The photos are current, the reviews keep arriving, the description reads well and every category is filled in. You are still not in the three-pack, while a competitor with a scruffier profile has sat there for months. You don’t have an optimisation problem. You have a ranking problem, and they are not the same thing.
The Google Business Profile ranking factors that decide the map pack come down to three signals Google names itself: local results are “mainly based on relevance, distance, and popularity”. Google says popularity, the industry says prominence, same thing. A fully optimised profile maxes out one of those three. That is why the work felt finished and the ranking never moved.
This piece explains what each signal is, why Google weighs it the way it does, and why your weighting differs from the case study you read last week. It is not a task list. If you want the steps, the companion guide on how to rank higher on Google Maps has them.
The Three Signals Google Actually Uses: Relevance, Distance and Prominence
You have read the list. Relevance, distance, prominence, three bullet points, next article. Ask what affects Google Maps ranking and the same list comes back, never an explanation of what Google is working out when it applies it.
Google’s definitions are short. Relevance is how well a profile matches the search. Distance is how far each business sits from the searcher, using whatever Google knows about their location. Prominence is how well known a business is, judged partly on how many sites link to it and how many reviews it has.
Google adds one more line: “There’s no way to request or pay for a better local ranking on Google.”
Now the part that gets left out. Google is not scoring businesses. It is managing risk for someone who wants a result they can act on in the next few minutes. Each pillar retires a different risk.
Relevance retires “wrong sort of business”. Distance retires “too far to bother with”. Prominence retires “technically correct, bad experience”.
A result that clears all three is a safe bet. One that fails a pillar badly is a wasted click Google would rather not serve. That is why local pack ranking factors don’t add up neatly: excelling at one pillar buys no forgiveness for failing another, which is the closer-competitor problem below.
Which leaves the uncomfortable bit. A fully optimised profile satisfies relevance and little else: enough to be eligible, nowhere near enough to rank.
Three signals, one job. Most stuck businesses are losing on a pillar they have never audited.
Relevance: Why Your Primary Category Outweighs Everything Else You Fill In
Whitespark’s 2026 Local Search Ranking Factors report puts Primary GBP Category at number one for Local Pack and Maps, scoring 227, with proximity second at 225. When an article repeats “category is the number one ranking factor” with no source, that report is it.
Relevance is a matching problem, and the primary category is the strongest single statement of what your business is. Everything else, the services, the description, the attributes, refines a match the category has already framed. Local View’s tiering lands in the same hierarchy: S-tier is primary category, name keywords and reviews; A-tier is photos, secondary categories, address and website; B-tier is services, hours and attributes.
Then the signal nobody accounts for: whether you are open. “Business is Open at Time of Search” jumped to #5 in Whitespark’s 2026 Local Pack and Maps list, first surfaced by Joy Hawkins in a December 2023 webinar. A business at #1 can start slipping roughly 60 minutes before closing and be displaced by open competitors once the doors shut. Whitespark warns that padding your hours to look open around the clock breaches Google’s guidelines and annoys customers who arrive to a locked door.
Relevance is the cheapest pillar to max out and the easiest to mistake for finished work. Optimising gets you into the game, not past it. It decays quietly: profiles rank fine, then slip as photos age, services go stale and reviews sit unanswered, because nothing visibly broke.
If your rankings wobble across the day rather than across the week, look at your opening hours before anything else.
Distance: Why a Closer Competitor Beats a Better Business
Better reviews than them, better profile, better website. And a competitor two streets closer sits in the pack while you don’t. That is the pillar working as designed, not a glitch.
Your ranking is not a number. Google’s local results work as a grid, so a profile can sit at #1 for searches starting on its own street and be invisible three streets out. That is why local SEOs use geogrid trackers that scan many points across a service area. Checking your rank from your own office is the least representative reading available.
Proximity runs off where Google’s geocoding engine drops your map pin, not the address on file. Google assigns that pin a confidence level. When confidence is low, it doesn’t guess your building. It falls back to the city centre, and your business quietly relocates by miles.
Usual causes: unit numbers crammed into address line 1, new construction Google hasn’t mapped, and co-working addresses, the most geocoding-hostile of all.
Boundaries get decided in metres. Joy Hawkins documented a business verified in Olathe whose pin fell outside the city border. Google re-attributed it to neighbouring Lenexa, so it ranked badly for Olathe searches and well for Lenexa ones. The UK version is a unit on a road forming a borough boundary. Pin-correction steps live in the how-to guide.
Distance is the one pillar you cannot control, because Google shows different results depending on where the searcher is standing. Search Atlas’s Manick Bhan calls proximity a fixed baseline, not a competitive lever.
Distance costs you a pillar you cannot buy your way out of. It does not cost you every search: the grid means you lose specific points, not the whole map.
Prominence: The Pillar You Can Actually Move
You cannot move the building. You can move this. Prominence is the pillar owners control most directly and invest in least, the highest-leverage ground for a business stuck where it stands.
Search Atlas ran an XGBoost regression study on 3,269 local businesses across food, health and law. Its global model for positions 1 to 21 gives proximity 55.2% of predictive weight, review count 19.2%, domain power 5.9% and semantic relevance in reviews 5.3%. In the top 10 it shifts hard: proximity drops to 36.2% while review count climbs to 26% and review keyword relevance to 22.8%. Search Atlas’s own reading: proximity decides who gets into contention, reviews decide what happens inside it.
