How to Read a Google Maps Ranking Grid and Choose the Next Fix
Read local ranking grid points, compare the same locations, handle missing results, and turn an agency scan into a specific next action. Includes worked examples and a scan brief.
The short answer
- Each point represents a sampled search location. Confirm the query, business listing, result surface, time and scan depth behind it.
- Read actual positions and the provider's legend. “Not found within the checked depth” and “scan failed” are different states.
- Compare the same coordinates and settings. Report missing observations and name the denominator behind every percentage.
- Inspect the results at weak points, verify a specific business or page problem, and assign a next action. A grid alone cannot prove why a ranking changed or how many leads it produced.
A client can rank second beside the shop and disappear from the results a few neighborhoods away. A Google Maps ranking grid shows a business's observed position for one query at several sampled locations. Read the point settings and missing-result labels before diagnosing the colors. A red edge gives an agency somewhere to investigate; it does not identify the fix.
On this page
Start with what the grid actually measured
A grid repeats a location-based check across a set of points. A 3 × 3 grid has nine points; a 7 × 7 grid has 49. The number at a point describes the listing's position in that provider's captured result set, not a permanent rank for everyone nearby.
Keep Google Maps results, Google Search local-pack results and organic website positions separate. Being third in a Maps result set does not establish that the business appeared in the three-listing pack on a customer's Search page. If those distinctions are unfamiliar, start with how local SEO works across Maps and website results.
Before accepting a scan as the baseline, save:
- Exact query, language, engine, result surface and device or collection mode.
- Business identifier, listing URL and branch details. A shared brand name is not enough for a multi-location client.
- Every point's coordinates, grid center, spacing and distance units. Save the actual points rather than relying on an ambiguous radius label.
- Scan timestamp and time zone, checked result depth, provider and report URL or export.
- The client decision, included customer area and treatment of failed, skipped and not-found points.
Download the scan brief to keep those settings with the export. It is a manual planning document, not a software import format.
Choose points that cover the client's real customer area. If a point falls in water or outside the agreed market, document the exclusion before comparing results. Keep an expanded exploration grid separate from the stable monitoring grid. Tightening the map until it looks greener changes the question.
For vendor selection and scan budgets, use the local rank tracker comparison. This guide addresses what to do after collecting a scan.
Read positions and missing states before colors
A color is a shortcut supplied by the tool. It needs a legend and a result depth. BrightLocal's new-platform Local Search Grid guide currently separates positions 1–3, 4–10, 11–20 and over 20. It defines a dash as outside its top 100 and uses 101 when calculating that report's average. That rule belongs to that platform version.
Local Falcon's scan-report documentation describes a 20+ marker. Do not carry BrightLocal's depth or average calculation into a different provider's report. Confirm the definition beside the export you are using.
| Recorded state | What you can say | What remains unknown |
|---|---|---|
| Position 3 | The matched listing was third in the captured result set at this point. | Whether every customer saw it there, or contacted the business. |
| Not found through position 20 | A completed check did not return the matched listing within positions 1–20. | Its exact position below that depth. |
| Error or skipped | No usable observation was collected for that point. | The listing's position. |
| Different branch matched | The result belongs to another location. | The intended branch's position until identity is resolved. |
Never turn a failed check into position 21, zero, or a red “loss.” Never silently drop not-found points to make an average improve. If a provider substitutes a number, preserve that rule with the score. Our worked example below avoids a substituted average altogether.
Worked example: compare a nine-point grid fairly
Fictional agency example. An agency monitors one plumbing branch for “emergency plumber” using a 3 × 3 grid, 1 km between neighboring points, the same listing identifier and Google Maps collection mode. Both scans check through position 20. Dates A and B are illustrative comparison labels. These are invented observations, not scans of a real business.
Each cell shows A → B at the same named point. North is at the top. NF20 means a completed check did not find the branch within 20 results; Error means the check failed.
| Grid row | West column | Center column | East column |
|---|---|---|---|
| North | NW: NF20 → NF20 | N: 8 → 5 | NE: 12 → 7 |
| Middle | W: 3 → 2 | C: 2 → 2 | E: 6 → 3 |
| South | SW: NF20 → NF20 | S: 3 → 2 | SE: 9 → Error |
Download all 18 example observations to reproduce the counts. Compass labels are schematic point identifiers; a real comparison also needs saved coordinates.
There are nine usable observations in A and eight in B. A has three top-three positions out of nine usable points (33.3%). B has four out of eight (50%). Those percentages use different point sets, so do not present their difference as a fair improvement.
