Product selection is where most stores are won or lost, and it happens before a single unit is bought. The screen below is not clever. It is just applied consistently, which turns out to be the harder part.
Demand comes first, and it has to be boring
We look for categories with steady twelve-month demand rather than a spike. A product that sells four hundred units a month every month is worth more to a new store than one that sold six thousand units in November and nothing since. Seasonality is not disqualifying, but it changes how much inventory capital gets committed and when.
Margin is calculated after every cost, not before
The number that matters is what remains after the referral fee, fulfilment, storage, inbound shipping, returns provision and advertising. A product showing forty percent gross margin on a supplier quote routinely lands at eleven percent once those are applied. If it does not clear our floor at that stage, it is out, regardless of how good the top-line number looked.
Competition is read as structure, not count
Twenty competing sellers is not worse than five. What matters is whether the category is controlled. If two sellers hold ninety percent of the reviews and both are the brand owner, the category is closed no matter how thin the listing count looks. If demand is spread across thirty mid-sized sellers, there is usually room.
Supply risk is the one people skip
A single supplier with no second source is a store with an expiry date. We want a qualified alternate before launch, agreed lead times in writing, and a sample from each. This is the least interesting part of the process and the one that most often prevents a dead quarter.
What survives
Roughly one in forty candidates clears all four. That ratio is the point of the exercise. The screen exists to say no cheaply, so that capital is only committed to the small number of products that survive being looked at properly.
