AI Dropshipping
Find a product, generate the photography and the store around it, and let the orders arrive.
Store activity
Orders, traffic and margin
| Event | Product | Order | Margin |
|---|---|---|---|
| New order | Linen summer dress | +€49.95 | +€21.40 |
| Return requested | Cotton bucket hat | — | — |
| New order | Ribbed knit set | +€59.95 | +€24.20 |
| Fulfilled | Linen summer dress | — | — |
| New order | Cotton bucket hat | +€24.95 | +€7.10 |
| Visitor spike | Instagram referral | — | — |
- Linen summer dress3 variants24 sold
- Ribbed knit set2 variants17 sold
- Cotton bucket hat1 variant9 sold
Example figures. These amounts show how the workflow behaves — they are not earnings, not ours and not anyone else's. Nobody made this money.
Open full screen · recording mode — sized for vertical video.
Read this before the opportunity
This one demonstrates far better than it performs. Its content score is 9.5 against a business score of 5.2 — a gap of 4.3 points. That pattern is exactly what the get-rich-with-AI industry is built on: strategies that look spectacular in a short video and disappoint in practice.
The complete workflow films in one take: product to photo to store to order.
Follow these prompts to start building your income stream
6 steps. Paste each prompt into Claude, use what comes out, move to the next. The last one ends in something live — not in a plan to make something live.
- 01
Work out if the numbers add up
Free sampleA sheet showing what you must charge to make money.
In this step you find out whether the product can survive advertising costs, because that is where dropshipping fails. Most stores are killed by arithmetic that was never done: the margin looked fine until returns, ad costs and payment fees were in it. You end with a number that decides whether to continue.
1 — The unit economicsWork out the unit economics for a dropshipping product before I commit to it. Product cost including shipping to the customer: [PRODUCT COST] Planned selling price: [SELLING PRICE] Payment processing: [PAYMENT FEES] Expected return and refund rate for this category: [RETURN RATE] Platform or store costs per month: [FIXED COSTS] Give me: 1. Gross margin per unit in cash and as a percentage. 2. What is left after returns and payment fees. 3. The maximum I can pay to acquire one customer and still make money. 4. How many sales per month before the fixed costs are covered. 5. What happens to all of the above if the return rate is double my estimate. 6. The break-even ad cost per click at a 1%, 2% and 3% conversion rate. Show the arithmetic step by step so I can redo it for another product.
2 — Judge the resultHere are my numbers: [PASTE RESULTS]. Tell me plainly: 1. Whether the maximum customer acquisition cost is realistic for paid traffic in this category today, and what typically pushes it higher. 2. What has to be true for this to work — and how likely each of those things is. 3. What the honest alternative use of the same budget would be. 4. Whether I should pick a different product, and what property the next one should have. If the answer is that this product cannot carry paid traffic, say so directly.
Claude for the model; keep it in a spreadsheet you can rerun per product.
- [PRODUCT COST]
- Including shipping to the customer and any supplier fees. The number the supplier quotes is not the number.
- [RETURN RATE]
- Ask the supplier; if you cannot find out, use a pessimistic figure and see whether it still works.
- [PAYMENT FEES]
- Roughly 2–3% plus a fixed amount, depending on your processor.
Then: Redo the calculation for two more products before choosing. The first product that survives the maths is rarely the first product you liked.
Where the prompt stops: Nobody can tell you your real advertising cost in advance; it is discovered by spending. Treat every figure here as a planning assumption, and confirm it with the smallest possible live test.
- 02
Pick a supplier and order from them
In the packageOne supplier you tested yourself, with delivery times you can publish.
In this step you order from your own supplier like a customer would. Shipping time and packaging are what your reviews will be about, and they are the two things you cannot fix afterwards with better copy. You end knowing what your buyer actually receives.
- 03
Build your store
In the packageA store that looks real, with a product page that answers the questions.
In this step you build the thing people land on. The product page carries the whole sale: what it is, what it does, what it does not do, when it arrives, and what happens if it goes wrong. Generic template copy is what makes a store look like the hundred other stores selling the same item.
- 04
Make the product photos
In the packagePhotos of the real product, in a setting that sells it.
In this step you produce the pictures. AI makes the lifestyle shots affordable, but the line matters: enhancing the setting is normal, changing the product is misrepresentation and it arrives back as returns and chargebacks. You end with a set of images you could defend to a buyer holding the item.
- 05
Run your first ad test
In the packageA small test with a budget and a stop date, written down first.
In this step you spend money to find out whether anyone buys. The kill switch is the point: deciding in advance what result means stop is what separates a test from a slow loss. You end with either a working product or a clear, cheap answer that it is not.
- 06
Ship the first orders
In the packageOrders shipped, questions answered, and a decision on whether to scale.
In this step the first parcels go out, and that is where a dropshipping store is really tested. Support speed decides your chargeback rate, and chargebacks decide whether you keep a payment account at all. You end with a support system and an honest verdict on the product.
5 of the 6 steps are in AI Money
The sample above is a real step, not a teaser — it is the same depth as the rest. The package adds the other 5 for this strategy, and the same treatment for the other eleven.
Your first 7 days
- 1Decide on a supplier who publishes an EU responsible person, before choosing a product.
- 2Write your supplier disclosure and put it in the store from day one.
- 3Build one product page with generated photography and honest delivery times.
- 4Calculate the margin per unit after returns and payment fees, on paper, before spending on ads.
Your first 30 days
- 1Run a small, deliberately capped ad test and treat the budget as lost.
- 2Measure cost per acquisition against your actual margin, not against revenue.
