Local SEO is the work that gets a business found on Google by people searching in a specific area, such as a shop, a dental practice or a garage, and it is played on three different rankings: the local pack on the map, the organic results and AI answers. In the local pack the Google Business Profile listing weighs 32% and links 8%; in local organic the weights flip, with on-page at 33% and links at 24% (Whitespark, Local Search Ranking Factors 2026).
So it isn't just a scaled-down version of ordinary SEO?
No. And above all, it is not a single ranking.
There are three, each won with different levers, and most of the strategies I see work on only one of them.
The typical mistakes:
- working on the Google listing in the hope of climbing the blue links;
- buying links to get into the map;
- publishing thirty "plumber in [town]" pages that only swap the town name.
And why should you listen to us?
Because we see local search through our clients' numbers. Contesini Consulenze runs two electricity and gas advice offices in the province of Mantua and had never used Google Ads.
A mobile landing page, tracking with Tag Manager and GA4, Performance Max campaigns focused on the area around the two offices: 299 enquiries at €7.20 each, 68% of them workable.
The lesson we took away applies to local SEO too: every click from outside the area is a wasted click. A €7 lead from outside the area is worth nothing.
In this guide:
- the three rankings and what weighs in each, with the Whitespark table;
- what the top organic position is really worth on a local query (not much: 5.2% CTR);
- listing, reviews, website, citations and links, in order of weight;
- what AI has changed and what it hasn't;
- how to measure it without being fooled by "average position".
Let's start with the three rankings.
The three rankings in local search (and the weights that separate them)
Someone searching "outdoor shop london" (1,300 searches a month in the UK, DataForSEO, September 2026) gets three blocks of results built by different mechanisms: the local pack with the map and three listings, the organic blue links below, and more and more often an AI summary above everything. The first step in any local job is to read the SERP and work out which of the three is worth playing for, as with any other check of search intent.
On that SERP the pack showed:
- Cotswold Outdoor Covent Garden: 4.3 stars, 687 reviews;
- Ellis Brigham Covent Garden: 4.7, 1,200;
- Outdoor Emporium in Camden: 4.8, 410.
And the AI summary?
Not in this case: the desktop check from London showed none, only a People Also Ask box with AI-generated answers.
So how do you know what weighs in each ranking?
Since 2008 Whitespark has published the Local Search Ranking Factors, a report that asks a panel of specialists to weigh the factors that move positions.
The 2026 edition, released on 6 November 2025, involved 47 specialists and 187 factors. And the weights say something clear.
| Factor group | Local pack (Maps) | Local organic | Visibility in AI answers |
|---|---|---|---|
| Google Business Profile listing | 32% | 7% | 12% |
| Reviews | 20% | 6% | 16% |
| On-page | 15% | 33% | 24% |
| User behaviour | 9% | 10% | 9% |
| Links | 8% | 24% | 13% |
| Citations | 6% | 7% | 13% |
| Personalisation | 6% | 8% | 9% |
| Social | 4% | 5% | 9% |
The Google listing is worth 32% in the pack and 7% in organic.
Links account for 8% in the pack and 24% in organic.
In other words: anyone working on the listing in the hope of climbing the blue links is spending time on the wrong ranking. And so is anyone buying links to get into the pack.
The right-hand column is new for 2026 and shows a third, different distribution, where reviews weigh more than twice as much as in organic (16% against 6%).
How Google chooses which businesses to show (and the factor you can do nothing about)
Google says in the official Google Business Profile documentation that local results are based mainly on three criteria: relevance, distance and prominence.
- Relevance: how well the listing matches what the person searched for. You improve it by completing the listing, not by stuffing it with keywords.
- Distance: how far the business is from the searcher. If the query contains no location, Google uses the device's position.
- Prominence: how well known the business is. Reviews, ratings and links pointing to the business count here.
On the same page Google adds a sentence worth rereading: there is no way to request or pay for a better local ranking.
Not even with the monthly fee they offer you on the phone?
Not even then. It is the ready answer for anyone who calls promising "the top spot on Maps".
Then there is distance, the factor you cannot work on and the one that decides the most.
In the Whitespark 2026 report, proximity to the point of search is the second single factor in the pack with 225 points, behind only the listing's primary category (227) and ahead of keywords in the business name (223).
Two identical ice-cream parlours eight blocks apart are playing two different games. And neither can do anything to change that.
What the top organic position is really worth on a local query (5.2% CTR)
On a non-branded local query the top organic position brings very few clicks: for an outdoor shop in Rome the CTR was 5.2% at an average position of 1.7, against 37.6% for branded queries at position 1.1. It is the number you won't find in the guides, and it changes the order of priorities.
