Authenticity signals
Reviewer authenticity signals — patterns in reviewer activity history
Review text similarity — detection of copy-pasted templates
Time-series anomalies — unnatural concentrated bursts of reviews
Mismatch between review content and the product — generic praise unrelated to the actual product
How filtering is applied
Review data exposed to AI Chat and Dashboard is cross-referenced against these authenticity signals to improve reliability.
When suspicious patterns are detected, the affected reviews are down-weighted in analysis.
Consumer sentiment results reflect the cleaned signal.
Perfect filtering is not possible
No platform can guarantee 100% removal of fake reviews. Trendier provides trend-level reliability.
We recommend interpreting reviews based on patterns and trends across many reviews rather than relying on a single review.
Common follow-up questions
Can I see the exact fake review ratio for a specific product?
Exact percentages are not provided, but you can identify suspicious patterns by cross-referencing review velocity anomalies with sentiment distribution.
How can I dig deeper into review data?
On Dashboard (Enterprise), you can explore SKU-level review data and consumer profiles (skin type, skin concerns, etc.) directly.
Related articles
Where does the data come from?
What is Trendier Dashboard?
How does Trendier identify trends?