- 3Handle the first returns yourself to learn what they really cost.
- 4Stop or continue on that number — not on the number of orders.
What you need
core means you cannot skip it. The rest only if you need it.
- ShopifycoreStorefront, checkout and order routing
- Photoroomcorefree planProduct cutouts and scene photography
- Recraftfree planConsistent brand imagery
- CapCutcorefree planAd creative editing
- KlaviyoThe email flows that make a second purchase possible
- GorgiasSupport without promising what you cannot deliver
- Molliefree planEuropean payment methods
Want the research?— what it is, what it costs, and how we scored it
What you can actually watch
A product photo being generated, a store appearing around it, then orders arriving one by one.
What it is
Your own web shop — usually Shopify — sells products that a third-party supplier ships directly to the customer. You hold no stock. AI generates the product photography, the store copy and the ad creative.
Why this opportunity
One part of this is genuinely strong: product imagery. Generating studio-grade product and scene photography without a studio is the best-evidenced AI application in the whole strategy. Time to first sale is also among the fastest in the set.
Who this is for
For someone who can buy advertising competently and read a margin sheet, and who will name their supplier publicly. It is not a low-risk starter business: in every service strategy here you risk your time, and here you risk money you cannot get back.
The business model
Retail price minus product cost minus advertising. Our research puts it plainly: in its usual form this is not a product business but an advertising-arbitrage business.
How money is made
The spread between sale price and supplier price, after ad spend, returns and chargebacks. Gross margin can look healthy; the margin that matters is the one after those three, and that figure is UNVERIFIED in the checks behind this page and thin in all qualitative evidence.
Who pays
End consumers on cold traffic from Meta or TikTok, mostly impulse purchases in the €15–60 band. No repeat purchase unless the brand means something, and a generic store means nothing.
Finding customers
Paid advertising. Not one option among several — the only one that works at scale for a shop with no brand, no SEO history and no email list. Acquisition cost is therefore the real cost of the business, not the tooling.
Delivery
Order, forward to supplier, supplier ships, customer waits, you handle support. The parts that cost the most — waiting time, returns, chargebacks — are the parts you control least.
Where AI actually helps
Unevenly distributed, and that is the diagnosis: high for product photos, model shots, ad variations and store copy; moderate at best for product research, because a model does not know what converts this week; high in volume for support, with the caveat that you are liable for what your bot promises (Air Canada v. Moffatt, BC Civil Resolution Tribunal, February 2024). Net: AI cuts the costs that were already low and leaves the ones that eat the margin untouched.
Skills you need
Medium to high, on an axis most people do not expect: not product knowledge but media buying, creative testing and margin arithmetic. Without those you burn budget without learning why.
What it costs to start
Tooling is verifiable and modest: a Shopify plan plus an image tool — Flair from $8/month with a commercial licence only from Pro+ at $26/month (Tier 1, flair.ai/pricing), Botika Pro $55/month (Tier 1, botika.com/pricing). Ad budget is the real number and it is UNVERIFIED: no Tier 1/2 source for CPM, CPC or CAC benchmarks in this category was found, and every circulating figure comes from people selling courses. What can be said without a number: ad budget is an order of magnitude larger than tooling, it is gone the moment a creative fails, and it is not recoverable.
How to price it
Retail pricing set against ad cost per acquisition. If you cannot state your CAC, you do not have a price — you have a guess.
Common mistakes
Hiding the supplier. Since 13 December 2024 EU product safety rules require a named EU responsible person in every listing; print suppliers publish one, Asian dropshipping suppliers generally do not. Shopify separately requires disclosure — "You should inform your customers if you use a third-party supplier" — and reserves the right to suspend any account without notice, while Stripe prohibits facilitating transactions on behalf of an undisclosed seller. Ship from a supplier you have not named and you are offside with both your platform and your payment provider, which is how stores get frozen with money inside them. Also: New York law in force since 9 June 2026 requires disclosure of AI-generated people in advertising, and that obligation is the merchant's.
Scaling
Scaling means spending more on ads, which means the model only scales while the creative keeps working. There is no compounding asset underneath unless you deliberately build a brand.
The version that holds up
Work only with suppliers who can be named and who publish an EU responsible person, and treat that constraint as the product-selection filter. It rules out most of what course content recommends, which is the point — what remains can be built into an actual brand with repeat purchase.
Time to first revenue
2–8 weeks
2–8 weeks to a first sale, assuming (a) an ad budget you can afford to lose, (b) multiple creatives tested, and (c) acceptance that early sales are unprofitable. That is the wrong measure, though: this figure marks the moment you start spending money, not the moment you make it. Time to first profit is unknown and UNVERIFIED.
How it scored
Ten axes, scored. Risk scored 9 out of 10, where higher is worse.
The state of this market
This one works, but only one way, and it is not the way most people set it up. The problem is not saturation, it is that the standard way this is done breaks two rules at once. Since 13 December 2024, EU product safety rules require a named EU responsible person in every listing; print suppliers publish one, Asian dropshipping suppliers generally do not. Shopify separately requires you to disclose the third-party supplier, and Stripe prohibits facilitating transactions on behalf of an undisclosed seller. Ship from a supplier you have not named and you are on the wrong side of both your platform and your payment provider — which is how stores get frozen with money in them. It is workable with suppliers who can actually be named, and that constraint changes which products are worth selling. Note too: New York law in force since 9 June 2026 requires disclosure of AI-generated people in advertising, with the obligation on the merchant.
Not sure this one fits you?
Answer a few questions about budget, time and skills and see which of the eighteen match your situation — and which to avoid.
Run the Opportunity Finder