Local SEO Ranking Factors: What Is Actually Proven

The statistic that 46 percent of all Google searches are local comes from a tweet. Not a study, not a Google announcement. One person reporting a remark made at an event in October 2018, written up by a journalist who said at the time that the number seemed too high.

Local SEO advice is built on a stack of numbers like this. Here is what is actually documented, what is expert opinion, what has been measured, and what dissolves on contact.

What does Google actually document?

One page, and it is short. Google states that local results are mainly based on relevance, distance and popularity.

That wording matters, because it changed. The sentence the entire industry quotes is relevance, distance and prominence. Google's current page says popularity in the summary sentence while retaining Prominence as a subheading. The famous line about a business farther away potentially outranking a closer one has also been deleted. Thousands of articles are still quoting text that is no longer on the page.

The three definitions, in Google's words:

• Relevance is how well a profile matches what someone is searching for, improved by providing complete and detailed business information.
• Distance is how far each business is from the searcher, using what Google knows about their location if they have not shared it.
• Prominence is how well known a business is, based partly on how many websites link to it and how many reviews it has. Google states directly that more reviews and positive ratings can help local ranking.

Google also states that there is no way to request or pay for a better local ranking, and that it keeps algorithm details confidential deliberately. It publishes no weightings of any kind. Anyone giving you a percentage is not quoting Google.

What do the experts think?

The main source in this field is Whitespark's Local Search Ranking Factors, published on 6 November 2025 by Darren Shaw. It surveyed 47 experts across 187 factors, each scored from 0 to 5.

Whitespark's own disclaimer is the most important sentence in it: none of the experts have special access to the internal workings of Google's local search algorithm. Participants were asked to estimate, and the results are meant to give direction rather than measurement.

The top five factors for the local pack, by score:

• Primary category on the Google Business Profile.
• Proximity of the address to the point of search.
• Keywords in the business title.
• Physical address in the city being searched.
• Business open at the time of search.

Reviews appear next: high ratings at sixth, quantity of reviews with text at ninth, recency at eleventh. At the other end, keywords in owner responses ranks 182nd of 187, and title tag length ranks 184th, which the report labels an SEO myth outright.

Two warnings about how this list is used elsewhere. These are integer scores, not percentages, and the percentage weightings circulating in blog posts are invented. Two articles quoting the same report claim GBP signals at 32 percent and proximity at roughly 55 percent simultaneously, which cannot both be true.

Second, Shaw publicly disagrees with his own panel on the seventh ranked factor, writing that he remains unconvinced that showing versus hiding your address is a ranking factor and suspects it is a bug. When the person running the survey dissents from its results, treat the list as informed opinion rather than findings.

One of those factors is worth acting on regardless, because it is also documented by Google: your primary category. That is set in the profile and it is free to get right.

What has actually been measured?

Very little, and the honest samples are small.

Reviews and ranking

Sterling Sky ran a controlled test on whether review count moves rankings, updated in February 2026. Three businesses sitting at nine reviews saw a small Maps improvement on reaching ten. Three businesses going from ten to eleven saw nothing. An earlier single listing test found a gain going from three to sixteen reviews and no further gain from sixteen to thirty one.

That is a real experiment with real controls and a sample of three. It is more evidence than almost anything else in this field, and it is still three. Joy Hawkins has been open about why: you cannot properly track clicks from the local pack, which makes measurement genuinely hard.

AI results versus the local pack

The best methodology found in this subject comes from a June 2026 study running 1,120 query executions across seven verticals and thirteen US cities. It found that 28.5 percent of local pack businesses were absent from Google's AI Mode, with a 71.5 percent overlap and a 95 percent confidence interval of 69.2 to 73.8 percent from 10,000 bootstrap resamples.

Restricting to the top three positions dropped the overlap to 48 percent. Businesses appearing only in AI Mode had 49 percent more reviews at the median. The study also notes that a bare query returned no listings in all ten cities tested while adding near me returned listings in all 38 cases, and that results shift day to day.

Note what this means practically: ranking in the local pack no longer guarantees appearing in AI results for the same query. They are separate selection systems.

