Fake Review Statistics 2026: What 63,533 Product Checks Show

Original data from 63,533 ReviewLens checks on Amazon, Walmart, eBay and Etsy, plus the best outside research on how common fake reviews are.

Key findings

• About 1 in 9 listings showed real warning signs. 11.3% of graded checks scored a C or worse.
• Strong manipulation is rare in what people check. Only 0.6% of graded checks got a D or F.
• Amazon looked cleanest. Since per-store grading began on October 22, 81% of Amazon checks earned an A, against 41% on eBay, 36% on Etsy and 25% on Walmart.
• Etsy had the most "cautious" grades. Over the same period, 28% of Etsy checks scored a C.
• Most checks found nothing alarming. The average check raised 0.7 red flags, and 42% raised none.

Grades across all stores

Where people check

Grades by store

Compare stores with care. Each store exposes different review data: Amazon shows a full star chart and verified-purchase labels, many eBay listings only carry catalog reviews, and on Etsy the unverified-buyer check doesn't apply. The Walmart and Etsy samples are also small. The gap is real, but it isn't a precise ranking.

What outside research says

• 11% to 15% of reviews are likely fake in consumer electronics, home and kitchen, and sports and outdoors, according to a 2023 study for the UK Department for Business and Trade. It modeled 2.1 million reviews across 9 UK e-commerce platforms and found that well-written fake reviews made people more likely to buy.
• Fake reviews are illegal in the US. The FTC's rule banning fake reviews and testimonials, including AI-generated ones, took effect on October 21, 2024, with civil penalties of up to $51,744 per violation at 2024 rates.
• The UK followed in 2025. The Digital Markets, Competition and Consumers Act 2024 made fake reviews a banned practice from April 2025.
• The best-known checker is gone. Fakespot shut down on July 1, 2025. See our Fakespot alternatives.

How we collected this data

A check is one ReviewLens analysis of a product page by a user of the extension. ReviewLens counts checks anonymously: it records the store and the grade, never the product, the page address or anything about the person. The figures come straight from those counts.

• Not a random sample. People choose what they check, and the same product checked by several people counts each time. Popular products are overrepresented.
• Grades measure warning signs, not proof. An A means the reviews didn't show the patterns ReviewLens looks for. It can't catch every fake.
• The method is public. Every check and threshold is on how ReviewLens scores reviews.

How to cite this data

We update this page as more data comes in. Journalists and researchers can ask us for figures by store or by month.