I took that shop's Search Console data for the three months from 1 June to 28 August 2026, and separated branded queries from non-branded local ones.
| Query type | Average organic position | Impressions | Clicks | CTR |
|---|---|---|---|---|
| Branded (shop name, shop name + city) | 1.1 | 2,309 | 869 | 37.6% |
| Non-branded local (the Italian equivalents of "hiking shop rome", "hiking boots rome", "salewa rome") | 1.7 | 1,605 | 84 | 5.2% |
Same position band. A sevenfold difference in click-through rate.
Isn't that just intent?
Partly: whoever types the shop name wants that shop, whoever types "hiking shop rome" is still choosing.
But seven times is too much to be intent alone.
On a local query the first blue link sits below the pack, below the map and often below an AI summary. By the time the user gets to it, the first screen is already over.
The corollary is blunt: if the business has a physical location, the first goal is the listing, not the organic position.
Organic is for informational queries, for the ones where the pack doesn't appear, and for feeding overall visibility on the SERP. But on the query that brings customers to the counter, the game is played on the map.
Is your top organic spot bringing customers or just impressions?
For the shop in Rome the same position was worth 37.6% of clicks on branded queries and 5.2% on local ones: the map makes the difference. Let's look at your listing together, and at the queries where you currently sit below the pack.
Book a strategy consultationThe Google Business Profile listing (the top five pack factors)
If the pack is the goal, the listing is the tool: five elements move it more than all the others, and they are the top five in the Whitespark 2026 report. Primary category, proximity, name, address and opening hours.
The primary category (factor number one, 227 points)
It is the top factor overall. Choose the most specific one that genuinely describes the business: "sporting goods store" and "outdoor equipment store" compete on different queries.
And how is it usually set?
Badly. Most listings have the category set on the day of registration and never touched again, often the broad category Google suggested on the first attempt.
Secondary categories are added afterwards, and they help.
A BrightLocal study of 20 June 2023 on 1,050 US locations, measured with a 7x7 grid at 500-metre spacing, found an average Maps ranking of 7.6 for listings with no additional categories and 5.9 for those with four.
It is a US survey, but the category mechanism is the same everywhere.
The business name (third factor, and the most common trap)
Keywords in the name are the third factor in the pack. Which explains why so many listings look like "[Surname] Dental Implants Invisalign Manchester".
The point: the name on the listing must be the real trading name.
Adding services and cities breaks Google's guidelines, and it is one of the edits competitors report most often and most successfully.
If the trading name already contains the service, write it. Otherwise don't: the gain of two weeks is paid for with a suspension.
Address and opening hours (fourth and fifth factors)
A physical address in the city being searched is the fourth factor (213 points). Being open at the time of the search is the fifth (189).
Wrong opening hours don't just lose phone calls: they lose positions in the pack.
Holiday closures, summer hours and bank holidays belong on the listing, not only on the door.
Reviews, and which ones count (those from the last three months)
Reviews are worth 20% of the pack, the second factor group after the listing, and recent ones count most. BrightLocal publishes its Local Consumer Review Survey every year: the 2026 edition, released on 11 February 2026 with a sample of 1,002 US adult consumers, reports these figures.
- 97% read reviews of local businesses.
- 74% prioritise reviews from the last three months.
- 31% only use businesses with at least 4.5 stars, up from 17% the year before.
- 19% expect a reply the same day.
The data was collected in the United States and should be read as a trend, not as a snapshot of the UK market.
So which number really matters?
The one on recency.
A profile with 200 reviews all from 2023 is worth less than one with 40 reviews spread over the last twelve months. In the Whitespark report, a steady flow of new reviews is among the top fifteen pack factors, separate from the total count.
How to ask: the short link generated from the listing dashboard, given to the customer while they are still in the shop or in the message that follows the appointment.
How to reply: to all of them, five-star ones included, taking genuine disputes off the public page. Review management is part of the experience and trust signals Google also assesses outside local search.
The website, where the weights flip (on-page 33%, links 24%)
In local organic the factor ranking is different: on-page at 33%, links at 24%, the Google listing at 7%. It looks like the general ranking factors, and that is no accident.
Is the top factor a keyword?
No, a structure: a dedicated page for each service, 210 points in the Whitespark 2026 report.
So the "Services" page with ten paragraphs stacked one under the other needs splitting up. A dental practice offering implants, orthodontics and hygiene needs three pages, each with its own title tag and heading structure.
The second factor in local organic is the geographic relevance of keywords (190 points): the city appears where it makes sense for it to appear, not repeated at the end of every paragraph.
Location pages: only where there is a real branch
On location pages there is a rule few people say out loud: build them where there is a real branch or real work.
Thirty pages for "plumber in Salford", "plumber in Stockport", "plumber in Trafford" generated by swapping the town name? Duplicate content with a variable inside, and Google treats them as such.
If you cover a whole county, one page explaining the area served works better than thirty clones.