The statistics that are not real

Four figures appear in almost every local SEO article. Here is where each actually comes from.

46 percent of all Google searches are local

Origin: a tweet from 10 October 2018 reporting a Google representative's remark at an event. Written up by Barry Schwartz at Search Engine Roundtable a week later under a headline that was itself a question, with Schwartz noting that all he had was one tweet and that 46 percent seemed super high.

Google never published it. There is no methodology, no definition of local intent, and no update in eight years.

88 percent of local mobile searches result in a call or visit within 24 hours

This one cannot be traced to any primary research at all. The attribution repeated across dozens of sites points to a content marketing agency rather than a research house, and other sites attribute the identical figure to Google. No study, no sample, no date.

76 percent who search nearby visit a business within a day

This is genuinely Google published, via Think with Google around 2016, which puts it a decade old with no sample size or method stated alongside it. It gets conflated with a better documented 2014 Google and Ipsos study of 4,500 respondents, which is a different piece of research.

Near me searches grew 500 percent

Real, but far narrower than it sounds. Google's data compared the second halves of 2015 and 2017, and only for mobile searches containing near me plus a buying phrase. It is cited as though it covers all near me searches and as though it were recent. It is nine years old.

What to do with all this

Work the documented layer first, because it is free and it is the only part Google has confirmed.

• Get the primary category right. It is top of the expert list and one of the few things Google names directly.
• Complete every field. Google states that businesses with complete and accurate information are more likely to show up.
• Build reviews steadily. Google documents review count and rating as part of prominence, and the measured evidence suggests early reviews matter more than later ones.
• Build a dedicated page for each service. This tops the expert list for localised organic results, which is a separate ranking surface from the map pack.
• Ignore anything sold to you on the strength of a percentage. There are no published weightings.

Key takeaways

• Google's own page now says relevance, distance and popularity. The prominence wording everyone quotes has changed, and the line about distant businesses outranking closer ones was deleted.
• Google publishes no ranking weightings and states there is no way to pay for better local ranking.
• Whitespark's 2026 survey is 47 experts estimating across 187 factors. Its numbers are scores, not percentages, and the percentages in circulation are invented.
• The best measured review test has a sample size of three businesses, and it suggests early reviews matter more than later ones.
• A June 2026 study of 1,120 queries found 28.5 percent of local pack businesses missing from Google's AI Mode. They are separate systems.
• The 46 percent local searches statistic originates from a single tweet in 2018.
• The 88 percent call or visit statistic cannot be traced to any research at all.

Common questions

What are the most important local SEO ranking factors?

Google documents only three: relevance, distance and popularity, with no weightings. Practitioner surveys rank primary category, proximity, keywords in the business title and reviews highest, but those are expert estimates rather than measurements, and the survey's own author disputes one of its top entries.

Do reviews affect local rankings?

Google states directly that more reviews and positive ratings can help local ranking. The best controlled test, on three businesses, found a small improvement crossing ten reviews and none after that. So reviews matter, and the returns appear to diminish faster than most agencies imply.

Is proximity the biggest local ranking factor?

Experts rank it second behind primary category, and Google names distance as one of three factors without weighting it. The often quoted claim that proximity accounts for a specific percentage of ranking decisions has no source. You cannot change your distance from a searcher anyway.

Does appearing in the local pack mean appearing in Google AI results?

No. A June 2026 study of 1,120 queries found 28.5 percent of local pack businesses absent from AI Mode, dropping to 52 percent absent when restricted to the top three. They select independently, and businesses showing only in AI Mode had notably more reviews.

A test for any local SEO proposal

Ask where each number in it comes from. If the answer is a percentage weighting for a ranking factor, it does not exist. If it is 46 percent or 88 percent, you now know the provenance. What survives that question is usually a short list of profile work, service pages and reviews, and it is worth knowing what that actually costs before you buy it.

EchoPulse builds local pages, profiles and review flows for local and service business owners, and quotes work rather than ranking promises. The $299 Pilot runs 14 days with no contract.