Dropshipping unit economics in 2026: the €69 worked example

Almost every store that failed, failed on a spreadsheet before it failed in reality. Ten minutes of arithmetic would have shown it.

This piece is that arithmetic, built line by line on a single €69 product, plus the two changes that turn the same product from a loss into a business.

First, what actually broke

Three numbers moved, and only two of them moved in your favour.

The old model was arbitrage: buy attention cheaply, sell a commodity at a markup. Attention is not cheap any more and it is not going back. So the margin has to come from somewhere else.

That is the whole difference between the two versions of this business:

Note both halves of that table. Branded is roughly double the margin and ten times the timeline.

The only equation that matters

Contribution margin per order must exceed what it costs to get a customer. There is no creative solution to that arithmetic — only pricing, bundling, or a different product.

So let us build one. A €69 product:

At a €40 acquisition cost, that loses €6.35 on every order. The business as described does not work, and scaling the advertising only finds the loss faster.

Change one thing: the order bump

Add a €19 accessory at checkout, taken by about a third of buyers. Average order value goes to €75.30, and the margin climbs to €38.15. Better. Still thin.

Change one more: the second purchase

About one customer in five buys again within ninety days, at a €30 margin. Effective margin per acquired customer: around €44.

Now a €40 acquisition cost works, with room.

The practical rule that falls out of this: stay above €45, and ideally above €70. Cheap products cannot carry paid advertising, and a lower price feels like the safe choice precisely because it is the easy one.

What to aim for

Product choice is about 70% of the outcome

Nothing later rescues a product that fails these tests. There are nine criteria in the full scorecard; these four eliminate the most ideas:

1. It can be sold above €45. That alone removes most of what people consider.
2. Product cost is under a third of the selling price. Under a quarter is better.
3. It has a natural second purchase — a refill, an accessory, a size up. That is the repeat order the table above depends on.
4. It is demonstrable in three seconds. Show someone a silent three-second clip and ask what it does. If you have to explain, your cost per click will be punishing.

Score honestly out of nine. A "sort of" is a no. Under seven, keep looking — the search is free and the mistake is not.

And the product doing well in everyone else's advertising right now is the worst possible choice. By the time you can see it, the audience is saturated and a dozen stores are undercutting each other.

The €160 step that kills one product in three

Before you spend anything on packaging: order samples from three suppliers. €100 to €200, two weeks.

Roughly a third of products fail on quality alone. In the worked example, three samples cost €160 including shipping: one arrived with a wobbling mechanism, one had a chemical smell that never faded, one was genuinely good. That is a normal result — and every product that fails here would have failed at the customer stage instead, more expensively, with a refund attached.

DHgate is where that sampling stage happens. It is a testing tool, not a brand: once a product has proven itself, at around thirty to fifty orders, you move to a sourcing agent, and the difference is not subtle.

The compliance layer people skip

The least interesting chapter, and the one most likely to decide whether you still have a business in a year.

In the EU, a consumer product cannot legally be placed on the market unless an economic operator established in the EU is responsible for it — with their name and postal address on the product or its packaging. If you are the seller, that obligation is yours even if you never touch the goods. In 2026 this moved from a documentation exercise to enforcement: blocked listings, withdrawal orders, recalls.

Import VAT now applies to every parcel regardless of value, too. A customer paying an unexpected charge at the door is not a sale — it is a chargeback.

So send one message before anything else: can you provide the technical documentation and safety assessment for this product? The answer tells you whether you have a business. And stay out of categories where the answer is usually no: batteries and electricals, anything with a medical claim, cosmetics, supplements, children's toys.

This is not legal advice — it is the shape of the problem. Check what applies to your product and country.

The store, and the advertising

Most of your paid traffic lands directly on a product page and never sees your homepage, so that is where the effort goes: photo and price above the fold, a one-line promise, a buy button without scrolling, then the video, three benefits with images, reviews with photographs, specifications, delivery and returns in plain language, the real objections answered, and the buy button again.

Guest checkout, always. Shipping costs visible early. And the local payment method your market expects — in the Netherlands that is iDEAL, and missing it costs more sales than any design decision. Shopify is the default for the boring reasons: it hosts the store, takes the payment and handles the tax.

For creative, native beats polished and a real kitchen beats a studio. The lever nobody uses is shooting five angles of one product rather than five products: the problem angle, the comparison, the sceptic, the unboxing, and "if you work from a kitchen table". Once an angle wins and you sell in more than one language, Wizstar versions it with translation and lip sync rather than reshooting.

The test, with a number written down first

Set the testing budget before you start and treat it as tuition: €500 to €1,000 to prove or disprove one product. Then write down your kill number — the cost per purchase above which you stop — because in the moment you will want to give a failing creative one more day, and one more day is how €500 becomes €2,000.

The worked example: an under-desk footrest at €69, supplier cost €19, margin €35.35, plus a €19 desk mat as an order bump. Kill number €38. Budget €750. Five creatives, €20 a day each, three days.

Two under the kill number. Three switched off without discussion, because the number was written down in advance.

Round two: two new versions of the winning angle, €250 over four days. Eleven purchases at €26 each, 31% took the order bump, average order €74.90, margin about €40. Roughly €154 of profit on €250 of advertising. Thin, real, and pointing the right way.

The largest lever in the whole exercise was free: one afternoon filming five angles instead of one.

What the first order is actually for

At €40 to acquire a customer, the first order barely pays for itself. Everything you earn comes from what happens next:

• Average order value. Going from €69 to €90 does more for the business than a 20% improvement in your advertising, and it takes an afternoon: an order bump, a bundle, free shipping just above your current average.
• Five emails, written once. Order confirmation, shipping with tracking, "it has arrived — here is how to get the most from it", day seven ("how is it going?"), day fourteen (the related product, with a reason).
• A genuine second product. Not a random addition — the thing this customer needs next.

Riibase keeps the customer side organised once that starts to matter, and GuidelyPro is worth a quote if your store grows past the point where visitors can compare products at a glance. If you sell clothing, Style3D removes a round of expensive physical samples.

The honest summary

Branded dropshipping in 2026 is a real business with real margins and a real timeline: twelve to twenty-four months, and €1,500 to €3,000 to test properly, part of which you should expect to lose.

If losing that would hurt, this is the wrong model right now — and that is not an insult. Two models in the hourly-rate ranking start at effectively zero and can fund this later.