Find the average rating of a five-star system from the number of votes at each star level. Enter how many 5-star, 4-star, 3-star, 2-star, and 1-star ratings you received — the calculator returns the weighted average rating, the total number of votes, and a percentage breakdown.
The average rating calculator works out the overall star rating of a product, service, business, or piece of content from the number of votes it received at each star level. Instead of a simple average of scores, it computes a weighted mean: each star value counts once for every vote it received. Type in how many 5-star, 4-star, 3-star, 2-star, and 1-star ratings you have, and the calculator instantly returns the average rating out of five, the total number of votes, and a percentage breakdown showing how the votes are distributed.
This is exactly how the star ratings you see on shopping sites, app stores, review platforms, and restaurant listings are calculated. Whether you run an online store, manage a business profile, teach a class using rubrics, or are simply curious how a headline rating is put together, this tool does the math for you and shows every step. Below you will find the formula, worked examples you can reproduce, guidance on interpreting the result, and answers to the most common questions about average ratings.
A five-star average is a weighted average, where the "weight" of each star level is the number of votes it received. The formula is:
rating = (5×r₅ + 4×r₄ + 3×r₃ + 2×r₂ + 1×r₁) ÷ (r₅ + r₄ + r₃ + r₂ + r₁)
Here r₅ is the number of 5-star votes, r₄ the number of 4-star votes, and so on down to r₁, the number of 1-star votes. In words: multiply each star value by how many people gave that rating, add all those products together, and divide by the total number of votes. The result is a number between 1 and 5 (or 0 if there are no votes) that represents the typical rating a new visitor would expect.
The denominator — the sum of all the votes — is the total number of ratings, which the calculator also reports. The total matters just as much as the average: an average of 4.8 from 5,000 votes is far more trustworthy than 5.0 from a single vote.
Imagine a product with the following reviews: 120 five-star, 60 four-star, 25 three-star, 10 two-star, and 15 one-star. First, multiply and add:
(5×120) + (4×60) + (3×25) + (2×10) + (1×15) = 600 + 240 + 75 + 20 + 15 = 950
Then divide by the total number of votes:
total votes = 120 + 60 + 25 + 10 + 15 = 230 → rating = 950 ÷ 230 ≈ 4.13
So the average rating is about 4.13 out of 5 from 230 votes. These are the calculator's default values, so you can press Calculate to see the full breakdown and then replace them with your own numbers. Notice how the large block of 5-star and 4-star reviews pulls the average up, while the 15 one-star reviews drag it down only slightly because they are outnumbered.
A common mistake is to average the five star values (5, 4, 3, 2, 1) and get 3, ignoring how many people chose each. That would only be correct if exactly the same number of people voted at every level. In reality, votes are almost never evenly spread — most products collect far more high ratings than low ones, or vice versa. The weighted average respects this by counting every individual vote. Each person who leaves a 5-star review adds a "5" to the pile; each 1-star reviewer adds a "1." Dividing the grand total by the number of voters gives the true center of gravity of all the opinions, which is why every major rating platform uses this method.
Alongside the average, this calculator shows what percentage of all votes each star level represents. This distribution tells a story the single average number cannot. Two products can both average 4.0 stars yet feel completely different: one might have almost every review at 4 stars (consistent, predictable quality), while the other might be split between many 5-star and many 1-star reviews (polarizing quality that delights some and disappoints others). Looking at the percentage of 1-star and 2-star reviews is especially useful, because a cluster of low ratings often points to a specific, fixable problem — a shipping issue, a confusing feature, or a mismatch between expectations and reality.
Context matters, but a few rules of thumb apply across most platforms:
Interestingly, research on consumer behavior suggests that a perfect 5.0 can actually seem less trustworthy than a 4.7 or 4.8, because a handful of critical reviews makes the positive ones feel genuine. A high-but-not-perfect score backed by a large number of votes is often the most persuasive combination.
