How to read a geo-grid scan, and what to fix first

A geo-grid scan shows where your Google Business Profile appears for one keyword at every point across an area, so the way to read it is as a map of territory, not a single score. Start with the shape: a strong core around your business that fades towards the edges is normal. Then look for the things that are not normal: weak pins right next to you, one side of town owned by a single competitor, and pins where you are closer to the searcher than a business that beats you. Those are the pins you can actually win, and they tell you what to fix first. The pins where every winner is closer than you are mostly down to distance, and no amount of profile work will reliably flip them.

What a geo-grid actually shows

Local results depend heavily on where the person searching is standing. Google's own documentation names three main things that decide local results: relevance, distance and prominence, with distance meaning how far each business is from the customer doing the search.¹ That is why checking your position from one spot (your office, or the middle of town) gives you one number that can change completely a few streets away.

A geo-grid gets round this by searching from lots of places at once. You choose a keyword, a centre (usually your business) and a grid, for example 7 by 7 points spread across 5 km. Each of those 49 points is its own search, run with the location set to that point. The result is a map of pins, each coloured by where your listing came in from that spot.

On a typical Maps grid the colours run from green (positions 1 to 3, the ones that show in the Map Pack) through yellow and orange to red, with grey for pins where you were not in the results at all. The exact bands vary by tool, but the logic is the same: green is where you win customers, grey is where you do not exist for that search.

How to read the patterns

Most grids fall into a handful of recognisable shapes. Learning to spot them is most of the skill, because each one points to a different cause.

Strong centre, fading at the edges

This is the normal shape, and it is not a problem to solve. You are strong close to your business, and as the searcher moves further away, closer competitors take over. The question is not whether it fades but how quickly. If you are green for 2 km in every direction, that is a healthy footprint for most high street businesses. If the green stops at the pin on top of your building, relevance or prominence is weak and distance is doing all the work.

Holes near your business

A ring of green with a yellow or grey patch a street or two away is the most valuable pattern on the map, because distance cannot explain it. Usually a competitor sits right inside that patch and beats you on their doorstep, or the competitor that wins there simply matches the search better (a closer category match, more reviews, a more complete profile). Open the pin, see who wins, and compare their listing with yours.

One competitor owning one side

If the whole east of the grid is lost to the same business, look at where they are. Very often they are physically on that side, and you are seeing the line where their distance advantage beats yours. That part is not fixable by profile work alone. If they are not closer and still win, that is a relevance or prominence gap, and their profile is your benchmark.

A lopsided grid that does not match your location

If your strongest pins are clearly off to one side of your business, check the basics first: is the scan centred where your listing actually is, and is the map pin on your profile in the right place? A misplaced pin, or a centre set on a postcode rather than your door, can make the whole map look skewed.

Grey almost everywhere

A grid that is grey even beside your premises usually means the search does not match what your profile says you are. A plumber with "Heating engineer" as the only category can vanish for "plumber near me" while doing well for heating searches. It can also mean the scan did not match your listing at all (a different name, or a different address), or that the profile has a problem such as a suspension. Rule those out before you start changing things.

Pins that flicker between scans

Some pins will be green one week and yellow the next with nothing changing on your side. These borderline pins are where small improvements show up first. Pins that are always green or always grey tell you less about what is changing.

Choosing grid size and spacing

The map you get depends on the grid you draw. Too tight and everything is green, which feels good but tells you nothing. Too wide and the outer rings are grey by default, which can make a healthy business look like it is failing.

1. 1

   Match the area to where your customers really come from

   A café in a city centre might draw from a kilometre or two. A plumber with a van covers a whole town or more. Draw the grid over the area you serve, not the biggest area the tool allows.
2. 2

   Pick spacing that matches how the map changes

   In a dense city, results can change every few hundred metres, so tighter spacing shows real detail. In suburbs and towns, a kilometre or so between points is usually enough. In rural areas, wider spacing is fine because there are fewer competing businesses.
3. 3

   Use more points, not a wider grid, when you want detail

   A 7 by 7 grid has 49 points. A 13 by 13 grid over the same area has 169, so each pin covers a smaller patch. More points cost more to run, so save the dense grids for your most important keywords.
4. 4

   Keep the grid the same every time

   The value is in comparison. If you change the centre, size or spacing between scans, a change in the map might just be a change in where you looked. Fix the grid for each keyword and leave it.

If your area runs along a coastline or a long high street, a grid that can be turned at an angle wastes fewer points on the sea or on fields.

Maps, Google Search and AI Mode grids are different things

The same keyword can show different businesses in Google Maps, in the local pack on Google's results page, and in an AI Mode answer. Google's developer documentation notes that AI Mode and AI Overviews may use different models and techniques, so the responses and links they show will vary.² Measure the place your customers actually search, and do not assume one map stands in for another.

