Your Google Star Rating Is Not a Simple Average

The number on your Google Business Profile looks like arithmetic. Add more five-star reviews, the score goes up. Get a one-star, it drops. The logic seems simple enough.

Then you watch it not behave that way. A business collects a burst of genuine positive reviews and the rating moves less than expected. An owner gets a fake one-star removed after a successful flagging campaign and sees almost no change to the displayed number. A competitor climbs from 4.1 to 4.6 in what appears to be far fewer new reviews than the math should require.

The gap between what the arithmetic suggests should happen and what actually appears on the profile is where your strategy lives. If you understand why the displayed rating behaves differently from a simple mean, you stop chasing the wrong levers and start pulling the ones that actually matter.

Your Rating Is a Weighted Number, Not an Arithmetic Mean

Google's support documentation states that your rating is calculated from all your reviews. That statement is accurate, but it leaves out most of the complexity.

The first complication is filtering. Google's spam detection removes reviews it considers inauthentic - sometimes immediately, sometimes days or weeks after they appear. When a review gets filtered out, it may have already been visible on your profile briefly, but it is no longer contributing to the displayed score. The count of reviews you can see and the count of reviews Google is using in the calculation are often not the same number.

The second complication is weighting. The displayed number is not the same as the mathematical average of the visible stars on your profile. If you have ever added up all the star values from your visible reviews and divided by the count, you have probably noticed the result does not precisely match what Google displays. The difference is usually small - a few hundredths of a point - but it is consistent and persistent.

Google has not published the exact formula. What the observable behavior shows is that more recent reviews carry more influence over the displayed score than older ones do. This is intentional design. A profile with two hundred five-star reviews from three years ago and eight one-star reviews from last month is sending a different signal than its raw arithmetic average implies. The weighting makes the score more sensitive to what is happening now, which is what prospective customers actually care about.

The practical consequence: your displayed rating is not a permanent record of your all-time performance. It is a rolling signal that responds to recency as much as to volume.

Why Your Star Average Has an Expiration Date

Owners who treat their review profile as "established" once they reach a comfortable score tend to discover this the hard way.

A business that collected eighty reviews in its first two years and then stopped asking - because the owner felt the profile was in good shape - will see its protective cushion thin out over time. Not dramatically, and not month to month. But the reviews collected three or four years ago carry progressively less algorithmic weight. If you stop collecting new five-star reviews, your score becomes more exposed to the impact of any new negative reviews that arrive, because the reviews that used to absorb the blow are no longer pulling as much weight.

Two businesses can sit at identical displayed ratings today and be in very different positions. One is collecting three to five new five-star reviews per month on top of a healthy base. The other is coasting on a stack of older reviews with nothing recent coming in. Six months from now, after one difficult week of two or three unhappy customers, those profiles are going to diverge. The active collector absorbs the hit. The coaster takes a visible knock.

The score you see today reflects your review history. The score you see in six months reflects your review cadence right now.

The Recovery Math: How Many 5-Star Reviews It Takes to Move a Damaged Score

This section is where the arithmetic becomes useful, even if Google's actual algorithm departs from a simple average. The math is directionally correct and it gives you a realistic sense of what recovery actually requires.

Suppose your profile has sixty reviews and sits at 4.0 stars. A simple average means those sixty reviews sum to 240 total star points (60 x 4.0). To reach a displayed score of 4.3, you need the new average to be 4.3.

Adding new five-star reviews: each adds 5 to the numerator and 1 to the denominator. Let n equal the number of five-star reviews needed:

(240 + 5n) / (60 + n) = 4.3

240 + 5n = 258 + 4.3n

0.7n = 18

n = approximately 26 reviews

Twenty-six new five-star reviews to move from 4.0 to 4.3 - when you already have sixty. That surprises most people. It surprises them more when they run the same calculation with a larger starting base. At 200 reviews and a 4.0 average, the same 0.3-point improvement requires over eighty new five-star reviews.

The higher your review count, the harder it is to move the number in either direction. This is what makes established profiles stable - it also makes recovery from a bad stretch a sustained effort rather than a quick campaign. You cannot outrun a run of bad reviews with a single ask to your contacts. You need consistent collection over weeks and months.

One note of optimism: because Google weights recent reviews more heavily than old ones, the real-world number required to move your score may be somewhat lower than this simple arithmetic suggests. But "somewhat lower" does not change the fundamental shape of the problem. Recovery is a marathon, not a sprint.

What Does Not Move Your Rating (Despite What You May Have Heard)

There is a short list of actions that business owners invest time in, believing they will shift the displayed score, when they do not.

