Amazon Review Analyzer

On Amazon, the Verified Purchase badge means Amazon's records confirm the reviewer bought the item through the platform. That raises the evidence quality of a review — the person at least held the product — but it is not a guarantee of honesty. Sellers have been known to refund buyers off-platform in exchange for positive reviews, badge and all.

Reviews without the badge are not automatically fake either; long-time customers sometimes review products bought elsewhere, and Vine reviewers carry a program disclosure instead. Treat the badge as one signal among many: useful in aggregate, weak as a single-review verdict.

Amazon pools reviews across every variant of a product — every color, size, and bundle shares one rating. That 4.8-star average might belong mostly to the blue medium, while the red large you are eyeing has a dozen one-star reviews about faulty zippers. Always filter reviews to your exact variant before trusting the number.

The scan looks for variant-specific complaint language, but double-check it yourself: open the review filters, select your variant, and read what remains. If the variant filter empties the review list, you are buying on the reputation of a different product entirely.

Review velocity — how fast reviews accumulate — is one of the most honest signals on a listing. A healthy product gathers reviews steadily over months. A sudden burst of dozens of five-star reviews in a single week, especially on an older product, often traces back to a discount-for-review campaign or something less innocent.

A brand-new product with hundreds of glowing reviews deserves the same skepticism: real adoption takes time. None of this proves fakery on its own — launches and viral moments create genuine spikes too — but unusual velocity is a reason to weight the critical reviews more heavily in your decision.

For Amazon listings, five captures cover nearly everything the scan needs: the rating histogram and total review count, the product information block with the seller name and current price, the top critical reviews, the top positive reviews, and any answered questions that mention defects or sizing.

Include the variant selector in at least one capture so the analysis matches the version you would buy. Skip the 'frequently bought together' and sponsored sections — they add noise. If reviews span multiple pages, capture page one and the most critical page; the scan handles the rest.

Run the scan, then work through five checks. First, the verdict: Buy, Maybe, or Avoid. Second, the biggest complaint — does it affect your use case? Third, the variant filter — do the reviews match what you are buying? Fourth, the fake-review risk — is the rating itself trustworthy?

Fifth, the value question — would you still be happy at this price if the rosiest reviews turned out to be exaggerated? If the answer is yes across all five, buy with confidence. If two or more raise doubts, keep shopping. Five minutes now beats a return label later.