Prominence is built from reviews, citations, links, brand mentions, website authority and behavioural signals. Whitespark’s 2026 report has review and behavioural signals rising, on-page and link signals slipping, citations in slow decline. Whitespark scores “Recency of Reviews” (#12) and “Sustained Influx of Reviews Over Time” (#14) separately from raw count. A steady drip of new reviews signals a living business, while 20 in one weekend then silence looks fake, which is why a big review total can sit there doing nothing.
That 22.8% for review keyword relevance means the language inside reviews carries weight, not just the star average. A review naming the service and the town does work that “great service, highly recommend” does not.
Those are study weightings, not algorithm weightings, a distinction the final section unpacks. Prominence is still the only pillar that repays sustained effort, and the top-10 data says it decides positions once you are in contention.
Why These Ranking Factors Weigh Differently for Your Business
The GBP ranking factors league table is an average, and no business is average. The pillars weigh differently for a solicitor and a salon, and advice built for one wastes the other’s money.
Search Atlas’s sector breakdown shows it. Law is proximity-dominated: 67.4% of predictive weight for positions 1 to 21, 69.4% in the top 10. Reviews move a firm far less than location. Beauty runs the other way: proximity is only 31.1%, while review volume (27.0%) and review keyword relevance (26.3%) outweigh it, so a salon further out can outrank a nearer rival.
Health inverts by position: reviews lead at 33.4% for positions 1 to 21, distance at 36.1% in the top 10.
Service-area businesses sit outside this entirely: no address, no pin, still eligible across their declared service areas. Without that anchor, MapRanks notes, prominence and relevance carry more of the load and strict proximity less.
Andy Walker’s three-line diagnostic: “Great reviews but wrong primary category. Relevance problem. Optimised profile but one location covering a massive city. Distance problem. Strong website but no citations. Prominence problem.”
The pack is the tell. If the businesses above you are consistently the closest, reviews will not rescue you; if they sit further out on deeper, service-specific review bases, you have a lever.
Most useful when you are splitting a limited budget across pillars. Skip it if you already know your pin or category is wrong: that is a fault, not a weighting.
Where the Ranking Factor Percentages Come From and How Far to Trust Them
Nearly every article on Google Business Profile ranking factors quotes “the 2026 survey” without saying which one: Whitespark’s Local Search Ranking Factors report, run by Darren Shaw since 2017, created by David Mihm in 2008.
47 experts score 187 ranking factors from 0 to 5 across four impact areas, including AI search visibility, new for 2026. Factors group into eight categories: GBP, on-page, review, link, behavioural, citation, personalisation and social. Whitespark’s caveat reframes it: “None of the experts have special access to the internal workings of Google’s local search algorithm.”
Search Atlas publishes its limits too: proximity may be overweighted by its fixed-radius grid, it measures correlation not causation, the dataset is sector-imbalanced, and it is one May 2025 snapshot. A study that publishes its own weaknesses beats one that doesn’t. Read every percentage here as directional evidence, not algorithm weighting.
Then the industry’s favourite orphan stat: NAP consistency makes you “40% more likely to appear in the local pack”, credited to BrightLocal, with no traceable study behind it. That is how a number becomes a fact by repetition.
Shaw models the alternative. Removing your address usually tanks rankings, but he has seen cases where it didn’t, and says he doesn’t know why.
If an article quotes a percentage without naming who measured it, treat it as a hypothesis and check your grid.
FAQ
Why does a closer but weaker competitor still outrank me?
Because distance is one of three pillars rather than a tiebreaker, and it can outweigh a review or relevance advantage outright. Search Atlas puts proximity at 67 to 69% of predictive weight for law firms. There is also a scope problem in the question: they may only outrank you for searches starting near their own premises, not across the map.
What is the difference between the local pack and local organic results?
Two separate ranking systems. The local pack, the map three-pack at the top, is ranked from Google Business Profile data on relevance, distance and prominence. Local organic, the blue links underneath, is ranked by Google’s broader web algorithm on content quality, backlinks, domain authority and on-page relevance. A business can do well in one and badly in the other: location is a hard constraint on the pack, no barrier in organic.
Does hiding my address as a service-area business hurt my rankings?
Usually, yes. Most practitioners see rankings fall when the address comes off, because it removes the proximity anchor Google was working from. It is not a settled rule: Whitespark’s Darren Shaw has seen cases where an address came off and rankings did not move, and he says openly that he doesn’t know why. Treat it as usually risky, with an unexplained exception.
How does Google’s AI Overview decide which local businesses to name?
Differently from the local pack. RankHarvest reports AI Overviews appearing in roughly 40% of local searches as of May 2025, pulling from sources beyond your profile: Yelp, Tripadvisor, Reddit, Instagram and company websites. They favour reviews carrying specific service language and location detail over generic praise, and use structured local schema to extract facts. It is a synthesis, not a re-ranking.
For the fixes, the companion guide on how to rank higher on Google Maps has the steps. If you would rather someone worked out which pillar is costing you, that is what our Google Business Profile management is for.