The local rank tracker that shows the whole map not one misleading average number

A national keyword like 'best noise cancelling headphones' returns essentially one SERP. Google personalizes it lightly by history and device, but every searcher in the country sees roughly the same top ten. Rank tracking that reports one number for that query is honest — one number is a fair summary of one SERP. A local keyword like 'orthodontist' does not work that way. Google's local algorithm weighs proximity as a top-tier ranking factor. Two searchers three miles apart get different local packs. The 'rank' is not a scalar; it is a two-dimensional field across the geography of the service area. Reporting one number for that field is like reporting one number for the temperature of a city — technically defensible, practically useless. The correct unit of measurement is a grid. SeoNexus runs geo-grid scans from real GPS coordinates in a configurable pattern. A 3×3 grid gives nine data points, a 5×5 gives twenty-five, a 7×7 gives forty-nine. Every point runs its own simulated search from that coordinate, records the top twenty results, and stores the position of your business. The output is a heatmap — the exact shape of your local visibility. Red cells at the north edge, green cells at the south, a competitor's storefront pin exactly in the middle of the dead zone. That is the picture you needed, and no single-number tracker will ever produce it. Once you can see the distribution, the questions change. You stop asking 'am I ranking?' and start asking 'why does my south side rank three and my north side rank fourteen?' That is a question with a specific answer — a nearby competitor's newer profile, a service you don't list, a category mismatch — and it is only visible on a grid.

Once you have a grid, you need summary numbers so you can trend the data over time and roll it up across a portfolio. The three numbers everyone in serious local SEO uses are AGR, ATGR, and Share of Local Voice. AGR — Average Grid Rank — averages your position across only the cells where you rank in the top twenty. It is the friendliest number and the most common vanity metric, because cells where you don't rank at all get quietly ignored. You can 'improve AGR' by disappearing from cells you were losing, which is not improvement. ATGR — Average Total Grid Rank — averages your position across every cell, counting non-ranking cells at position 20 or beyond. It is a harder, honest number that gets worse when you fall off the grid and better when you climb into more of it. This is the number an operator should watch, because it cannot be gamed. SoLV — Share of Local Voice — is the percentage of cells in the grid where you appear in the top three (the map pack). Because 44% of local clicks go to the three-pack, SoLV is the number that predicts real-world map-pack traffic more tightly than either AGR or ATGR. If SoLV goes from 12% to 47% over a quarter, calls and direction requests move with it, and you can prove causation. SeoNexus tracks all three per keyword and per location, rolls SoLV up across a portfolio by tag, and lets you set alerts on any one of them. Most rank trackers only track AGR because it looks best in the sales demo. That should tell you something.

A rank tracker that stops at the scan hands the operator a problem, not a solution. 'You are red in these 12 cells' is a starting question, not an answer. The answer requires cross-referencing why — what does the winner have that you don't? Which specific profile fields, categories, services, attributes, review signals, or citation gaps explain the delta? Local Falcon and similar tools stop at the grid. They give you a beautiful heatmap and leave the diagnosis as an exercise for the reader. That is fine for the SEO consultant billing by the hour to interpret it. It is bad for the operator who needs to fix things this week, and worse for the agency scaling across dozens of clients where nobody has time to hand-diagnose every red cell. SeoNexus is built the other way. Every geo-grid scan is joined to the profile audit for that location — a 40+ point check across categories, services, attributes, hours, photos, post cadence, review velocity, Q&A, and suspension risk. When a cell is red, the platform surfaces the plausible causes ranked by likely impact: 'winner in this cell has 340 reviews, you have 84,' 'winner has category Optometrist plus Eye Care Center, you have only Optometrist,' 'you have no service listed for pediatric eye exams and 4 of 6 winners in this cell do.' The operator gets a fix list, not a puzzle. The client gets a report that reads 'here is what changed, here is what we did, here is the rank movement it produced.' That loop — scan, diagnose, act, re-scan — is the difference between a rank tracker and a rank-improvement platform.