Whitespark's Local Search Ranking Factors survey puts review signals at roughly 20% of what decides local pack placement, up from about 16% in 2023. Steady movement in one direction over several years, in the one part of search where a small business can realistically compete.
The practical consequence is not that reviews became more persuasive. It is that reviews became infrastructure. A business with the wrong opening hours and a business with no recent reviews now fail in a similar way: both are quietly ranked below a competitor who got the basics right.
"Review signals" is more than the star average
The phrase covers a bundle, and the bundle is what matters, because teams tend to optimise the one component they can see on their dashboard and ignore the rest.
- Count. How many reviews the profile has, relative to nearby competitors in the same category.
- Velocity. Whether they are still arriving. A profile that stopped collecting looks stale regardless of its total.
- Recency. How old the most recent ones are.
- Diversity. Whether reviews also exist off Google, on the directories and platforms that matter in your category.
- Response rate. Whether the owner replies, which is both a signal and the thing readers check.
- Keyword content. What customers actually name in their text, which is how a profile ends up associated with a service it never listed.
A business with 300 reviews, none in the last year, no replies, and nothing on any other platform is weaker on this bundle than one with 80 that arrive steadily and get answered. The total is the least interesting number in the set.
Reviews and listings are one problem
The mistake worth avoiding is treating review management as a separate project from listing accuracy. Both feed the same ranking, and one undermines the other when they drift apart.
Duplicate profiles split a location's reviews across two entries, so a business with 200 reviews presents as two businesses with 100 each. A category set wrong means the profile competes for the wrong queries, and reviews mentioning what you actually do cannot correct it. Hours that are wrong generate the specific complaint that damages both rank and rating at once, because a customer who drove to a closed shop writes about exactly that.
What to do, in order
Audit the profiles first. Before any review work, confirm one profile per location, correct primary category, correct hours including holidays, and no duplicates. This is unglamorous and it is the highest-leverage thing on the list, because everything downstream inherits the errors.
Restart velocity. If reviews stopped arriving, that is the gap to close before worrying about the average. A steady small flow beats a burst, and a burst risks the platform's filtering removing the lot.
Reply to everything, quickly. Response rate is cheap to fix and it is one of the few signals fully within your control. It also compounds: the reply is what the next reader sees.
Read what customers name. If reviews for your Bengaluru outlet keep mentioning a service that is not in your categories, that is free information about how you are actually being found, and usually a category worth adding.
Keeping expectations honest
Two caveats, because 20% gets quoted as though it were a dial.
Proximity still dominates. Someone searching two kilometres away will see businesses near them regardless of how good your profile is. Review signals decide placement among the businesses already in contention, which is real but bounded.
And the figure comes from a practitioner survey, not from Google. It reflects what experienced local SEOs observe, which is the best available evidence and is not the same as a published weight. Treat it as a direction of travel rather than a coefficient.
The direction has been consistent for several years, and it points the same way each time: the profile you keep current beats the profile you set up once.