A San Diego law practice went from 10 profile visits a month to 81. Here’s the whole mechanism.

Search for your own business from your own office and you will probably rank well. That tells you almost nothing about what a customer three kilometres away sees.

A San Diego criminal defence attorney found this out the direct way. Criminal defence in a major US metro is one of the harder local search environments there is: competing firms have decade-old profiles, hundreds of reviews, and marketing budgets a single practitioner cannot match. This client's reputation in court was well ahead of his visibility on the map, a common and expensive gap. This is the real, already-published record of what changed, with the caveats included.

What a ranking screenshot doesn't show you

A screenshot taken from a business's own address proves the business can find itself. It does not prove a stranger three neighbourhoods over can. Google's own guidance is explicit that local ranking runs on three factors, relevance, distance and prominence, and that there is no way to pay for a better position.¹ Distance is exactly the factor a flattering screenshot hides.

Before touching anything, we ran a baseline geo-grid scan, coverage measured across the metro rather than from one address, so every later claim could be checked against a real starting point. The opening weeks of the window were the slowest part of the curve. February showed minimal interaction: the profile was live but effectively invisible. That is the honest shape of local SEO. Category and profile work land first, prominence built on reviews and citations compounds behind it, and the visible payoff tends to arrive in weeks six through twelve, not week two.

What we actually changed

Nothing on this list is exotic. That is close to the whole point.

• Primary category and services. The single strongest relevance field on a profile, and the first thing corrected, using Google's predefined service entries rather than free text.
• Service area and hours accuracy. Being open at the moment of the search is now rated among the top five Local Pack factors, and legal searches skew toward evenings and weekends, when somebody has just been arrested or charged.
• Baseline geo-grid scan, then recurring scans. Coverage measured before anything was touched, so every later claim could be checked against it rather than taken on faith.
• NAP and citation cleanup across legal directories, where attorney listings are unusually prone to old addresses and stale tracking numbers.
• Review cadence and responses. The client committed to asking, consistently, and to responding to everything that arrived.
• Schema and entity clarity in the final weeks, so assistants could verify the practice, its location and its practice areas independently of the profile itself.

The same signals now decide whether AI assistants mention you

AI assistants do not read advertisements. They read structured, verifiable facts published across the web, and they weigh third-party citation consistency more heavily than a business's own claims about itself.² Category accuracy, consistent citations and clear entity data are not a separate AI-search project bolted on afterward, they are the same foundation the Local Pack position was already built on. More on AEO and GEO →

What we'd flag if we were grading our own work

Calls did not rise smoothly. There was a real dip in the middle of the window before April's jump to 18, which we did not predict and which took a cycle to diagnose. Three months is also a short window, and it spans a seasonal pattern legal practices genuinely experience, so some portion of the movement is not attributable to us. We are reporting profile interaction data because that is what we can evidence. We are not claiming a revenue figure we did not measure, and we would not trust a case study that did.

If you want the mechanism without the full engagement

The category, hours, citation and review work above ran as a full 12-week roadmap. Not every business needs that. If the specific gap is proving distance, dated evidence that a business is actually active in the neighbourhoods it wants to rank in, Map Check-Ins is the productized version of that one mechanism: field photos, timestamped and pinned to a location, published to the site and the Google Business Profile as jobs happen.

The honest part

Nobody can guarantee a top spot on Google Maps, including us, and a case study that implies otherwise is not one to trust. What we can do is show the actual mechanism and the actual numbers, slow start included, and let the shape of the curve speak for itself.