Search Grid Methodology

Why a single rank number is not a measurement

Distance is a local ranking factor, so results differ by where the searcher is standing. A firm can be first at its own office and fifteenth four miles away. Checking rank once, from one place (usually the office) produces the most flattering number available and tells a firm almost nothing about whether prospective clients across the city can find it.

The method

1. Define the grid. A lattice of geographic points across the firm's service area.
2. Define the query set. The charge-specific terms the firm actually needs, in local statutory vocabulary. See criminal defense keywords.
3. Scan before work begins. Every query from every point, producing the baseline surface.
4. Do the work.
5. Rescan the identical grid: same points, same queries, and compare.

The discipline is in step five. Changing the grid, the query set or the radius between scans makes the comparison meaningless, and is the easiest way for a report to show improvement that did not happen.

The published scans

SBD Marketing publishes six markets measured this way, all in windows under 90 days:

How to read “total grid performance gained”

The headline figure aggregates movement across the tracked keywords over the whole grid. It is a measure of visibility surface, not of traffic, calls or cases. A firm whose grid performance rose 316 per cent became visible across substantially more of its city for the terms it cares about. Whether that converted into signed matters depends on the firm's intake, its reviews, its fees and its competitors: none of which a grid measures.

How often to scan

Scanning too often produces noise and scanning too rarely misses the movement that mattered. Monthly is the cadence that suits most defense engagements: frequent enough to catch a decline while it is correctable, infrequent enough that ordinary volatility averages out between readings.

Two exceptions are worth making. A baseline scan belongs at the very start, before any work, and a firm that skips it has given up the ability to demonstrate anything later. A scan also belongs immediately after a significant change, such as an office move or a profile suspension, because those produce real discontinuities and the before-and-after is the evidence.

The rescans behind the published case studies were taken inside ninety days of their baselines, which is stated with each figure so a reader can judge what the window supports. A ninety-day window shows profile and technical work landing, and it is short for content work, which is a limitation of those figures and is described as one.

What the grid replaced

Before grid tools, local visibility was reported as a list of keywords with a position beside each. That format is still the most common thing a law firm receives, and it is close to uninterpretable for local queries, because a single position implies a single answer to a question whose answer varies by where the searcher stands.

The position in such a report is measured from somewhere, and the report rarely says where. Most often it is the firm’s own address or a city centroid, both of which flatter the result. A firm reading “DUI lawyer: position 2” has learned that the firm ranks second at one unstated point, which is compatible with being absent across most of its market.

The grid replaced this by measuring many points and reporting the distribution. That is the reason the figures published on this network are grid averages with their scan dates and query sets attached, and the reason a firm should treat a keyword-and-position report as a starting question instead of an answer. See SEO measurement standards.

How a scan is actually taken

A grid scan queries the local results from a set of coordinates laid over a city and records the firm’s position at each one. The tooling simulates a searcher standing at each point, which is what makes it able to show a distance gradient that a single check cannot.

The choices that shape a scan are the grid’s size, its spacing and its centre. A seven by seven grid at one-mile spacing covers a different question from a nine by nine at half a mile: the first describes a metro, the second describes a neighbourhood. Neither is more correct, and comparing one with the other is meaningless, which is why the parameters have to be recorded with the result.

Each point returns a position or an absence, and the absences carry as much information as the positions. A grid that is mostly blank at the edges has described the firm’s reach precisely, and a report that averages only the points where the firm appeared has thrown that away. The scans behind the published figures count every point.

The parameters that have to be fixed

A grid comparison is only meaningful where the two scans are the same instrument. Five parameters have to be held constant between the baseline and the rescan, and moving any of them silently produces improvement that never happened.

The grid’s centre point has to stay put, since moving it toward the office flatters every subsequent reading. Its dimensions and spacing have to stay fixed, because a tighter grid over a smaller area concentrates on the ground the firm already holds. The query set has to be the same terms, which is where reporting most often drifts. The scan cadence should be regular, so that comparisons are not made between a peak and a trough. And the scan dates have to be published, because a figure without a date cannot be checked.

These are the parameters this network publishes for its own figures, and they are the ones a firm should ask any agency to state before accepting a grid comparison. An agency that cannot produce the centre point, the spacing and the query list for both scans has not shown a measurement.

Reading a grid honestly

A grid shows a pattern and the pattern carries more information than the average. A firm strong at the centre and absent at the edges is proximity-bound, and the work that helps is content and organic visibility across the metro. A firm middling everywhere is usually short on prominence, and reviews and citations are the lever. A firm strong on one side of a city and weak on the other is frequently looking at a competitor concentrated in that half.

Averages hide all of this, which is why an average alone is a weak report even when it is honestly produced. The published figures on this network are averages because they compress into a table, and the underlying grids are the actual evidence. A firm reviewing its own results should look at the pattern before the number.

Two cautions apply to any grid figure, including the ones published here. A large percentage gain from a low base is a small absolute movement, so a firm going from near-invisible to modestly visible can report a striking percentage. And a grid measures visibility, which is a leading indicator of cases signed and not a substitute for counting them. The client’s own intake numbers remain the measure that matters.

What the grid cannot show

Publishing the limits alongside the figures is the point. A measurement whose weaknesses are stated can be argued with, which is what distinguishes it from an advertisement. See case study methodology and SEO measurement standards.

See also: Case study methodology · SEO measurement standards · Map pack rankings