Responding to reviews. Owner responses are visible to future customers and they are a real signal to Google's local ranking algorithm - but they do not change the star rating. A thoughtful reply to a one-star review does not alter the star contribution of that review. The star is still a star. Responding is worth doing for other reasons, including how it reads to prospective customers and what it signals about your responsiveness. Just do not expect it to move the number.

Flagging and removing a single bad review. When Google agrees a review violates its policies and removes it, the rating adjusts to reflect the removal. But the adjustment is almost always small, because the removed review was one data point in a pool. A single one-star removed from a profile with eighty reviews moves the average by a fraction of a point - often less than a tenth. Business owners who spend weeks trying to get one bad review flagged are usually disappointed by how little the number changes when they succeed. The effort is not always wasted - a genuinely fake or policy-violating review should be reported - but it is not a score recovery strategy.

Asking Google to adjust your score manually. There is no mechanism for this. Google does not accept requests from business owners to override the displayed rating.

The sentiment of the text in your reviews. Your displayed star rating is based on the star values, not on whether the written text is positive or negative. A review that says "they fixed what three other plumbers couldn't solve" and gives five stars contributes five stars to your average, regardless of how the language reads. The text in reviews influences local search relevance for specific keywords, but it does not change the displayed number.

The Velocity Problem: When Collecting Too Fast Works Against You

Google's spam filter is triggered not just by the content of individual reviews but by collection patterns. A profile that jumps from four reviews to forty reviews in two days looks different from one that collects two or three reviews per week over several months - even if every one of those forty reviews is completely genuine.

When a spike of reviews arrives in a short window, Google's detection becomes more aggressive. Reviews start disappearing from the visible count. Some never appear at all. From the business owner's perspective, customers report leaving reviews that never show up. The count barely moves despite real activity.

This is one of the strongest arguments against batch review campaigns - asking your full contact list at once, or running a social media push that drives a sudden surge of review activity. The intent is legitimate, but the pattern looks like manipulation to Google's filter, regardless of whether the reviews themselves are real.

A consistent cadence - sending review requests to customers as they complete their experience, week after week - produces a growth pattern that avoids triggering the filter. If you are using a tool that automatically sends requests after individual transactions rather than batch-blasting your contact list, this problem largely takes care of itself. QuickFeedback, for instance, sends each request tied to a completed job or service rather than in groups, which keeps the velocity pattern looking natural to Google's systems.

The goal is a review profile that grows at a steady, sustainable rate rather than in spikes that plateau. Steady growth compounds over time. Spike-and-plateau growth invites filtering that erases some of what you worked for.

The Display Threshold: When Your Profile Shows No Rating at All

New profiles and profiles with very few reviews do not display a star rating. Google requires a minimum number of reviews before surfacing a score. The exact threshold is not published, but observations from the local SEO community consistently point to around five reviews as the minimum for a rating to appear.

Until a profile clears that threshold, it shows no stars at all - which is a worse signal to most customers than even a modest score. A blank rating reads as unverified, inactive, or new. Customers scanning search results instinctively skip profiles without a visible rating.

This is the one situation where a short-term concentrated push makes complete sense. Getting from zero to five genuine reviews as quickly as possible - by reaching out directly to your most recent satisfied customers and asking them individually - is the right move. Any displayed score is better than none. This is not a contradiction of the velocity advice above: the filter is far less likely to strip reviews from a profile with no history, and the five-review window is small enough that timing risk is minimal.

Once you are past the threshold and have a score showing, shift to the consistent cadence approach and do not look back.

The Metric That Matters More Than the Decimal

The displayed star average is the most visible signal on your profile, but it is a lagging indicator. It tells you what your collection history has been, not what your collection rate is right now.

A more useful number to track is review velocity: how many new reviews you are receiving per month. This tells you whether your collection system is functioning, whether your timing and channel are working, and whether customers are willing to review you when asked. It gives you early warning when something in the process breaks - before that break shows up in the displayed average.

A business at 4.2 stars collecting eight new reviews per month is in a structurally stronger position than a business at 4.4 that has not collected a new review in three months. The first business's rating is exposed to new negatives, but it has the momentum to absorb them. The second business's 4.4 is fragile - one bad stretch of two or three one-stars, with no recent positives to offset them, can push the displayed number in a direction that takes months of consistent collection to recover from.

Set a monthly velocity target that fits your business size. For a single-location service business, one new review per week is achievable and provides meaningful cushion. Treat a month below that target as a process signal - something to investigate and fix - rather than just watching the displayed average and hoping it stays where you want it.

The average is what customers see. The velocity is what keeps it healthy. Track both, but invest your energy in the one you can actually control.