AI Mode answers have no fixed order, so a position number for them would be invented. The honest measure is whether you were mentioned or recommended, and at what share of the points checked. If a tool gives you a numbered "place" inside an AI answer, ask what it is actually counting.

How far proximity limits what you can fix

Distance is one of the three main inputs to local results, and it is the one you cannot change without moving.¹ That gives you a very practical way to sort lost pins into two piles.

If your business genuinely serves customers in an area where you have no premises, the answer is not a fake address or a virtual office. Google's guidelines say a service-area business should have one profile for its central location with a service area set, and the address hidden.³ A second profile is only appropriate for a real second location that meets customers.

What to fix first

Once you have sorted your lost pins, work through the fixes in this order. Each step is worth doing on its own, and the early ones are quick.

Two of these come straight from Google's documentation: complete, accurate information and reviews both feed into local results.¹ The category advice follows Google's own guideline to choose the fewest, most specific categories that describe the core business.³

A quick checklist for your next scan

• The grid covers the area you actually serve, not the widest area available.
• The centre sits on your listing's real location.
• You have marked the pins near you that are not green, and opened each to see who wins.
• You have split lost pins into 'closer than a winner' and 'every winner is closer'.
• Your categories match the keyword, compared with the businesses beating you.
• You know the rating and review count of third place at your fixable pins.
• Sponsored listings are not being counted as positions on a Google Search grid.
• The next scan is booked on the same grid, so you can compare like with like.

How SearchOps geo-grid scans work

Everything above can be done by hand, a pin at a time. SearchOps does the searching, the matching and the comparison for you, and keeps the history so that changes mean something.

1. 1

   Draw the grid you need

   Grids run from 3 by 3 up to 21 by 21, covering anything from 1 km to 100 km across, centred on your business or a point you choose. You can turn the grid to follow a coastline or a high street. AI Mode grids go up to 9 by 9.
2. 2

   Pick the engine

   Google Maps reads the top 20 businesses at each point. Google Search reads the local pack and any AI Overview, with sponsored slots tagged as ads and kept out of the position count. AI Mode reads the answer and records whether you were named or your site was cited, never a position.
3. 3

   Read the share, not just the map

   The scoreboard shows the share of points where you make the top 3 (or are named, for AI Mode) with a likely range, and only marks a change as a real gain or drop when it is bigger than the normal swing between runs.
4. 4

   See who wins where, and why you lose points

   Open any pin to see the businesses there, switch to a rival's grid, or view who wins each point. For Google Maps, Geo-Grid Insights sorts your lost pins into those where you are closer than a business in the top 3 and those where every winner is closer, and points out categories the winners use that you do not and the review gap to third place. Findings can go straight onto your Tasks list.
5. 5

   Schedule the same grid and compare

   Schedule any engine daily, weekly, every two weeks or monthly, at the time you choose. Every run uses the same grid, so territory movement shows exactly where you started or stopped making the top 3. Export to CSV or share a link with a client.

The full detail of what each engine reads is on the geo-grid feature page, and scan credits are set out on the pricing page. If you are working through the fixes above, the categories guide is the best place to start.

FAQ

What is a good geo-grid result?

There is no single target, because it depends on the keyword, the competition and how wide you drew the grid. A better question is whether your share of points in the top 3 is growing over repeated scans of the same grid, and whether the pins near your business are green.

Why is my average position misleading?

An average only counts the pins where you appeared, so a business that is first at five pins and absent from forty can show an excellent average. The share of points where you make the top 3 reflects the whole area, including where you are missing.

How often should I run a scan?

For most businesses, weekly or every two weeks is enough to see the effect of changes without reading too much into normal movement. Run it at a time of day when your customers actually search, and keep that time the same.

Why do two scans of the same grid differ when nothing changed?

Local results move a little from one search to the next, and borderline pins can flip either way. That is why a single scan is a snapshot, and why a change should only count as real when it is bigger than the normal swing between runs.

Can I win the pins at the far edge of my grid?

Sometimes, but if every business in the top 3 there is closer to the searcher than you, those pins are mostly decided by distance. Put your effort into the pins where you are closer than a winner, and widen your reach from the inside out.

Does a strong Maps grid mean I show up in AI Mode?

Not necessarily. They are different results and can name different businesses for the same search. Scan the engine your customers use, and read AI Mode as a share of checks where you were mentioned, not a position.

The short version

Read a geo-grid as territory. Expect the fade at the edges, look hard at the gaps near your business, and split your lost pins by whether distance explains them. Work on the pins where you are closer than a winner: categories and relevance first, then the review gap to third place, then a complete profile and wider prominence. Keep the grid the same, scan on a schedule, and only count a change once it is bigger than the normal swing.