Local ranking factor percentages: what's real, what's not

Search a few local SEO articles published this year and a specific breakdown keeps turning up: GBP signals are 32% of the local ranking algorithm, on-page signals 19%, reviews 16%, links 15%, behavioral 8%, citations 7%. It reads like a real, sourced chart. Whitespark and BrightLocal do run real, long-running annual local ranking factor surveys, which makes the number easy to believe without checking. This piece checked it, found what Whitespark’s real chart actually says instead, and confirmed what Google itself is willing to name.

The breakdown that circulates#

The 32/19/16/15/8/7 figures are usually attributed, loosely, to “Whitespark or BrightLocal 2025 data.” Neither site publishes that combination. Whitespark’s real, dated 2026 report does contain a chart with the same category names, GBP signals, on-page signals, review signals, link signals, behavioural signals, citation signals, plus two categories the circulating claim leaves out entirely, personalization and social signals. Its real numbers for local pack and Maps rankings are GBP signals 32%, review signals 20%, on-page signals 15%, behavioural signals 9%, link signals 8%, citation signals 6%, personalization 6%, social signals 4%.

Only the GBP figure, 32%, matches. Every other number is shifted: review signals is 20% in the real chart, not 16%, on-page is 15%, not 19%, and link signals is 8%, not the 15% the circulating claim assigns it. That is not a rounding difference or an old survey year drifting slightly, it is a different set of numbers wearing the same category labels. A single correct figure sitting inside five incorrect ones is exactly the shape a plausible-sounding fabrication takes: close enough to a real source that no single number looks obviously wrong.

What Whitespark’s real report actually measures#

The real chart is not decorative. Whitespark’s own methodology description states that “47 of the best and the brightest local search experts are invited to complete a comprehensive 2-hour survey,” scoring 187 individual factors from 0 to 5 across four areas: local pack and Maps ranking impact, local organic ranking impact, conversion impact, and AI search visibility impact. The percentages in the group-weighting chart are built from those summed, ranked scores.

That is a real, citable methodology, and it is also a specific kind of measurement: 47 practitioners reporting what they believe correlates with ranking, not a controlled experiment isolating what Google’s live algorithm actually weighs. Google’s local algorithm is not observable from outside, so no third party, however careful, can publish a causally measured weight for any local ranking factor. Whitespark’s report is useful precisely because it is honest about being a consensus survey rather than something it is not.

What Google itself documents#

Strip away every third-party percentage and Google’s own position is short and comes with no numbers at all. Its Business Profile documentation states that local results are based mainly on three factors.

Relevance is “how well a Business Profile matches what someone is searching for.” Google says complete and detailed business information helps it match a profile to the right searches.

Distance is “how far each business is from the customer who’s searching,” calculated from a searcher’s shared location or, absent that, from what Google can otherwise infer about where they are.

Prominence is “how well-known a business is,” including signals like inbound links and review volume; Google states plainly that “prominent places are more likely to show up in search results.”

Google attaches no percentage, no weighting, and no ranked order to these three. The same documentation also states there is no way to request or pay for a better local ranking, closing off the idea that any of this can be purchased outright.

Why we care#

The lesson is not really about local SEO percentages specifically. It is about how a real number, one correct figure inside a real, methodologically sound survey, gets copied into a new combination with five wrong ones and keeps circulating because the source organization’s name is real and its survey does exist. Whitespark’s actual report is more useful than the version built from it: it tells you the categories are real, GBP signals really do lead, and it is honest that the whole exercise is 47 experts scoring what they believe matters, not a measured test of Google’s code.

That distinction matters for how this site treats any local ranking claim going forward: a percentage attributed to a real survey gets checked against that survey’s actual, current publication before it ships, not against what a downstream blog says the survey found. Google’s own three named factors, relevance, distance, and prominence, are the one part of this topic that needs no such verification, because Google is the primary source for its own algorithm.