A 7×7 grid scan costs 49 credits per keyword, per run. Most operators buy a rank tracker before they have decided any of the three numbers that determine whether the data is worth reading.
That is the problem with this category. Every buyer’s guide on page one hands you eight tools, a price column and a verdict. None mentions that the tool matters far less than how you configure it, and that a badly configured grid does not merely waste money, it produces a chart that actively misleads you about your own restaurant.
We run weekly grid scans on live restaurant locations and publish the baselines before touching anything, so this comes from the measurement side rather than the sales side. Here is what a Maps rank tracker actually does for a restaurant, what it costs once you multiply it out, and the restaurant-specific trap that invalidates a large share of the grid data operators are looking at today.
What a Maps rank tracker actually measures
Your Google Maps ranking is not a number. It is a surface.
Search “restaurants near me” from a table in your dining room and you might be first. Walk four blocks north and you might not appear at all. Proximity is the strongest input in local search and it is measured from wherever the searcher stands, so a tool that says “you rank number 4” has told you almost nothing, because it has not told you where.
One practitioner puts it plainly in a widely watched walkthrough: “if you registered your Google Business Profile at your address, I’ll make one up, 1010 Main Street, you’re ranking at that address when you type in plumber near me. However, when you drive a couple of miles and you go to a friend’s computer or you check it on your cell phone somewhere else, you realize that you’re not ranking anymore.”
A geo-grid tool solves this by sampling. It lays a lattice of points over a map centred on your restaurant, runs your keyword as a simulated search from each point, and records the position you held at each one. The output is the familiar heat map: green where you place well, yellow where you slip, red where you are nowhere.
Two headline numbers usually come out of it. Coverage (some tools call it visibility, or share of local voice) is the share of grid points where you appear at all, or appear in the top few. Average rank is your mean position across every point sampled.
One caveat almost no buyer’s guide mentions: grid tools generally sample the Google Local Finder, not the three-pack in ordinary search results. As one experienced local SEO put it, “the local pack extension in search is usually very similar but not always exactly the same as local finder results.” Close, usually. Identical, no. Treat the grid as a high-quality proxy, not a literal readout of what guests see.
The two headline metrics move in opposite directions
Here is the part that broke my assumptions, and it comes from our own published data rather than a vendor deck.
We track three live restaurant locations in Brampton, Ontario, on the same grid configuration and the same weekly scan cycle, with every baseline published before we touched anything. If coverage and average rank were two views of the same underlying performance, they should rise and fall together.
They do the exact opposite.
Read the two columns against each other. Coverage climbs steadily, 53.1 percent to 73.1 to 76.3. Average rank gets worse at every single step, 5.09 to 7.77 to 8.2. The location covering the most ground holds the weakest positions on it, and the pattern holds across all three without exception.
That is not a coincidence, it is a trade. A restaurant either dominates a tight core, appearing at fewer points but sitting second or third where it does appear, or it spreads thinly across a wide area, showing up almost everywhere at sixth or eighth. New York Sliders East is the first shape. Railway Chai House is the second. Neither is wrong. They are different businesses, and they should be optimised differently:
Pick one before you start optimising. Anyone reporting whichever of the two happened to improve this month is telling you a story, not a metric, and you can only spot the difference if you decided in advance which one you are paid on.
Scan design: three dials, three distortions
Grid size, radius and cadence are all chosen by you. Each one bends the answer, and only one of the three is really about honesty.
Radius is the honesty dial
This is the one nobody selling you a tool wants to discuss. Tighten the radius and your map goes green. Widen it and the identical restaurant, on the identical day, looks broken. Nothing about your performance changed. You changed the ruler.
A local SEO said this out loud in a thread where someone was celebrating an all-green grid: “set the distance steps between pins closer and your grid will look green, too.” And, more bluntly, about what the chart is really showing: “To hell with what the grid chart says! All it’s doing is representing how proximity affects competition. You’ll always run into a wall at a certain point where you just aren’t going to rank because there are more competitors with better proximity.”
