Calculate weighted average product rating from star counts

Enter the count recorded at each star level from one through five. The calculator multiplies every count by its star value, combines those weighted scores, and divides by the complete review total. It also adds every bucket to report the exact number of reviews represented. The calculation is deterministic, requires no product identifiers, and rejects an empty distribution rather than presenting a misleading zero rating.

Enter the complete rating distribution

Start with the five counts shown by your store, marketplace export, review provider, or reporting database. Put the number of one-star reviews in the one-star field, then do the same for two, three, four, and five stars. Counts must be whole numbers of zero or more because each value represents actual reviews, not a percentage or an already calculated score. Include buckets with no reviews as zero; do not omit them. The calculation works equally well for a new item with one review and a mature catalog listing with millions, provided every count remains within the safe integer range. Use figures from the same product, variant scope, channel, and reporting moment. Mixing a lifetime five-star count with a monthly one-star count creates a mathematically valid result that describes no real population. If a source supplies only percentages, obtain the underlying counts first whenever possible, since rounded percentages may not reconstruct the true review total. Once the five compatible counts are ready, submit them together as one distribution.

Understand the weighted calculation

A simple average of the five bucket counts would answer the wrong question. The calculator instead gives each bucket its star value: one-star reviews contribute one point each, two-star reviews contribute two points each, and so on through five points for every five-star review. It adds those contributions to obtain a weighted score, then divides that score by the total number of reviews across all five buckets. For example, ten five-star reviews influence the result five times as much as ten one-star reviews because they contribute fifty rating points instead of ten. The returned average is rounded to no more than six decimal places, which keeps output stable while preserving far more detail than a typical storefront display. The total review count is returned separately and is never inferred from the average. If all five counts are zero, there is no population and therefore no defined weighted average. The capability reports invalid input in that case instead of manufacturing a zero-star result that could be mistaken for genuine customer sentiment.

Use the result consistently

The average rating is suitable for catalog quality checks, internal dashboards, product-feed enrichment, marketplace reconciliation, and display logic. Decide separately how your storefront presents it: a system may show one decimal place, two decimal places, or a graphical star fill, while retaining the more precise returned value for comparisons. Always keep the total review count beside the average because confidence and commercial meaning differ sharply between a rating based on two reviews and the same rating based on twenty thousand. When comparing channels, calculate each channel from its own five buckets before combining data. To obtain an overall rating across channels, sum corresponding star counts first and run the combined distribution; averaging channel averages directly can overweight a channel with very few reviews. In automated pipelines, preserve the source timestamp and product key alongside the result even though they are not inputs to this calculation. Browser use is free, while an API request costs $0.002. The same deterministic arithmetic supports both paths, so repeated submissions of the same counts return the same figures.