Aggregate Rating
An aggregate rating is the averaged score across every review of a product, and the schema.org property of the same name that lets search engines render that average as a star snippet beside a listing, summarising many ratings into one citable number.
As a metric, the aggregate rating collapses all of a product page's reviews into a single figure, usually shown as a value out of five next to a review count. As schema.org markup, the aggregateRating type carries that figure in a machine-readable form, and it is the piece search engines read when deciding whether to draw rating stars under a result. The average is only half the signal: the count beside it tells a reader how much weight the average deserves. A flat 4.9 from six reviews and a 4.6 from eight hundred are not the same claim, and shoppers read them differently even when the headline number looks higher.
Google requires the markup to include a ratingValue and either a reviewCount or ratingCount, and it enforces a strict rule: the count and value in the structured data must match the reviews actually visible on the page. Marking up a rating that a shopper cannot see, or inflating the count, is against policy and can cost you rich results entirely, so the safe path is to generate the markup from the same reviews you display.
Consider a Shopify store selling a single hero product, a merino base layer. The product page shows 214 published reviews averaging 4.7 stars. The aggregateRating block should report exactly ratingValue 4.7 and reviewCount 214, recalculated whenever a new review is approved or an old one is hidden. If the merchant later splits the listing into three colour variants, the decision is deliberate: either keep one shared aggregate across all variants, or markup each variant separately with only its own reviews. Mixing the two, for instance showing the full 214 on a variant page that only displays nine reviews, is the exact mismatch that drops the snippet.
The term matters for AI search because answer engines treat the aggregate as a compact, quotable fact. When ChatGPT, Perplexity, or a Google AI Overview is asked to compare base layers, a clean ratingValue and reviewCount give the model something precise to cite, and the count signals how trustworthy that average is. A page with reviews rendered only as a JavaScript widget, with no corresponding structured data, often gives these systems nothing to read at all, so the product is summarised without its own evidence.
Getting existing reviews readable, corroborated, and cited by search and AI is the gap BeyondReviews closes, and a correct aggregateRating is the mechanism that turns honest review counts into the star snippets and answer-engine citations that earn the click.