Research, Sources & Methodology

Relevance is how well a profile matches what someone searched for. Distance is how far each result is from the searcher or the location term used. Prominence is how well known a business is, which Google says is informed by information across the web, including review count and review score.

Google also says that complete and accurate information helps a profile appear in more relevant searches, and that it does not publish the weighting behind these systems.

We therefore never claim to know Google's weighting. Where we prioritise one task over another, that priority is GBPly's own judgement — see the GBP Expert methodology below.

These are Google's own published figures, attributed in the playbook to Google internal data. They describe an association between profile completeness and engagement — they are not a statement that completeness is a weighted ranking factor.

Google's playbook also reports that 96% of customers are more likely to visit a business that displays its opening hours, and that 91% of consumers search online before visiting a local business.

The same playbook is the source for two further figures we quote on the industry pages: 65% of consumers are more likely to choose a business that responds to its reviews, and businesses adding photos see 42% more requests for directions and 35% more clicks to their website. These are general findings across local businesses — we apply them as context for a sector, never as sector-specific measurements.

Google's 2026 playbook states that categories are crucial for online visibility and that selecting them accurately helps a profile appear in the most relevant searches.

Independent research supports the same point: Whitespark's expert survey consistently places the primary category among the strongest local ranking factors. That is expert opinion, not a Google statement.

Setting a service area helps Google understand the geography a business serves. It does not guarantee ranking throughout that entire catchment — distance from the searcher remains one of Google's stated ranking inputs, and we never present service-area coverage as a promise of visibility across it.

These figures appear in Google's 2026 Business Profile Best Practices Playbook, sourced there to Google internal data. This is a first-party Google statistic, which is why we use it in place of older third-party photo statistics.

It describes an association between having photos and customer actions. It is not a statement that photo volume is a ranking factor.

The weekly cadence and post-type mix are Google's own stated best practice in the 2026 playbook. The visibility figures come from the Crate & Barrel case study cited in that playbook.

A single case study and a recommended cadence are not the same as a confirmed ranking factor. We recommend weekly posting because Google recommends it and because it keeps a profile current — not because Google has said posting frequency is weighted in local ranking.

Google's own guidance asks businesses to respond to reviews and prohibits soliciting reviews selectively, offering incentives for reviews, or discouraging negative reviews. GBPly's review request tooling sends the same request to every customer on a list — it does not filter by expected sentiment.

Reviews can only be removed by Google where they breach Google's content policy. A business disagreeing with a review is not by itself grounds for removal, and we never state otherwise.

We do not say that Ask Maps uses Business Profiles and reviews instead of websites. Google's own description is broader than that, and includes web content.

What follows from Google's description is straightforward and is what we actually claim: an accurate, complete, well-reviewed profile gives these systems better material to work with. We do not claim to know how Ask Maps weights its sources.

Scale figures: where we quote 'over 300 million places' and 'more than 500 million contributors', those are Google's own numbers from its Ask Maps announcement, describing the Maps ecosystem as a whole — not a statement about how much of an answer comes from any one source.

Launch date: Google announced Ask Maps in March 2026. Availability has rolled out gradually and by region, so what a searcher sees will differ from place to place.

This matters for two claims we make. First, a competitor or member of the public can suggest an edit to your profile without having any administrative access to it. Second, information you never entered can still appear on your profile.

Profile Guard exists because of this: it records what changed, when we observed it, and what the previous value was.

GBPly does not display per-post views, clicks or click-through rate, because Google no longer supplies them and we will not estimate a number and present it as Google data.

The Posts tab in Insights reports publishing activity — what was published, when, and by which profile — filtered by publish date. Profile-level views and actions continue to come from Google's Performance API.

Google's API surface is the part we can state without ambiguity: there is no supported endpoint for reading or posting questions and answers on a Business Profile, so no tool — including ours — can manage on-profile Q&A. Google's API sunset documentation is the reference for what remains available.

