How we score

What the score is

The Ratelens Score is a single figure from 0 to 100 that summarises three things we can observe about a product across Amazon, Flipkart and Nykaa: how buyers rate it, how genuine its discount is, and how consistent the platforms are about it. It is shown on every product with at least one buyer rating, and the three parts are always displayed so you can see how the number was reached.

What the score is not

It is not a test of how well the product performs. We do not own the products, do not run them in a lab, and make no claim about motor power, build quality or durability. The score reads the market's signals (ratings, prices and declared MRPs), not the product itself. Treat it as a shortcut through the listings, not a verdict on the hardware.

The three parts

Buyer rating: 45%

Each store's star rating, weighted by how many people rated there, so a 4.1 from three buyers cannot outweigh a 3.8 from three thousand. Crucially, a high rating from very few ratings is pulled back toward the middle: a 4.6 from eight buyers is not strong evidence and is not scored as though it were. A rating is trusted in full only once around a thousand people have left one.

Genuine value: 35%

The real discount off a trusted MRP. Where the platforms agree on the MRP we use that consensus figure; where they disagree we use the lowest declared value, never the higher one. This is deliberate: a higher MRP makes the same price look like a bigger discount, so the lowest figure is the safest to measure from. A listing whose only "discount" comes from an MRP no one else agrees with earns little here.

Transparency: 20%

Whether the product is verified on more than one platform and whether those platforms tell a consistent story. A product carried on all three stores with agreeing MRPs scores well; one where two platforms declare contradictory MRPs is marked down, because a shopper cannot rely on what the discount actually is.

Bands

80 and above is Excellent, 65 to 79 Good, 50 to 64 Average, and below 50 Below average. A component with no data (a product with no reliable MRP, for instance) is left out and the remaining parts are reweighted, rather than the product being punished for a gap in the data.

Why we publish the method

A score you cannot check is just an opinion with a number attached. Ours is built only from figures already shown on the page, so you can follow it yourself. If a score looks wrong, tell us at hello@ratelens.in.