That is the honest description of a geo-grid. Past a certain distance you are not going to outrank a restaurant that is simply closer to the searcher, because proximity dominates. A grid that is all green usually means the radius is too small to contain a real competitive question.
So set the radius to your genuine trade area, the distance guests actually travel to you, and then never move it again. The absolute number matters less than the fact that it stayed fixed. A trend line across a moving ruler is not a trend line.
The corollary: be sceptical of before-and-after grid screenshots in a pitch deck. A practitioner managing SEO across more than 10,000 locations called himself “slightly doubtful” of the dramatic red-to-green transformations circulating in the industry. You need not believe the worst to notice that the person choosing the radius is also the person judged on the result.
Grid size is the resolution dial
More points mean finer detail, and every point costs money. A 5×5 grid gives you 25 samples, enough to see the shape of your reach in a dense urban catchment. A 7×7 gives 49 and suits a suburban restaurant with a wider draw. A 9×9 is a reasonable one-off baseline, which is what Local Falcon’s own team suggests: “Personally I like to start with the 9×9 grid and a fairly tight radius in order to get an idea of the baseline (2 miles or-so). Once I’ve got that dialed for each query, I put the best ones into a Campaign and run it on a schedule.”
Beyond that you are buying precision you cannot act on. On the Local Search Forum a Los Angeles photographer described running “a 21 x 21 grid with a 30 mi radius” and asked, reasonably, whether that was right. It is 441 points per keyword per scan. We price that out below.
Cadence is the noise dial
Daily scanning feels rigorous and mostly measures Google’s own churn. Local rankings wobble hour to hour for reasons unconnected to anything you did. Weekly is the shortest interval at which your change is visible above that noise, and it is what we run. Same day, same hour, every week. Which brings us to the trap specific to restaurants.
What tracking actually costs, multiplied out
Every listicle quotes the sticker price and stops. Local Falcon from $24.99 a month. BrightLocal from $39. Both true, both nearly useless, because grid tools bill on consumption and the sticker tells you nothing until you have chosen your three dials.
The model is simple once someone states it. A member of the Local Falcon team spelled it out in a community thread: “One grid point is one credit. So if you have a 7×7 it’s 49 credits and if you added two keywords/queries that’s two searches of a 7×7 grid, so a total of 98 credits. It adds up fast but that’s why the credits are so affordable. You get 7,500 credits for 25 dollars.”
An independent user verified it with a live test: “One location, 11×11 grid. 1 keyword is 121 credits per scan. 2 keywords is 242 credits etc. If I want it to generate an some AI analysis on each scan then thats an additional 25 credits per kw.”
Note that last detail, because no comparison page mentions it. AI analysis is charged on top, per keyword, per scan.
So: credits equals grid points, times keywords, times scans per month. Here is that arithmetic at six realistic configurations, assuming six tracked keywords per location, about right for a restaurant covering its cuisine term, two or three occasion terms and a neighbourhood variant.
The finding surprised me, and it runs against the direction every affiliate listicle pushes you.
For a single restaurant scanning weekly at sane settings, the entry tier is not just adequate, it is enormous overkill. A 5×5 grid, six keywords, four scans a month is 600 credits. The $24.99 plan includes 7,500. You are using eight percent of it. Spread across the plan, proper weekly grid tracking for one restaurant costs roughly two dollars a month in credits. The subscription floor is what you are paying for, not the scanning.
The cost only becomes real in two situations, and neither is the one the upsell pages warn you about. Over-spec the grid and it bites immediately: that 21×21 configuration burns 10,584 credits a month for a single location and blows past the entry plan on its own. Or scan daily, and a modest 7×7 on six keywords hits 8,820 and does the same, while buying you mostly noise.
The practical read: buy the cheapest plan, resist the dense grid, and spend the difference on what the grid is diagnosing rather than on more measurement of it.
The restaurant-specific trap: your scan time is a variable
This is the section I most want operators to take away, because it silently corrupts a lot of restaurant grid data and no competing guide mentions it.
