Most product ideas look good at the start. You can already picture the listing, the supplier price seems low and a few competitors appear to be selling. The expensive part comes later, when stock sits in a warehouse because demand was thinner than it looked, or the margin disappears once fees are paid.
Validating demand before you list doesn't guarantee a winner, but it filters out most of the losers. Here's the process we follow when we research products for sellers.
Start with a clear shortlist
Research goes faster when you know what you're looking for. Write down your criteria before you look at a single product, so you judge ideas against them instead of against your enthusiasm.
- A price range that leaves room for fees, shipping and advertising.
- A size and weight that are cheap to store and ship.
- No complicated compliance requirements in the markets you sell to.
- A product you can describe and photograph well.
- Room to improve on what's already listed: better images, bundles or clearer information.
Check demand from several angles
No single number tells you whether a product will sell. Look for signals that agree with each other.
Marketplace search
Start typing the product into each marketplace's search bar and note the suggestions. Terms that appear in autocomplete are things people actually search for. Then look at how many results come up and how closely they match your product.
Sales signals
On eBay, filter search results to sold items to see what has actually sold and at what price; Terapeak, in Seller Hub, goes deeper. On Amazon, Best Sellers Rank shows how a product sells compared with others in its category. Treat it as a comparison, not a sales figure, and watch it over several weeks instead of trusting one snapshot. Steadily growing review counts are another sign that units are moving.
Seasonality and trends
Check whether interest is steady, seasonal or a short spike. Search trend tools show the shape of demand over time. A product that sells mainly in December can still work, but you'll need to plan stock and cash flow around it.
Size up the competition
Demand with no competition is rare, and usually means something is wrong. What you want is demand you can realistically compete for. Look at the first page of results and ask:
- Are the top listings from big brands, or from sellers like you?
- How many reviews do the leaders have, and what do the negative ones complain about?
- Are the listings weak in ways you can fix: poor images, missing information, unclear sizing?
- Is everyone competing on price alone?
Do the margin math before you commit
A product with strong demand can still lose money. Build a simple per-unit calculation before you order anything:
- Product cost, plus shipping to your warehouse or fulfillment center, duties and packaging.
- Marketplace fees: referral or final value fees and any per-order fees.
- Fulfillment costs, whether you ship orders yourself or use a fulfillment service.
- Advertising, especially while a new listing builds its sales history.
- Returns and damaged units.
Run the numbers for each marketplace separately, because fee structures and fulfillment options differ. If the numbers only work at a price above what competitors charge, you need a clear reason why buyers would pay more: a better bundle, higher quality or information the other listings don't give.
Check the supply side too
Demand is only half the picture. Before you commit, make sure you can get the product reliably: order samples, ask about lead times and minimum order quantities, and find out how consistent quality is from one batch to the next. A product that sells well but arrives late or varies in quality will cost you in returns and reviews.
Rule out the deal breakers
Some products aren't worth the risk. Check whether the category is restricted or needs approval on your marketplaces, whether the design or brand is protected by someone else's intellectual property, and whether the item counts as hazardous or needs certification. Fragile or oversized products aren't impossible, but they raise costs and return rates.
Test small, then scale
Order a small first batch, or list a limited quantity if you dropship, and give it enough time to collect real data. Track sessions, conversion, returns and buyer questions. The questions are especially useful: they show what your listing doesn't explain yet. Scale up only when the numbers hold.
Good product research is mostly discipline: clear criteria, several signals and honest math. If you'd like help finding and checking products for your stores, see our Amazon Management service.
