Google names three local ranking factors, and most of what the industry adds is inference, not fact

Three words on a help page, and a whole industry around them

Open Google's help page on improving local ranking and the core of it is short. Local results, it says, are based mainly on relevance, distance and prominence ^[1]. Relevance is how well a Business Profile matches what someone is searching for. Distance is how far each business is from the person searching, and if that person does not share a location, Google uses what it knows about where they are. Prominence is how well known a business is ^[1].

Set that beside what a business owner is likely to hear in a month of sales calls. Post on your profile every week. Build directory citations. Add your town to your business name. Add schema to the website. Answer every review within the hour, because Google is watching.

Some of that advice has evidence behind it. Some of it is a reasonable guess. Some of it has been tested and found to do nothing, and some of it works in the short term while breaking the rules that decide whether the profile stays on Google at all. All four arrive in the same confident tone.

This article sorts them. It sets out what Google has put in writing, where the industry's longer lists come from and how they are made, which claims have been tested and with what result, and where the evidence and Google's rules point in different directions. The companion piece on why a Maps position differs from street to street covers distance in detail, so this one concentrates on everything else.

What has Google actually put in writing?

The documented position is more specific than the three headline words suggest. On prominence, Google says the factor is based on information such as how many websites link to a business and how many reviews it has ^[1]. It says that more reviews and positive ratings can help a business's local ranking ^[1]. It tells owners to keep their hours, including special hours, up to date, and it says that businesses with complete and accurate information are more likely to show up in local results ^[1]. It says verification tells Google the owner is authorised to represent the business, which makes the profile more likely to appear ^[1].

It also says, in one plain sentence, that there is no way to request or pay for a better local ranking on Google ^[1]. Google's general explanation of how Search works repeats the point for all results, stating that nobody can pay it to rank higher and that results are not influenced by whether a business buys ads ^[2]. The same page explains that Google's systems recognise when a query has local intent, so a search for pizza returns nearby businesses ^[2].

Beyond the help page, Google has publicly confirmed two specific changes to local ranking. In December 2021 it described a local search update that ran from 30 November to 8 December as a rebalancing of various factors it considers in generating local results ^[8]. In December 2023 Danny Sullivan, Google's Search Liaison, said Google had long used whether a business is open as part of its local ranking systems, and that this had recently become a stronger signal for non-navigational queries ^[9].

That is close to the full public record. Google does not publish a list of local ranking factors, their weights, or which Business Profile fields carry weight. Everything beyond this point is someone other than Google describing what they believe or have observed.

Where do the long lists of ranking factors come from?

The most cited source is the Local Search Ranking Factors report, published by Whitespark. The 2026 edition, released in November 2025, invited 47 local search experts to complete a survey of about two hours, weighing and scoring 187 factors ^[3]. Each factor is scored between 0 and 5 for its effect on local pack and Maps ranking, local organic ranking, conversion, and visibility in AI search ^[3].

The respondents placed the primary Business Profile category first for local pack and Maps ranking, followed by the proximity of the address to the point of search, keywords in the business title, a physical address in the city of the search, and whether the business is open at the time of the search ^[3]. Grouped into themes, they ranked Business Profile signals highest, then review signals, behavioural signals and link signals, with citation, on-page, personalisation and social signals below ^[3].

The report is careful about what it is. Its authors write that none of the experts has special access to the internal workings of Google's local search algorithm, and that the factor list and responses are based on more than a decade of analysis, experience, research and testing ^[3]. That is an honest description of an informed opinion poll. It is not a measurement of Google's system, and nothing in the method could make it one.

The difficulty arises downstream, where the survey's output is repeated as if it were a specification. Vendor guides report, for example, that reviews have grown in importance for local pack rankings from 16% in 2023 to 20% in the latest survey ^[4]. Those percentages describe how much weight a group of practitioners gave to a theme. They do not describe how much weight Google gives it, and a reader who meets the figure without that context could easily assume otherwise.

None of this makes the survey worthless. Where many experienced practitioners independently notice the same pattern, that is information. It simply belongs at the weight of an outside consensus, not of documentation.

Which industry claims have tests behind them?

A smaller body of work goes further than opinion. Practitioners change one thing on a real profile, track rankings before and after, and publish the result. These are not controlled trials, but they are closer to evidence than a survey score.

Keywords in the business name. The local search firm Sterling Sky added a keyword to a restaurant's Business Profile name, removed it, and added it again. It reported that adding the keyword dramatically improved ranking for that term, and that removing it returned rankings to where they had been ^[6]. It also reported that the effect often appeared within hours ^[6]. The experts in the Whitespark survey place the same factor third for local pack ranking ^[3]. On the narrow question of whether it moves rankings, the evidence is about as strong as anything in this field.