For a matched comparison, set aside SE in both periods and disclose why. Across the eight common usable points, A has 3/8 top-three positions (37.5%); B has 4/8 (50%). That is one additional sampled point in the top three, or 12.5 percentage points. E crossed from sixth to third. Two points remained NF20. Recheck SE; its failure is not a loss or a recovery.
This is top-three point coverage, with each included point weighted equally. It is not the percentage of residents reached, geographic area covered, market share, or lead growth. A point in a dense commercial neighborhood and one in a sparsely populated corner contribute equally to this simple calculation.
Compare settings before interpreting movement
Put the two briefs beside each other. A new radius, shifted center, different branch, changed keyword or deeper result limit can create apparent movement. If the setup changed materially, label the new scan a new baseline. Where a subset remains genuinely comparable, identify those exact points and report that subset separately.
Keep a neutral record of implemented work and outside changes between scans. Record holidays or changed opening hours when relevant. Do not pick only the best-looking query or the most favorable date for the client slide.
If a manual search disagrees, compare its query, location context, surface and time with the saved scan. Inspect the listing identity and captured businesses at that point. An incognito window or a VPN does not establish that you reproduced a customer's exact location and results. BrightLocal also explains that different grid tools implement collection differently. Preserve parallel baselines when changing providers instead of splicing their scores into one trend.
Turn a pattern into an investigation, then a task
Google describes local ranking through relevance, distance and prominence. That explains why a listing can perform differently across an area. It does not make a red outer ring proof of missing links, insufficient reviews, a wrong category, or a penalty.
Use the pattern to decide where to collect evidence:
| Observation | Check next | Action only if confirmed |
|---|---|---|
| Strong near the branch, weaker farther out | Open results at representative weak points; compare the actual nearby businesses and service fit. | Correct a verified mismatch or improve useful service information. Do not promise a larger ranking radius. |
| Weak around the branch for a genuine core service | Confirm the intended listing, actual category, business details and result surface. | Fix inaccurate details with the owner's approval; document the change. |
| One service query weak, another strong at the same points | Check what the queries mean and whether the business and linked page actually serve both needs. | Use the local keyword research and page-mapping process before adding or rewriting pages. |
| A sudden broad drop | Follow the Google Maps ranking-drop diagnosis to compare nearby and outer points, report settings and listing availability. | Restore a valid measurement or investigate an evidenced issue with the agency audit checklist. |
Fictional two-branch example. Two dental practices share a brand name. The agency finds that its east-side scan matched the west-side listing. The first task is to correct the tracked identifier and establish a valid east-side baseline. Rewriting service pages from that faulty grid would answer the wrong problem.
For a real content issue, keep an action record such as: “The emergency-service page omits the actual out-of-hours process. Owner confirms service availability; writer adds accurate instructions; account manager verifies the live page and working contact route.” Save the evidence, owner, due date and closure test. A later ranking change is a new observation, not proof that the edit caused it.
Send the client a useful grid note
For the fictional plumbing example, a defensible note reads:
On eight comparable sampled points, top-three point coverage moved from 3/8 to 4/8. The east point moved from sixth to third; two points stayed outside the checked top 20. The southeast follow-up failed and is excluded from both periods in this comparison. We will recheck it and inspect the weak-point results before proposing changes. This does not measure leads or establish the cause of movement.
Put that note alongside the original export and action record. Use the monthly local SEO reporting template to connect it with separate website, Business Profile and confirmed enquiry data.
For ongoing organization, review OpenLocalSEO's rank-tracking workflow. We publish OpenLocalSEO; retain the original provider evidence alongside the campaign record rather than relying on a summary alone.
Questions agencies ask about ranking grids
What grid size and spacing should I use?
Choose them for the customer area and decision. A small grid can reveal differences near one branch; wider coverage may help investigate another area. More points are more observations, not automatically better evidence. Keep a stable baseline and document any expansion separately.
Does an all-green grid mean the campaign is finished?
It means the recorded positions meet that tool's green threshold for the chosen query and sampled points. Check whether the map covers relevant customers and whether enquiries reach the business. A tightly cropped grid can look excellent while answering a very limited question.
How often should an agency run a scan?
Choose a cadence that supports client decisions and the scan budget. Keep it consistent enough for comparison, and use a targeted recheck to investigate an error or significant change. There is no universal daily-scan requirement or promised time to improve.
Can I use the free Maps checker to generate this grid?
The free Maps result checker matches a business inside a list you supply and preserves supplied positions. It does not search Google, verify the observation location, or scan multiple points. Use it to inspect a saved list, with the query, branch and collection context recorded.
