Product research is the step new Amazon sellers most often rush, and it's usually the reason a first attempt at selling doesn't go anywhere. The idea is simple in theory: find something people already want to buy, that you can sell profitably, without walking into a category that's overloaded with strong competition. In practice, it means slowing down and checking a product against a few consistent criteria before committing money to it.
A common beginner mistake is picking a product because it's personally interesting, rather than because there's evidence people are actively searching for and buying it. Look for consistent search volume and steady sales activity over time, not a single viral spike. A product with stable, ongoing demand is usually a safer starting point than one riding a short-term trend.
Every product has competition; the question is whether it's competition you can realistically compete with. Check how many sellers are already listing similar products, how established their listings look (review counts, ratings, how long they've been selling), and whether the top listings are dominated by large, well-resourced brands. A category that's technically "in demand" but locked down by a handful of strong competitors is a harder place to start.
A product can look profitable on the surface and still lose money once you account for Amazon's referral fees, FBA fulfilment and storage fees (if applicable), shipping from your supplier, packaging, and any returns. Before committing, build out a simple cost breakdown per unit and be honest about your actual margin after every fee — not just the difference between your cost price and your selling price.
Beyond demand and competition, think about the physical realities: Is it fragile? Is it heavy or oversized, which affects FBA fees? Does it have quality or safety issues that could lead to returns or complaints? Is it a seasonal product, meaning demand drops hard for part of the year? These practical factors matter as much as the sales numbers.
Product research isn't about finding a guaranteed winner — there isn't one. It's about reducing avoidable risk before you spend money on inventory, so the products you do commit to have a genuine, evidence-based chance of working.