Proximity. After the December 2021 update, Sterling Sky reported that Google had heavily increased the weight it gives to distance from the searcher, that map packs had become more zoomed in, and that a wider variety of businesses was appearing ^[7]. It also reported that law firms with keyword-rich names lost heavily, in one comparison falling from more than 200 top-three positions to fewer than 30 ^[7]. Google itself described the update only as a rebalancing of factors ^[8], so the specific interpretation is the agency's, drawn from its own tracking.

Opening hours. This is the rare case where a practitioner observation was later confirmed by Google. Sterling Sky reported in late 2023 that Google appeared to be treating whether a business was open at the time of the search as a ranking factor, and Google's Search Liaison then confirmed that openness had become a stronger signal ^[9].

Review count. Sterling Sky reported a small but noticeable increase in Maps ranking for the main keyword when businesses it tracked went from 9 to 10 reviews, and no similar bump from 10 to 11 ^[12]. It also argues that a steady flow of recent reviews matters more than a large, static total ^[12]. Google's own statement is more general: more reviews and positive ratings can help ^[1]. The agency's finding is consistent with Google's statement, but the specific threshold rests on the agency's own observed cases.

Services on the profile. This example shows why dates matter. Sterling Sky tested whether adding predefined services to a Business Profile improved rankings in 2019 and concluded that it did not. It retested in 2022, adding a specific legal service for a law firm, and this time saw a measurable effect ^[11]. Both may have been right at the time; a test is a photograph of one moment.

Which claims have not survived testing, or were never tested?

Posts. Weekly Google Posts are a staple of monthly local SEO packages. Sterling Sky added one post a week for nine weeks to three inactive profiles, with no other changes, and reported no measurable effect on local pack rankings ^[10]. Its conclusion was that posts do not directly affect local ranking, though they may still be useful for communicating with customers ^[10]. The experts surveyed by Whitespark agree: the frequency and quantity of posts both sit below 140th place out of 187 factors for local pack ranking ^[3]. The case for posts as a ranking lever is weak.

Citations. Consistent listings of name, address and phone number across directories were for years a central local SEO service. Whitespark's own report now says citations have been decreasing in importance as a local search strategy for the past decade, while suggesting that AI search may be giving them new relevance ^[3]. Google's help page does mention links from other websites under prominence ^[1], but it says nothing about directory counts or consistency scores. Accurate listings are sensible housekeeping, because a wrong phone number loses a customer whatever the ranking effect. A claimed ranking gain from citation volume is not documented by Google.

Behavioural signals. The surveyed experts rank behavioural signals, such as clicks and requests for directions, third among the eight themes ^[3]. The strongest public evidence that Google uses interaction data came from the United States antitrust case, in which Google's Pandu Nayak described a system called Navboost as one of its important signals, drawing on user interactions with results, and said location was one of the ways its data could be sliced ^[18]. That testimony concerned web results. It shows that interaction data matters to Google Search in general. It does not tell a local business how, or how much, such data affects the local pack.

The 2024 leak. In 2024, thousands of internal Google API documents became public, and a great deal of commentary treated the attribute names in them as confirmed ranking factors. Google's response was to caution against making inaccurate assumptions about Search based on out-of-context, outdated or incomplete information ^[19]. The statement is self-interested but fair: a field name shows that something is stored, not how it is used.

Schema markup. Structured data is often sold as a local ranking factor. Google's documentation says structured data enables a feature to be present but does not guarantee it, and that a structured data manual action removes eligibility for rich results without affecting how a page ranks in web search ^[20]. Structured data has real uses, but Google's wording does not support selling it as a local ranking lever.

Where do the evidence and the rules point in different directions?

The most uncomfortable finding in this subject is that several of the tactics with the strongest evidence are the ones Google forbids or warns against.

Keywords in the business name are the clearest case. Google's guidelines state that including unnecessary information in a business name is not permitted and could result in the suspension of the Business Profile, and they give marketing taglines and service descriptions added to a name as examples ^[5]. The honest summary is that the tactic tends to work until it is detected, reported by a competitor or caught by an update, at which point it can cost the business its profile. The December 2021 update, as reported by Sterling Sky, hit keyword-rich names particularly hard ^[7].

Opening hours are a second case. When Google confirmed openness as a stronger signal, Danny Sullivan added that the signal might change as Google continued to evaluate it, and said he would not recommend businesses change their hours to appear open, given the signal may continue to be adjusted ^[9].

