Is dropshipping still profitable? The numbers, without the sales pitch

«Dropshipping is dead.» That sentence has been published every year for a decade, written by people selling a course, and contradicted daily by stores taking money.

The truth is less dramatic: the model is neither profitable nor unprofitable in itself. It is profitable when a precise calculation works out, and it is not when that calculation is wrong. The difference between the two comes down to a subtraction most people never do before spending their first pound.

Here is that subtraction, the numbers that feed it, and how to check before you pay.

What the numbers actually say

The failure rate, and what it measures

People often quote that an overwhelming majority of stores fail. That is true, but misread. The figure does not measure how hard the model is, it measures how many people open a store without having checked anything at all.

Opening a store costs a few tens of pounds and two hours. A market with an entry price that low mechanically produces a majority of abandonments, exactly like gym sign-ups in January. The failure rate mainly tells you the barrier to entry is low, not that the door is shut.

The real margins

This is where most projections collapse. A 60% gross margin on paper becomes, after product cost, platform fees, payment fees, returns and customer service, a considerably slimmer net margin. On highly competitive products it often falls below 15%.

Remember the rule: if your gross margin is less than three times your product cost, you have no room to buy traffic. And without bought traffic you need another source, which changes the whole project.

Acquisition cost, the real judge

The one number that decides everything is this: what a customer costs you, against what they bring you. Everything else follows from it.

In saturated markets, acquisition cost has multiplied several times over in a few years, because everybody buys the same placements at the same moments. That is the sector's real transformation, far more than any product fad. It is also why looking at what others spend, and for how long, has become a prerequisite rather than a luxury. We set out what is genuinely observable in our article on competitors' ads.

What has changed, and will not change back

The end of the novelty advantage

There was a time when simply showing an unknown product was enough to sell it. That window has closed: the same product video circulates across every network within days, and the discovery effect now lasts weeks rather than months.

The direct consequence: a product does not belong to you, only your execution does. The page, the offer, the service, the price.

Delivery times

Buyers have grown used to receiving things quickly. A three-week delay announced in small print now produces refund requests, payment disputes and negative reviews that cost more than the product's margin.

That single point wiped out an entire generation of stores built on the cheapest possible shipping.

The three models all called «dropshipping»

Product testing

You list quickly, buy traffic, keep what converts. It is the best known and the hardest: it demands an advertising budget, a tolerance for losses, and it builds nothing lasting if you stop there.

The built brand

Same logistics, opposite intent: you pick a segment, you look after the image, you keep customers. Profitability comes from the second order, not the first. It is slower and markedly sturdier, because acquisition cost is diluted across the customer's lifetime.

Niche arbitrage

You aim at a small, badly served audience where advertising competition is weak because the market looks too small to the big players. Volumes are modest, margins comfortable, and the position holds for a long time. It is the least talked-about model and the most accessible without a budget.

The calculation to do before starting

Take a sheet of paper and write four numbers.

The intended selling price. The cost of the product delivered to the customer, shipping and packaging included. The cost of acquiring a customer, which you will estimate from what others do in that market. And the number of orders one customer will place within a year.

If selling price minus product cost minus acquisition cost is negative, your project only stands if that fourth column is greater than one. Which is to say, if it rests on retention. Plenty of people discover that constraint after six months of losses, when it fits into four numbers written before starting.

Where the money is actually lost

Rarely where you expect. It is not the platform fees or the price of a domain name.

Losses come from ad tests stopped too late, traffic sent to a page that does not convert, refunds caused by delays, stock ordered ahead and never sold, and above all time spent on a niche nobody had checked. The most expensive line in a failed project appears on no invoice at all: the months spent working in the wrong direction.

The signs of a market still worth taking

Demand already exists

You want a market where people buy today, not one to be created. Creating demand costs infinitely more than capturing it, and it is brand work, not shop work.

The competition is soft

Numerous competitors are not a problem; good ones are. Look at their pages, their delivery promises, their replies to reviews. A market where everyone does the bare minimum is a market where being serious is enough to win.

The product can carry a margin

The price the market accepts must leave room for acquisition cost. Products under twenty pounds with no repeat purchase are almost always traps, because nothing remains after advertising. The full selection method is in our article on finding a winning product.

How to check all this without spending anything

Everything above can be observed from the outside, before investing.

You can see which stores already exist on your idea, what they lead with, how long they have been running ads, which apps they use and how fast their catalogue moves. You can estimate their traffic and above all where it comes from, which tells you whether the market is bought or earned. Our two detailed methods are in analysing a competitor's store and estimating traffic.

A day of that work saves you six months of error. It is the best ratio I know of in this trade.

The mistakes that cost the first thousand

Choosing the product before the market. A good product in a saturated market loses to an ordinary product in a soft one.

Believing advertising will fix a weak offer. It only speeds up the verdict, and charges you for it.

Looking at others' revenue instead of their margin. A store doing a hundred thousand a month can be losing money, and many are.

Killing a test after two days, or letting it run for two months. Both mistakes cost, the second far more.

Starting without writing the four numbers. It is by far the most common mistake, and the only one that makes all the others inevitable.

Frequently asked questions

Do you need an advertising budget to start?

Not necessarily, but then you need time. With no budget, your acquisition comes from search, unpaid social, or an existing audience. Those paths work and cost months instead of money. What does not work is having neither.

How long before the first pound of profit?

Count several months, and be careful to distinguish first sale from first profit. The first sale comes quickly and proves nothing. The first profit arrives when acquisition cost drops below margin, which takes iterations.

Is it better to target a niche or a product?

A niche, almost always. A product wears out and gets copied; a niche lets you renew the catalogue for people who already know you. That is what turns a shop into a business.

Is dropshipping legal?

Yes, it is an ordinary logistics method. What causes trouble is the surrounding practice: false delivery promises, counterfeits, missing legal notices. The model is not at fault, the shortcuts are.

Where to start this week

Three steps, in this order, and none of them costs money.

First write the four numbers from the calculation above, using your most cautious assumptions. If the result does not hold, you have just saved six months. Then take ten stores already present on your idea and record three things for each: how long their oldest ad has been running, where their traffic comes from, and how fast their catalogue changes. You will know whether the market is bought, earned, or already taken. Finally choose your model from the three above, according to what you have, time or money.

By hand that survey takes an afternoon per store. Evidio does it from the browser in one click on any Shopify store: ad pixels and active networks, estimated traffic and its sources, installed apps, featured products, technical make-up, with an export so your ten competitors sit in the same table. The extension is free, and our plans exist only to track that change over time.