In November 2023 Google began factoring whether a business is currently open into local pack placement, documented at the time by Joy Hawkins and the Sterling Sky team. For a 24 hour business this is irrelevant. For a plumber it barely registers. For a restaurant that opens at 5pm it is enormous, because you are closed for most of the clock and “open now” filters exactly the high-intent queries you care about.
Which means the time your scan runs is not a scheduling detail. It is an input to the result.
In our own scan logs, the same location measured inside service hours and again in the small hours of the morning differs by roughly eleven coverage points. Same restaurant, same week, same grid, same keywords. The only thing that changed was whether the doors were open when the simulated search fired.
Now consider how tracking usually gets set up. A scan is scheduled once, often on an overnight cron because servers are quiet then, and runs at 2am forever. That scan measures a profile Google is actively demoting for being closed. It stays internally consistent, so the trend looks plausible, but the absolute number is depressed and any comparison against a competitor scanned at a different hour is meaningless.
Three rules follow, and they cost nothing:
- Schedule scans inside service hours. Pick a slot that represents the trade you actually want, a Friday evening for a dinner house, a weekday late morning for a brunch spot.
- Only ever compare open to open. A week-over-week delta is only real if both scans ran at the same hour on the same weekday.
- If you run multiple locations with different hours, they need different schedules. One global scan time across a portfolio guarantees that some of your locations are being measured while closed, and you will misread that as underperformance.
Check what time your scans run before interpreting a single one of them. It is the cheapest accuracy win in local restaurant SEO, and most operators are failing it without knowing.
The shortlist, priced honestly
Local Falcon is the category reference and the right default if grid tracking is the job. From $24.99 a month for 7,500 credits up to $199.99 for 63,000. Deep grid features, competitor reports, scheduled campaigns, and you can scan a competitor without access to their profile. One user’s verdict: “Localfalcon UI ugly as hell… Functionally they all work the same via proxies to run queries.” The second half of that sentence is the important half.
BrightLocal starts at $39 a month with a 14 day trial and no card required. Its grid product is less specialised than Local Falcon’s, but it bundles citation tracking, review monitoring and reporting. If you want one login for local SEO generally and can accept a slightly weaker grid, this is the trade.
Whitespark, Localo and Local Viking round out the practical set. Local Viking (sold as Local Brand Manager) is worth a look if you want posting and review tools attached; one user stayed with it for that reason alone despite preferring Local Falcon’s visuals.
The free option is real, if unglamorous. Search your keyword in an incognito window from several points in your trade area and note where you land. One walkthrough recommends exactly this: “if you don’t want to invest in a tool, you could simply just search your keyword… and just note where I saw myself ranking in the maps.” Tedious and imprecise, but for a single location that only needs to know whether things are moving, better than nothing.
Restaurant Velocity includes a local rank grid and competitor tracking as two of its eight workflows, alongside review replies in your own brand voice, Instagram and Facebook comment replies, weekly Google posts, a scored and captioned photo library, a 33 factor profile audit, and Manual, Assist or Auto control modes you set per workflow. It is $50 per location per month, a founding price for the first 50 customers that stays locked for as long as you stay, moving to $99 per location afterwards. Every workflow is included at every location, there is no contract, and the 14 day free trial means nothing is charged until day 15.
Now the honest placement, because you should not take a recommendation from someone who will not tell you when to buy something else. If you are an agency tracking grids across many clients and many industries, buy Local Falcon. It is purpose-built for that, its credit model scales cleanly across unrelated verticals, and our app is not an agency grid tool. We built the grid for restaurant operators who want the measurement attached to the work that responds to it, not for practitioners who need a dedicated instrument. Different jobs.
Start your 14-day free trial if you want the grid and the fixes in the same place. Buy the dedicated tool if measurement is the whole job.
What to actually do when the pins go red
A grid diagnoses. It does not treat. Most operators stall here, because the chart shows where you are losing without saying why, and the obvious response, turning red pins green, is often the wrong goal.
Proximity dominates local ranking and you cannot change your address. Some red pins are permanently red, and chasing them is how marketing budgets die. The useful question is not “how do I go green everywhere” but “where am I losing ground I should be winning,” which usually means the middle ring: close enough to compete, far enough to be contested.
