Local Map
The Map tab shows one prompt measured at every point of a grid across your market. Each pin carries two separate facts: whether an answer engine named the business there, and where Google ranks it there. It appears on local projects only, and only once a Google Business Profile is attached.
The six pin states
A pin answers one question: did an answer engine name this business here? Which engines answered is shown on the pin as marks — solid where that engine named you, faded where it answered and did not, absent where it never ran. The Google Maps position rides alongside as a number, because it says which kind of problem a miss is, not whether there is one.
The AI named you here
An engine recommended this business by name at this point. Nothing to fix — this is the point worth screenshotting for a client.
Google Maps ranks you, the AI didn’t name you
The listing is strong enough for Google’s map pack, but the AI passed over it. The listing is doing its job; the web presence behind it is not. Content and structure.
Google Maps ranks you below 20
In the pack but far enough down that being named is rare. Google Business Profile — reviews, categories, photos, completeness.
Not in Google Maps here
Google returned a pack at this point and this business was not in it. Usually a service area or primary category that does not cover this point.
Google Maps returned no list here
Google answered at this point with fewer businesses than the three its own local module shows — we have measured markets that return exactly one, with the provider itself reporting zero results. Being absent from a list that short is not evidence of anything, so the map declines to call it a miss.
Not measured yet
No run has produced a reading here. Explicitly not a miss — a point added since the last run, or a run that has not happened. It fills in after the next weekly run.
Reading the summary line
The header reads “Named at 3 of 7 measured points · median rank 12”. The denominator counts points that have a reading, not points you configured — a point with no run behind it is not a place the business failed to be named. If some points have no reading yet, a second line says how many.
Median rank, not average
Skipping the points where the business is not in the pack at all, so one absence cannot drag the number to zero. When the business is in no pack anywhere the rank segment is omitted entirely rather than showing a zero.
Distance rings
Two rings, at half and full the distance to the furthest point, labeled in miles. The center is the centroid of the points, not the first one.
Switching it on
Grid tracking is two switches, not one. The project sets the shape of the grid; each prompt opts into it individually. Turning it on at the project level alone produces no map.
1. Attach the Google Business Profile
Required, and it gates everything else. AI answers name local businesses without linking to them, so a domain match finds nothing — we match on the Google listing instead. Search the name in Project settings and pick the profile.
2. Set the grid on the project
Choose 3×3 (9 points) or 5×5 (25 points) and a radius from 3 to 25 miles. Density and radius are project-wide, so one map never mixes spacings.
3. Opt each prompt in
Your grid is geometry, not a switch: a local project with a business and a market has one, and its radius and density say what shape it is. Whether a prompt uses it is answered on the prompt, where off means one reading at your market. Once your grid has points, new prompts with a market are measured across it by default, priced before you save. The control appears only on prompts that have a market — a Global prompt has one answer, so nine readings would return the same thing nine times. Prompts added before the grid existed keep running once until you switch them over, because gridding them for you would multiply what they cost without asking.
4. Wait for the next run
Points are resolved once when you save, not on every run — so moving the center or changing the radius requires saving again. Readings appear after the next weekly run.
When the map is empty
Five different reasons, and they need different fixes. The third is the one to watch, because it is the only one that does not announce itself on the Map tab.
No Google Business Profile attached
The map refuses to draw at all, and says so with a link to Brand & Market. This is deliberate: without an identity to match on, every pin would read “not in the pack” — which claims the business is invisible in its own town when the truth is that we do not know which business is theirs.
The project is national
The tab says the map is for local projects. Change market reach to local on the Brand & Market tab and give the project a market.
Grid on, but no points resolved
If no point resolves to a real place — a tight radius over water, a remote area — the prompt quietly keeps running as a single city-level reading. The save message says so at the time. If you are on Pro and seeing one reading where you expected nine, this is why; widen the radius and save again.
Fewer points than you asked for
Points landing on water, farmland, or a place the SERP provider cannot resolve are dropped. The map says “2 of your 9 points fall on open land or water” rather than quietly measuring fewer places than you configured.
Configured, but nothing has run
Two messages, depending on which half is missing. With no grid points: “No points measured yet”, with both setup steps named and a link to Brand & Market. With points but no prompt ticked: “Your grid is ready, but no prompt is measured on it yet”, with a link to the prompt list — the one remaining step. Grid changes take effect on the next scheduled run.
Limits worth knowing
Priced below what it runs
A gridded prompt costs more than an ordinary one, because it really is more measurements: a 3 × 3 counts as four tracked prompts and a 5 × 5 as twelve. There is no separate grid allowance — it comes out of the same pool.
Included on every paid plan
Grid tracking is on every paid plan. A plan without it measures every prompt at one point and resolves no grid at all, rather than failing on save, and a downgrade stops the fan-out at the next run rather than leaving it running.
Keep points five miles apart
Below that, neighboring points tend to return the same answer and you are paying for repetition. Widen the radius or drop to 3×3.
Which engines run locally
A local project runs ChatGPT plus Google’s AI Overviews, AI Mode and Gemini. ChatGPT and Gemini are each asked again at every grid point; Google’s map pack is read at every point too.
Set up the business identity and grid on your project first.
Go to Brand & Market