Most of the ratings you encounter every day are weighted averages just like the one this calculator produces, though some platforms add their own refinements. Online marketplaces and app stores typically display the plain weighted mean of all reviews, sometimes rounded to one decimal place and shown as filled, half, and empty stars. Some sites weight recent reviews more heavily so the score reflects current quality, or give more influence to verified purchases to reduce the impact of fake reviews. Others, like certain movie and restaurant platforms, blend user ratings with critic scores. Despite these tweaks, the foundation is always the same weighted-average calculation: sum the products of value and count, then divide by the total number of votes.
Average ratings are usually shown to one or two decimal places, and often visualized with stars. A rating of 4.13, for instance, might appear as four filled stars with a small fraction of a fifth. Be careful when comparing rounded scores: two products both displayed as "4.5 stars" could actually be 4.45 and 4.54, which round to the same figure but represent meaningfully different feedback. That is why this calculator also reports the total number of votes and the full distribution — the extra detail helps you compare ratings fairly rather than relying on a single rounded number.
The weighted-average method is useful anywhere people score things on a fixed scale. Teachers can average student feedback on a lesson, event organizers can summarize satisfaction surveys, HR teams can consolidate performance-review scores, and researchers can condense Likert-scale questionnaire responses into a single figure. Any time you have a count of responses at each point on a scale — not just one to five, but any range — the same logic applies: multiply each scale point by its count, sum the products, and divide by the total responses. This calculator is set up for the familiar five-star scale, but the principle carries over to any rating system.
It helps to see the weighted average built up one piece at a time, because the logic is the same no matter how many votes you have. Start by writing down your five counts — the number of 5-star, 4-star, 3-star, 2-star, and 1-star votes. The first job is to work out the total value contributed by each star level, which is simply the star number multiplied by how many people chose it. A level with 40 four-star votes contributes 4 × 40 = 160 to the running total, while a level with 3 one-star votes contributes only 1 × 3 = 3. Add up all five of these products and you have the numerator of the formula — the grand total of every point every voter awarded.
The second job is to add up the five counts to get the total number of votes, which becomes the denominator. Dividing the first number by the second gives the average. Because the calculation is just multiplication, addition, and one division, you can reproduce it by hand or in a spreadsheet, and you will get exactly the same figure the calculator shows. Seeing it laid out this way also makes it obvious why levels with more votes have more influence: a star value only counts as many times as there are people who chose it, so the crowd’s overall opinion — not any single review — determines where the average lands.
The most frequent error is confusing the number of votes with the star value. The boxes in this calculator ask for how many people gave each rating, not the rating itself — so if forty people left five stars, you type 40 in the 5-star box, not 5. Another mistake is forgetting a level that has votes, or double-counting reviews that appear in more than one place. It also pays to make sure you are averaging comparable things: mixing ratings from different time periods, product versions, or platforms can produce a number that does not really describe anything. Finally, remember that rounding hides detail. If you only ever look at the one-decimal figure, you may miss the difference between a score that is climbing and one that is slipping. Checking the total votes and the full percentage breakdown alongside the average keeps you from drawing the wrong conclusion from a single tidy-looking number.
Multiply each star value (1 to 5) by the number of votes it received, add the products together, and divide by the total number of votes. For example, (5×120 + 4×60 + 3×25 + 2×10 + 1×15) ÷ 230 ≈ 4.13.
It is a weighted average. A simple average of the values 1–5 would ignore how many people voted at each level. The weighted average counts every individual vote, which is why platforms use it.
Generally, 4.5+ is excellent, 4.0–4.5 is very good, 3.5–4.0 is decent, and below 3.0 usually needs attention. A strong score backed by many votes is more convincing than a perfect score from just a few.
Yes. An average is only as reliable as the number of votes behind it. A 5.0 from two reviews says far less than a 4.6 from two thousand, which is why this calculator also shows your total votes.
The tool is built for the five-star scale, but the same weighted-average method works for any rating scale: multiply each scale point by its count, sum, and divide by the total number of responses.
It depends on your total votes. With few reviews, a single 1-star rating can drop the average noticeably; with thousands of reviews, the effect is tiny. Volume protects your score over time.
This Average Rating Calculator is provided for general informational and educational purposes. It computes the standard weighted average of a five-star rating system. Individual platforms may apply additional weighting or filtering to their displayed scores, so results here may differ slightly from a specific site's published rating.