The loop we run is four weeks long and deliberately boring:
- Baseline. One scan, correct radius, inside service hours. Write down both metrics, coverage and average rank, and which one you are optimising.
- Change one thing. Categories and attributes, review velocity and reply rate, photo freshness, posting cadence, or the on-site content that supports the term. One at a time, or you will not know what worked. Our guides to local SEO for restaurants and optimising a restaurant Google Business Profile cover the levers in order.
- Wait, then re-scan open to open. Same weekday, same hour. Four weeks is usually enough for profile-side changes to register.
- Keep or revert, then repeat. Log every change with its date so the trend line has an explanation attached.
Two unglamorous levers move the middle ring more reliably than anything else in our experience. Review velocity and reply rate feed prominence directly, which is why how you handle review replies shows up in grid data weeks later. And profile completeness, the categories, attributes, hours and photos most restaurants set once in 2019 and never revisit, is the cheapest correction available. To automate the loop rather than run it by hand, that is what the automated Google Business Profile tool stack is for.
One last thing. Keep the bad weeks. We publish every snapshot in our live case-study journals, including the weeks where the numbers went the wrong way, because a record that only contains good weeks is a brochure. The value of a rank tracker is the honest series, not the best screenshot in it.
If the measurement and the fixes belong together for you, Start your 14-day free trial of Restaurant Velocity and the first baseline scan runs in about five minutes.
Frequently asked questions
How many credits does a 7×7 grid use?
49 credits per keyword, per scan. One grid point equals one credit, and each keyword is a separate pass over the whole grid. Two keywords on a 7×7 is 98 credits. On a weekly cadence with six keywords, that is 1,176 credits a month for one location, which fits comfortably inside a 7,500 credit entry plan.
What grid size and radius should a restaurant use?
For a dense urban restaurant, a 5×5 grid over roughly a 1.5 mile radius. For a suburban location with a wider draw, 7×7 over about 3 miles. Set the radius to the distance guests genuinely travel to you, then leave it fixed forever. Changing the radius changes the result without anything about your performance changing, so a moving radius destroys the trend line.
Does the time of day affect my Google Maps grid results?
For restaurants, yes, significantly. Google has factored whether a business is currently open into local pack placement since November 2023. In our own scan logs the same location differs by roughly eleven coverage points between a scan run during service hours and one run overnight. Schedule scans inside your service hours and only compare scans taken at the same hour on the same weekday.
Should I optimise for coverage or average rank?
Whichever one your revenue depends on, and decide before you start. Walk-in and tight-radius delivery restaurants should optimise average rank in the core, because appearing eighth two miles away books nothing. Wide-delivery, catering and destination restaurants should optimise coverage. The two genuinely move in opposite directions. Across three Brampton locations we track weekly, coverage climbs 53.1 to 73.1 to 76.3 percent while average rank worsens 5.09 to 7.77 to 8.2 at every step.
Is Local Falcon worth it for a single restaurant?
At $24.99 a month, if you will actually use the data, yes. A single restaurant on sensible settings consumes about eight percent of the entry plan’s credits, so you are paying for the subscription floor rather than the scanning. The caution is not cost, it is action: a tracker you check and never act on is an expensive chart. If you want the measurement attached to the work that responds to it, an app that does both is the better buy.
Can I track my Google Maps ranking for free?
Roughly, yes. Search your keyword in an incognito window from several points around your trade area and note where you appear. It is imprecise and time-consuming, and it will not give you a trend line or a competitor comparison, but for a single location that only needs to know whether things are moving, it is better than nothing. Repeat it at the same time of day each week or the readings are not comparable.
Why do grid results differ from what I see when I search?
Two reasons. First, grid tools generally sample the Google Local Finder rather than the three-pack shown in ordinary search results, and the two are similar but not always identical. Second, you are almost certainly searching from inside or near your own restaurant, which is the single most favourable point on the entire grid. Your personal search result is the best-case sample, not the average.
