Many new online stores make the same costly mistake, in the same order: they pick a product because it looks popular, order stock, start ads, and only then find out whether people will buy it at a price that leaves a profit. By that point the money is already spent.
You can turn that order around. Before you buy stock or spend on ads, check five things: demand, competition, your real margin, the supplier, and timing. None of them guarantees a product will sell, but together they tell you whether a product is worth testing, and how much you can afford to risk on it.
1. Check that demand is real and not just a spike
A product that went viral last week is not the same as a product people keep buying. Look for steady or growing interest over months, not a single sharp peak.
- Search interest over time. Google Trends is free. Compare the last 12 months with the last five years. A line that climbs and holds is a better sign than one tall spike that falls away.
- Proof that people already buy it. Look at marketplace listings for the same kind of product: how many sellers there are, how many reviews they have, and how recent those reviews are. Recent reviews mean recent sales.
- What buyers complain about. Read the three-star and one-star reviews. Repeated complaints show you what a better version, a better listing or better support could fix.
- Where the conversation is. If people ask about the product in forums, groups and video comments, there is interest you can reach. If no one talks about it, you will have to create the demand yourself, which costs more.
2. Size up the competition honestly
Competition is not automatically bad. It usually means there is a market. What matters is whether you can offer a reason to buy from you.
- Who sells it now, and at what price range?
- Are competitors running ads for it? Ad libraries, such as Meta’s public Ad Library, show which ads are live. Ads that keep running for weeks usually pay for themselves.
- What could you do differently: a clearer product page, a bundle, faster shipping to a specific country, better photos, or a narrower audience?
If the only way to win is to be the cheapest, think twice. Price wars leave very little margin for ads, returns and mistakes.
3. Work out your real margin, not the sticker margin
The gap between the supplier price and your selling price is not your profit. Your landed cost is everything it takes to get one unit to the customer, and your margin has to cover the cost of finding that customer too.
Example only. These are illustrative numbers to show the method, not real supplier or ad prices:
| Product cost from supplier | $6.00 |
|---|---|
| Shipping to the customer, per unit | $2.50 |
| Duties, taxes and import fees, per unit | $0.80 |
| Packaging and inserts | $0.40 |
| Landed cost | $9.70 |
| Selling price | $24.99 |
| Payment processing fees (about 3% plus a fixed fee) | about $1.05 |
| Advertising cost to win one sale | $8.00 |
| Profit per order | about $6.24 |
On paper this product “makes $15 a sale”. In reality it makes about $6, before returns, refunds and your own time. If ad costs rise to $14 a sale, it makes almost nothing. Run the numbers with your own quotes, and decide in advance the highest ad cost per sale you can accept.
4. Check the supplier before you check out
- Order a sample. Look at quality, packaging and how long delivery really took.
- Compare at least two or three suppliers on price, minimum order, shipping times to your main country and how quickly they answer questions.
- Ask about problems before they happen: what happens with damaged items, delays or stock running out.
- Check any rules for your market. Some products need safety marks, labels or certificates in certain countries, especially electronics, cosmetics and children’s products.
5. Plan around the season
Many products sell in waves: back to school, summer, the winter holidays, gift seasons. If demand peaks in December, you need stock, product pages and ads ready weeks before, not in the middle of it. Look at a product’s search interest across several years to see when its peaks usually start.
6. Test small before you commit
Once a product passes the first five checks, test it with a budget you can afford to lose:
- Build one clear product page with honest photos, sizes, delivery times and returns.
- Set a small, fixed test budget and a limit for ad cost per sale, based on your margin from step 3.
- Run the test long enough to get real numbers, then compare results with your limit.
- Scale only what earns its place. Drop what doesn’t, and keep notes on why.
A quick checklist before you order stock
- Interest is steady or growing over at least 12 months.
- Other sellers have recent reviews, so people are buying now.
- You can name at least one clear reason to buy from you.
- Your profit per order still works after realistic ad costs.
- You have held a sample and checked delivery times.
- You know when the product’s peak season starts.
- Your test budget is money you can afford to lose.
Doing this research in one place
Doing these checks by hand means jumping between many tabs and spreadsheets. We are building NicheTrendFinder to bring them together: trending products and niches, demand signals, competitor signals, supplier research, a landed-cost calculator and a seasonal calendar. It is still in development, and you can explore the demo now. Early access will open in batches.
If you are building an AI-powered product rather than selling physical stock, AI Product Development & AI Entrepreneurship covers how to test whether a problem is worth solving before you build.
Think of one product you were sure would sell that didn’t. Which of these six checks would have warned you first? Share it in the comments so other sellers can learn from it.
