats cross border, which is obvious. The point is that roughly a third of the British market has quietly reclassified its nearest 84 million consumers as somebody else's problem, and the decision lives in a configuration file rather than in a strategy document.

So the question for a British seller is not whether Germany is worth entering. It is whether anybody ever actually decided to leave. Most shops in this bracket never held that meeting. A platform migration, a carrier renegotiation or a VAT panic in 2021 trimmed the country list, and nobody has reopened it since. That is not a strategy. That is sediment.

The shop that charges more to send it than to make it

One brand gave us the more honest version of the same decision.

A ceramics and homeware brand, well known, premium, exactly the sort of gift purchase that travels well and carries a healthy margin, did offer German delivery. The shipping quote came in higher than the product in the basket. We did not complete the order, and neither would any customer.

This is not a blocked market. It is a blocked market with extra steps, and it is worse for the seller than an honest "we do not ship here", because the customer has already spent the browsing, the choosing and the checkout before finding out. That is a visitor you paid to acquire and then insulted on the final screen.

And if your shipping rate can exceed your item price anywhere in your range, you are not offering international delivery. You are running an acquisition funnel that terminates in a shrug. Take your cheapest product, pick your most distant country, and go and see what your own checkout quotes.

The two domain problem

At least eight brands in our list run separate UK and EU storefronts, and we placed orders with six of them. In principle this is correct architecture: local pricing, local VAT, local returns, local law. In practice it produced the two most confused cases in the entire test.

In the first, we ordered from a cycling retailer's website. The parcel arrived from a different cycling retailer: same group, different brand, different name on the box. When we tried to return it we were suddenly dealing with a company we had never transacted with, and the two shared a support function that could not tell us which of them owned the transaction. Support ultimately did not permit the return at all. The item is still here. This is the fiftieth company in a test of forty nine brands, and we met it only because a parcel turned up with its name on the label.

In the second, a menswear brand operating across two domains. The order went through, payment cleared, the parcel arrived. The confirmation email never did. The returns portal requires an order number, and no order number exists. Both domains were contacted. Neither produced one. That parcel cannot be returned, because from the shop's side there is no evidence the customer was ever there.

Neither of these came from unusual customer behaviour. We paid, we received goods, and we could not exercise a basic consumer right. In one case because the seller's identity was ambiguous, in the other because the confirmation email that anchors the entire downstream flow simply never fired.

A split domain setup makes an implicit promise: the customer will always know which company they are dealing with, and either side can recognise the other's orders. In two of the six split setups we bought from, that promise failed, and both times it failed at the return stage, which is the most expensive place available.

What actually works: the boring layer

Every one of the 31 orders shipped DDP. All duties and taxes were settled before delivery. Not one courier turned up asking a German consumer for money on the doorstep. The nightmare scenario of 2021, the surprise customs invoice at the door, did not occur a single time.

One caveat worth stating plainly, because it runs through the rest of this series: DDP status settled the money, not the movement. One of those 31 parcels was still held at the border and never completed its journey. Prepaid duty is not the same thing as cleared freight.

Tracking worked from Britain to Germany in 31 cases out of 31. In the German domestic test that number was 19 out of 20. The cross border study scored higher on tracking than the domestic one did.

And customs delayed exactly one parcel. That order came from a footwear brand charging the highest delivery fee in the test, 25 euro, went into a hold and never came out. One incident in 31.

So the layer everybody worries about is solved. Duties, tracking, clearance: handled, standardised, invisible, absorbed by carriers and software while the industry was still writing think pieces about it.

Which makes the rest of this series harder to read. Everything that broke, broke somewhere a competent operator controls directly: the delivery estimate on the checkout page, the return label that does not exist, the refund that arrives six weeks later missing most of its value.

Check this against your own operation

Five questions, thirty seconds.

Open your own checkout right now. How many EU countries are genuinely selectable - and then type in a foreign postcode, because that is a different test. When did somebody last review that list on purpose rather than inherit it?

Is your cross border shipping fee smaller than your cheapest product? Are you sure at the bottom of the range, not the average?

If you run several brands off one back end, do they all have the same country list? Ours found three that said no and one that said yes, inside one group.

If you run split UK and EU domains, does a customer on the EU site always know which legal entity they bought from, and can the return flow on one site recognise an order placed on the other?

Does every completed order generate a confirmation email with an order number, and what is the fallback path for a customer whose email never arrived?

What share of your EU traffic hits a checkout that cannot serve it? That number exists in your analytics today, and almost nobody has looked at it.

If any of those made you wince, do the cheapest market research available: order from your own shop, to a foreign address, this week. Then order from two competitors. This entire test cost us the price of the products and a few afternoons of opening boxes, and it told us more than any policy audit could.

"The most expensive customer is the one who reaches your checkout and discovers you do not sell to them. You already paid for that visit."

Pawel Zakielarz, CEO, ShopReturns

Before you open another market, find out what the returns from it will cost. We will work it out on your own data.

Get in touch

Next in this series: Customs is solved. The calendar is not.

About the study

Between May and August 2026 we tested 49 British brands selling online, direct shops only and no marketplaces, across fashion, footwear, outdoor, cycling and beauty. Each was approached as an ordinary consumer with a German delivery address. Where the shop accepted the order we placed one real, self funded purchase, then returned it through that shop's standard customer flow with no contact identifying us as researchers. Thirty one orders in total, fulfilled by thirty two companies, because one order arrived from a business we had not bought from. For each we documented the delivery promise against actual delivery, the carrier, DDP or DDU status, cross border tracking, return label availability, return destination country, return shipping cost and who bore it, refund amount, refund timing and every support interaction. Because one order per brand is an incident rather than a statistic, all results are reported as market and segment level patterns, and no individual brand is named.