The two ten-minute technical checks
- Consistent NAP: name, address and phone written the same way on the website, the Google listing and directories, commas included. In the Whitespark report, NAP matching between website and listing is among the top fifteen pack factors.
- LocalBusiness markup on the contact page, with latitude and longitude inside.
Nobody does them. They take ten minutes.
Citations and local links (6% in the pack, 13% in AI answers)
Citations are mentions of the business details on other websites, with or without a link, and in 2026 they weigh 6% of the pack but 13% of visibility in AI answers. In the UK the directories that still count are few, and Google itself shows which ones: on "outdoor shop london" the "Find results on" box above the pack lists Yell, Yelp, Cylex and Thatsup.
Add:
- Bing Places;
- the vertical portals of each sector: TripAdvisor for restaurants, Checkatrade for trades, Rightmove and Zoopla for property;
- Apple Maps, which feeds searches made from an iPhone without going through Google.
Weren't citations on the way out?
In the pack, yes. They have become useful again for a different reason from the one they were created for: language models read those pages to understand who you are and where you are.
The same logic as classic link building applies, with one difference: here data consistency counts for more than domain authority.
As for real links, local sources are the most obvious and the least used: sponsoring local teams and associations, chambers of commerce and business groups, local papers, suppliers that publish a list of stockists.
A "where to buy" page from a brand the business sells is a relevant local link you can get with one email.
Local SEO and AI search: what has changed and what hasn't (the pack still shows 93% of the time)
On strictly local queries the AI Overview appears in only 15% of cases and the local pack in 93%: AI has taken the informational side of search, not the side that brings customers into the shop. The data comes from a Whitespark study of 12 May 2025 on 540 queries in three US cities (Houston, Phoenix, Denver) and six sectors, from plumbers to optometrists.
There is a lot of fear here and not much data. And the result is more interesting than the headline it was given when it did the rounds.
| Query type | AI Overview present | Local pack present |
|---|---|---|
| Strictly local ("plumber in Denver") | 15% | 93% |
| Informational ("how long does an eye exam take") | 92% | 6% |
| Hybrid ("average cost of dental implants in Phoenix") | 97% | 17% |
| Average across the 540 queries | 68% | 39% |
On the query that brings a customer into the shop the pack is intact, and the AI Overview appears in one case in seven.
What AI has eaten is the informational half of the funnel: the questions people ask before choosing.
That is a problem for anyone who built local traffic on content (the medical practice blog, the garage's maintenance guide). It is not yet a problem for the listing. It is worth reading alongside the data on zero-click searches, which describe the same thing from another angle.
The figure that shifts the medium term (from 6% to 45% in a year)
In the same BrightLocal survey of February 2026, 45% of the 1,002 US consumers interviewed say they use tools such as ChatGPT to get recommendations for local businesses. The year before it was 6%. And 40% say they trust the suggestions they receive.
Seven times in twelve months.
A speed not seen since the smartphone replaced the Yellow Pages.
And in the UK?
There is no equivalent survey, and anyone quoting the US one as if it were British is cheating.
What UK data does show is how crowded the field is: according to the Department for Business and Trade, there were 5,690,265 private sector businesses at the start of 2025, up 3.5% on the year, and 75% of them had no employees besides the owners.
The share with a well-kept Google listing is a minority. That is the competitive advantage of anyone who has read this far.
To get cited by the models the same things apply as for the pack, with two additions: consistent data everywhere and pages that answer questions with numbers in them (indicative prices, lead times, areas covered). The method is in the guide on how to appear in ChatGPT answers and the one on AI Overviews.
How to measure it, and why almost everyone measures it badly (branded searches inflate everything)
Local SEO is measured with four tools: the listing's own metrics, Search Console filtered by city, grid rank tracking, and the split between branded and non-branded queries. In order of usefulness:
- The listing's own metrics: calls, direction requests, website clicks, and where the views come from (Search or Maps). They are the only ones that measure the pack.
- Search Console, filtering the queries that contain the city name and those with "near me", looked at apart from the rest. It is the exercise that produced the table from the shop in Rome.
- Grid rank tracking, which queries the pack from dozens of different points across the city instead of just one.
- Separating branded from non-branded, without which the KPIs tell the story of word-of-mouth growth and you mistake it for SEO.
With Contesini we applied the same rule from day one: tracking with Tag Manager and GA4 before the campaigns, and every enquiry assessed by the sales team. That is how we know 68% of the 299 enquiries were workable, not just how many there were.
And timing?
Nobody can promise it. The order of magnitude for a new listing in a competitive market is a few months, and what the article on how long SEO takes says still applies.
The variable that shortens the time in local search is not budget, it is the density of competitors within a one-kilometre radius.
