We published the bar instead of hiding it.

Nine out of ten products fail. It is the most repeated number in e-commerce, and it is not true. What is true is worse, and more useful.

Y-Score is not a vibe. It is 82 factors across nine dimensions, condensed to a 23-row shortlist with published thresholds — and a refusal to average away the things that kill deals.

Four numbers decide whether a product can survive being advertised. Here are the thresholds we use, with the actual figures.

Good margins on a product nobody urgently wants is a slow, well-funded failure. The demand block asks five questions, and one of them outranks the rest.

Play the ad with the sound off for three seconds. If the benefit has not landed, no budget will save it.

The order converted. Now shipping, returns, the supplier and the producer’s own web shop take their share. This is where good products quietly become bad businesses.

Most scoring systems let a fatal problem get averaged into an acceptable number. Six things end the evaluation instead, and none of them produce a score at all.

An application that says the product is loved scores lower than one that shows the reorder rate. That is deliberate, it is enforced in code, and it is why we ask for almost no documents.