On the consumer side, the 'Ask a question' surface has been withdrawn from Business Profiles. Google's own documentation is inconsistent about this: older Maps help material and legacy developer reference pages still describe place Q&A, while current Business Profile support guidance and Product Expert answers treat consumer Q&A as phased out.

Because of that inconsistency we describe the position as observed rather than announced: Q&A is not available on the profiles we manage and not available via the API. GBPly therefore treats Q&A as an owned FAQ knowledge base — answers you approve once and publish as Business Profile posts and as FAQ structured data on your own website, where you control them and where AI and local-search surfaces can read them.

Study type: expert opinion survey, not controlled testing and not a Google statement. It is useful for showing where practitioner consensus sits — for example the consistently high placement of the primary business category, and the comparatively lower placement of some Posts and photo variables.

Where we cite it we name Whitespark and the survey year. We never convert a survey result into 'Google confirms'.

Study type: online consumer survey, self-reported. Sample size, geography and question wording change year to year, so figures are not directly comparable between editions.

Where Google's own 2026 playbook covers the same ground — 91% of consumers using reviews to evaluate local businesses, 65% preferring businesses that respond — we now prefer the Google figure and cite the consumer survey only where Google publishes nothing equivalent.

Study type: third-party click-stream and SERP-feature analyses, aggregated and republished by several local SEO vendors. The underlying data collection is not fully published, and the click-share split traces back to older analyses rather than current Google data.

We keep these figures because the directional point — local results sit above organic and take a large share of attention — matches what our own customers see. We label them as dated third-party research wherever they appear, and we do not present them as Google figures.

Study type: originally traced to Google presentations and Think with Google mobile-behaviour material rather than a current published dataset, then repeated across the SEO industry.

We retain the figure as an order-of-magnitude indicator of how much search is local, labelled as a dated third-party figure. Where a page needs a current, verifiable number instead, we use Google's 2026 playbook figures — for example 91% of consumers searching online before visiting a local business.

Study type: Google consumer research published through Think with Google. The finding is genuine but it is from Google's earlier micro-moments research rather than current data, so we present it with that context.

Study type: vendor analysis of listing databases. Methodology and geography are not consistently published, and the figure predates several years of Google prompting owners to claim profiles.

We use it only as background context for why profile monitoring matters, labelled as dated third-party research.

Study type: vendor analysis. The commonly quoted 'one in three' style figures do not have clearly published sampling, so we treat them as directional.

The important distinction, which Google's own documentation supports, is that a suggested edit from a member of the public or a competitor is not the same as someone having administrative access to your profile. We describe the risk in those terms.

Assumed monthly workload: 4 Posts, 4 image uploads with descriptions, 40 review request emails, and replies to around 8 reviews.

Manual, roughly 6 hours: creating and publishing 4 posts ~2 hours; uploading 4 images with descriptions ~40 minutes; sending 40 review request emails ~2 hours; responding to ~8 reviews ~1 hour 20 minutes.

In GBPly, roughly 20 minutes: AI-assisted posts with batch scheduling ~8 minutes; batch image upload for the month ~3 minutes; paste an email list and auto-send ~3 minutes; one-click AI review responses ~6 minutes.

This is a GBPly estimate based on the workflow steps in the product, not a measured study of customers. Your figure will differ with volume, how much you edit AI drafts, and how many profiles you manage. It excludes strategy, photography and one-off setup.

Each checklist item is weighted High (10), Medium (5) or Low (2). The score is the share of available weight a profile has earned, so an incomplete high-priority item costs more than an incomplete low-priority one.

These weights are GBPly's prioritisation of effort versus likely benefit. They are not Google's factor weights, and Google does not publish factor weights. Completing every item does not guarantee a ranking change.

Verification status is treated separately: where a profile is not verified through Google's Voice of Merchant state, verification tasks are pinned to the top, because unverified profiles cannot benefit from the rest.

It is derived from publicly observable profile signals — review volume and rating, photo coverage, completeness of core information, category setup, and recency of activity — combined into a single 0-100 indicator.