In local search, average position is a number that doesn't exist (use the grid)
In the local pack there is no such thing as "the position" for a keyword: the same query, run from two points in the same city three kilometres apart, returns two different packs. The business that is first in one neighbourhood may not appear at all in the next.
So what does the tool reporting "position 4 for plumber manchester" measure?
An average of different races.
And the average of different races is the result of none of them.
The right unit of measurement is the grid: query the pack from twenty or thirty points mapped across the area and look at where you are in and where you are out.
It serves two purposes:
- working out where to open the second shop, because the gap in the grid is a map of unserved demand;
- stopping the chase for a number that goes up and down without anyone having done anything.
Our local SEO service starts exactly here: the grid, the listing, and the queries that currently see you on page two. If you would rather do it yourself, a morning and the Google Business Profile dashboard are enough.
Do you know which neighbourhoods your listing disappears from?
The same search run three kilometres away returns a different pack. With Contesini we focused everything on the area around the two offices and got 299 enquiries at €7.20 each: we start from your grid and tell you where you are out.
Book a strategy consultationLocal SEO sector by sector (with the data from each SERP)
We have applied these principles to real data from several sectors, with SERPs, volumes and CPCs measured:
- Dentists: SEO for dentists, with SERPs, volumes and CPCs for dental practices.
- Hotels, where local SEO counts for more than classic SEO: on a "city centre hotel" search the first element on the page is the Google Hotels box, fed by the Business Profile listing. The data is in SEO for hotels.
- Private healthcare, where the local pack and branch pages decide visibility: SEO for medical practices and clinics.
- Psychology, nutrition and physiotherapy practices, SERPs and keyword difficulty included: SEO for psychologists.
- Accountants and other advisers: the SERP combines a local pack of firms with a block of directories below it. How a local firm gets onto that page: SEO for professional services.
- Restaurants, with real data from SERPs and restaurant websites, volumes, CPCs and branded keyword shares: SEO for restaurants. Restaurants are also a sector where the listing beats ads on generic searches; where Google Ads does pay off is in Google Ads for restaurants.
- Property, where the local pack sits above everything, often with several branches of the same agency brand in the same pack: real estate SEO.
- Law, where the local pack and the organic results are made up of individual firms, with few portals in between: SEO for lawyers.
- Construction, where the local pack decides almost everything and many builders ranking organically have only a handful of keywords: SEO for construction companies.
Recommended reading: how it works for garages and tyre shops.
Frequently asked questions
It depends on the type of business: to appear in the local pack you need a verifiable address, but Google also allows service-area businesses, meaning those that go to the customer (plumbers, electricians, home services). In that case the address is hidden and the areas served are declared. With neither a location nor a service area you cannot get into the pack, and only local organic is left.
It depends on the ranking, and that is exactly the difference: in the local pack the Google listing is worth 32% and links 8%, while in organic the weights flip, with on-page at 33% and links at 24% (Whitespark 2026). In traditional SEO distance does not exist as a factor; in local it is the second heaviest.
No. Google says so in the Google Business Profile documentation: there is no way to request or pay for a better local ranking. You can buy ads, which are a different thing and are labelled as such.
It depends on your competitors: there is no threshold. What counts is the number compared with others in the same radius, the average rating, how often new ones arrive and whether they contain text. A profile with 40 reviews spread over the last twelve months works better than one with 200 reviews that stopped in 2023, not least because 74% of the consumers surveyed by BrightLocal in 2026 (US sample) prioritise reviews from the last three months.
It depends: only if there is a real branch or real work in that town to describe. Pages generated by swapping the town name are duplicate content and do not rank. If you cover a whole county, one page explaining the area served works better, with concrete references to jobs done in those towns.
No, not on the queries that bring customers into the shop: in the Whitespark study of May 2025, strictly local queries showed the local pack in 93% of cases and an AI Overview in 15%. The impact is on informational queries, where the AI Overview appears in 92% of cases. The new game is elsewhere: in 2026, 45% of the US consumers surveyed by BrightLocal use tools such as ChatGPT to get recommendations for local businesses, up from 6% the year before.
Where to start (a morning and the listing dashboard)
Not one ranking, but three: the listing at 32% in the pack, links at 24% in organic.
Not the top organic position, but the map: 5.2% of clicks against 37.6%.
Not 200 reviews from 2023, but 40 from the last twelve months.
Not average position, but the grid.
Now, two concrete things.
One: open your Google listing and check the primary category. If it is the broad one suggested on the day you registered, pick a more specific one and add secondary categories: in the BrightLocal study, listings with four additional categories averaged 5.9 instead of 7.6.
Two: in Search Console, filter the queries containing your city name and split branded from non-branded. If the CTR of the second group is far below the branded one (for the shop in Rome: 5.2% against 37.6%), your game is not in the blue links: it is on the map.
If you want us to look at your listing, grid and queries together, book a strategy consultation.
See you next time!
