How is a competitive visibility gap measured in local search?

Write the query exactly as captured. Do not replace several service phrases with a later report label. Record a named address, map point, grid point, or other reproducible location input, plus whether the location came from a device, a selected map point, an IP estimate, or another method.

Public Local Search Grid documentation shows competitors for a selected keyword and grid point, and separately provides a report-level competitor table for a selected keyword. Its displayed top-five list is that service's report behavior, not a Google-wide competitor rule.² If a measurement service supplies the set, record the service, query, grid point, report date, and displayed cutoff. If you select the businesses yourself, record the names before comparing the target with them.

Record whether the capture came from Google Maps, Google Search local results, or another explicitly named view. Keep Search Console website metrics separate from a Business Profile or local-result placement. Google defines Search Console average position as the average position of the topmost result from the site, and says manual results can differ by time, place, device, and recent history.³

For a trend, write down both run dates before calculating the difference. Public Local Search Grid documentation treats keyword and date selectors as part of the report and notes that its overview, date comparison, competitor table, and direct comparison sections work independently. Its top-100 and top-ten behaviors are limits of that service, not universal Google thresholds.⁴

Use this sequence:

1. Freeze the target profile and exact query set. Record spelling, language, intent, and location words.
2. Freeze the searcher location and how the location was supplied.
3. Record device, browser or app, time zone, interface language, and any location or sign-in context the capture exposes.
4. Choose one surface and record its exact mode.
5. Freeze the competitor set, its selection rule, and any service-specific cutoff.
6. Set the observation window and the visibility cutoff K.
7. Capture one row per observation unit. Mark incomplete or invalid captures according to the eligibility rule.
8. Count the target and every competitor with the same binary visibility rule.
9. Label the calculation with its sample, surface, cutoff, competitor set, and eligible-row rule.

The target visibility share is the target's visible eligible rows divided by N, the number of eligible rows. The competitor benchmark is the mean of the selected competitors' visibility shares. The gap is the benchmark minus the target share, expressed in percentage points:

N = sum(E_r)
target_share = sum(E_r * T_r) / N
competitor_share(C) = sum(E_r * C_r) / N
competitor_benchmark = mean(competitor_share(C) for every selected competitor C)
gap_pp = 100 * (competitor_benchmark - target_share)

A positive gap means the target was visible in fewer sampled rows than the selected competitor benchmark. A negative gap means the target was visible in more sampled rows under the same declared rule. Neither sign is a promise about a future position.

For two competitors, fill the table first, then calculate:

N = sum(E_r)
target_share = sum(E_r * T_r) / N
competitor_share(C1) = sum(E_r * C1_r) / N
competitor_share(C2) = sum(E_r * C2_r) / N
competitor_benchmark = (competitor_share(C1) + competitor_share(C2)) / 2
gap_pp = 100 * (competitor_benchmark - target_share)

Replace C1 and C2 with the complete frozen competitor set. If competitors were evaluated over different eligible rows, reconcile the rows before using a shared denominator or report separate denominators.

Those measures describe how people discovered or used one profile. They do not show how often the target and selected competitors were visible in the same query-location-surface sample. Keep the reporting period, profile identity, and activity metrics in a separate record.

Google's public local-ranking guidance says local results are mainly based on relevance, distance, and popularity, and that search-algorithm details are confidential. It also says there is no way to request or pay for a better local ranking.¹ Those boundaries are why the report should say, for example, “the target's observed visibility share was below the selected competitor benchmark in this sample,” rather than claiming a universal rank gap or a cause.