The address factor is a third. The survey places a physical address in the city of the search fourth ^[3]. Google's guidelines say that a virtual office, meaning a rented mailing address where the business does not operate, is not eligible for a Business Profile, and that a co-working office qualifies only if it has clear signage, receives customers during business hours and is staffed by the business ^[5].

Reviews are the fourth, and the one with a specifically British twist. Since April 2025, the Digital Markets, Competition and Consumers Act 2024 has banned fake reviews and concealed incentivised reviews, and the Competition and Markets Authority has published guidance on what that means ^[14]. The Chartered Trading Standards Institute's business guidance notes that businesses must take reasonable and proportionate steps to prevent fake or undisclosed incentivised reviews, and lists suppressing negative reviews and cherry-picking good ones as misleading practices ^[16]. The CMA's guidance says that asking for reviews without predetermining their content or sentiment is lawful, and that an incentive can be lawful if it is disclosed and the review still reflects a genuine experience ^[14].

Google's rule is stricter than the law. Its Maps content policy says merchants must not offer incentives such as payment, discounts, or free goods or services in exchange for any review, must not discourage or prohibit negative reviews, and must not selectively solicit positive ones ^[13]. It permits merchants to encourage reviews that represent a genuine experience, without offering incentives ^[13]. In January 2025 Google gave undertakings to the CMA to strengthen its enforcement, including sanctions against businesses that boost their ratings through fake reviews and prominent warning alerts on profiles where suspicious activity is found ^[15]. For a UK business, the practical rule is the stricter of the two: a disclosed discount for a review may be lawful, and it still breaks Google's policy.

There is a related myth about replying to reviews. Replying is often sold as a ranking factor. Google's help page says that replying shows customers their feedback is valued and that helpful replies can show a business is responsive ^[17]. It says nothing about ranking. It is good customer service, which is reason enough.

What this means for a local business

The most useful habit is to ask one question of every recommendation: is this something Google has documented, something a published test supports, or someone's opinion? Each category deserves a different level of trust, and a good adviser should be able to say which one a recommendation belongs to.

On that basis, the documented work comes first. Google says complete and accurate information helps a profile appear, that verification matters, that hours should be current, and that reviews and links from other websites feed into prominence ^[1]. None of it is exotic, and much of it is neglected. A wrong category or stale holiday hours will not be fixed by weekly posting.

Tested but unconfirmed claims come second, and they need judging against the rules. Choosing the primary category that most accurately describes the business is consistent with both Google's guidance on relevance and the experts' view of what matters most ^{[1] [3]}. Adding the services the business genuinely offers is consistent with Sterling Sky's more recent testing ^[11]. Adding keywords that are not part of the real business name is not, whatever the test results show, because the guideline is explicit and the penalty is the profile itself ^[5].

Reviews deserve a process rather than a campaign. Ask every customer in the same way, without offering anything in return and without filtering out the ones likely to be unhappy, which satisfies both Google's policy and UK law ^{[13] [14]}. Sterling Sky's testing suggests that a steady flow of recent reviews counts for more than a large, static total ^[12].

Opinion comes last. It can still be right, but a proposal built mainly on citation volume, weekly posts or schema rests on the weakest evidence, and an owner is entitled to ask for better before paying.

Two things are also worth saying plainly. No adviser can promise a position in the local pack, because distance is decided by where each searcher stands and Google states that ranking cannot be bought ^[1]. And the work that is best supported by evidence is also the work that makes a profile more useful to a customer, which is not a coincidence.

What remains uncertain

A great deal. Google does not publish the weights of its local factors, so every claim about relative importance, including the ordering in the Whitespark survey, is inference ^[3]. The factors themselves move: the services test gave opposite answers three years apart ^[11], and Google said of the openness signal that it may continue to be adjusted ^[9].

The tests that exist are small. Most involve a handful of profiles, often in North American markets, run by agencies that sell local search services. That does not make them dishonest; some publish results that cut against their own services, such as the finding on posts ^[10]. But no independent, large-scale study of local ranking factors in UK markets was found in researching this article, and the absence matters. A city with a dense centre and a different mix of competitors may not behave as a test in a North American suburb did.

The role of interaction data in the local pack remains unclear. The antitrust testimony establishes that Google uses interaction data in web ranking ^[18], but there is no public account of how it applies to Maps, and anyone who claims a precise answer is going beyond the record.

Finally, AI search is adding a new layer. Whitespark's 2026 survey now scores factors for their effect on visibility in AI answers, and suggests that citations may matter more there than they do for the local pack ^[3]. That is an early, opinion-based view of a fast-changing system. The sensible response is the same as for the rest of this subject: watch for evidence, and do not pay for certainty that nobody has.