It is GBPly's own score. It is not a Google score, not a measure of how a business currently ranks, and not a prediction of whether that business will buy anything. It answers one question: which nearby profiles look most under-managed.

We compare observable things: categories, review count and rating, photo counts, completeness of core information, posting recency, attributes and services.

Where we call a gap significant, or a competitor easier to close on, that is GBPly's heuristic reading of observable differences. It is not a statement of why Google ranks one profile above another — Google does not publish that, and no tool can read it.

Rank Replay shows what your grid positions were at each recorded scan and what profile changes we logged between those scans. It lets you see whether ranking movement coincided with a change.

Coincidence in time is not proof of cause. Local results also move with competitor activity, Google updates, seasonality and the searcher's exact location. We describe what the timeline shows, and stop there.

The underlying views, calls, direction requests and website clicks come from Google's Performance API. The baseline is the profile's own performance before active management, and the reported uplift is the difference between the two periods. Both sides use the median of each day's Maps + Search profile views, so a single spike in Google's data cannot move the figure.

We only ever quote profiles whose baseline was recorded at connection from the Google data covering the period before management began, with at least seven days in the window and the exact dates stored. Profiles whose baseline had to be reconstructed after a historical unit bug, or whose earliest data starts after connection, are tracked separately and are not used in published figures.

This is a before-and-after comparison, not a controlled experiment. It shows change since baseline. It does not isolate GBPly as the cause of that change, and we do not describe it as doing so.

Reviews are cumulative counts, so review growth needs no extrapolation: it is the count held today minus the count held when the profile connected. That starting count comes from the snapshot recorded at connection where one exists, and is otherwise derived from the date carried by each individual review, so it can be recomputed and checked.

Profiles have been connected for different lengths of time, so any comparable figure is stated per 30 days connected and as a median across profiles, never as a raw total per profile. Rating movement compares a profile's average rating over the reviews dated before connection with its average over all reviews today; profiles with no reviews at connection are excluded.

Deleted reviews and Google's own review removals reduce these figures, and losses are included rather than filtered out. This is a before-and-after measurement, not evidence that GBPly caused the change.

Call clicks, direction requests and website clicks are read from Google's Performance API. Each is reported on its own, never blended into a single 'actions' figure, and each side of every comparison uses the median of the profile's own daily values so a single spike cannot move it.

Cohort figures are the totals of per-profile medians, which are volume-weighted; the median per-profile percentage change is reported alongside them, because the two diverge whenever a few large profiles dominate the volume.

Same limitation as every other figure here: it shows change since baseline, not causation.

The weekly Posts cadence follows Google's own published best practice. The photo and review cadences are GBPly recommendations, chosen so a profile stays current without generating low-value content that could trip Google's content policies.

We do not promise a timeframe in which rankings will move. Anyone who does is guessing.

Performance metrics — profile views on Maps and Search, calls, direction requests, website clicks and search terms — are Google's numbers, shown as Google reports them.

Google publishes performance data with a lag of roughly two to three days, and publishes search terms monthly rather than daily. Our date filters anchor on the newest date Google has actually published, not on today, so the figures shown are complete.

Grid rank scans query Google Maps results from a set of geographic points at the moment the scan runs, and store the position we observed at each point. Results vary with the searcher's exact location and personalisation, which is why we record a grid rather than a single number.

Profile change history is built by comparing each synchronised copy of a profile with the previous one. Changes are detected on synchronisation, so a change is reported as at the sync that observed it, not the instant Google applied it.

Review request sending, opens where available, and portal activity are our own delivery records.

We cannot tell you how Google weights any factor, why a specific competitor outranks you, or when a change will move your position. Nobody outside Google can.

We do not guarantee rankings, traffic, calls or revenue, and we are not a Google partner or certified by Google.

Where our copy states an outcome, it is either something the product verifiably does, a figure published by Google, named third-party research, or a clearly labelled GBPly estimate. If you find anything on this site that does not fit one of those four categories, email support@gbply.net and we will correct it.