Why this exists

Most local SEO tools answer a question the business did not ask.

They report a keyword count, a field-completion percentage, a backlink total, or a rank for a query that was chosen after someone looked at the results. The business owner is left with a dashboard and no defensible answer to the only question that matters:

When a customer nearby needs the work I sell, am I visible?

Local Visibility Scorecard exists to answer that question with evidence.

Three failures the market usually blurs

A local business can fail discovery in at least three different ways:

1. Visibility failure — the business does not appear in the Local Pack, Maps, organic results, or AI answers for the demand it sells.
2. Profile failure — the business appears, but the Google Business Profile does not create trust or enable action.
3. Conversion failure — the customer reaches the owned website, but the site does not convert.

Treating all three as "SEO" produces bad advice. A business with a strong profile and a weak Local Pack presence may be told to add photos. A business with strong Maps visibility and a weak website may be told to publish more location pages. Neither diagnosis follows from the evidence.

The scorecard family separates the three:

LOCAL VISIBILITY
Can they discover you?
        ↓
GBP
Do they trust and act?
        ↓
WEBSITE
Do they convert?
        ↓
RAD
Remediation and operations

What this scorecard does differently

It measures outcomes, not activity

A check earns points only when an observable visibility outcome changes. "Published 12 pages" is not a score. "The business appears in the top 10 for 4 of 6 commercially important service clusters" is.

It locks the query universe before looking at results

Choosing keywords after seeing the rankings is the easiest way to manufacture a flattering score. This framework defines the market, services, intents, demand weights, and AI prompts first, hashes the universe, and refuses to score a run whose universe changed.

It separates visibility from resolution failure

If the target business cannot be resolved reliably, the run returns INSUFFICIENT EVIDENCE. It does not turn an identity-resolution failure into "rank not found" and penalize the business.

It treats AI answers as a real but different surface

AI answers are not one thing. A Google AI Mode answer, a ChatGPT model response, a citation in a source list, and a directory mention are different observations. The framework records them separately, repeats the observations, and never pools consumer/search surfaces with model/API responses.

It keeps owner data out of the public score

Search Console clicks, qualified leads, and revenue are not public. They are reported as owner-data questions. They never lower the public score.

It stays independently useful

The recommendation mappings are metadata. The scorecard does not require RAD, a CRM, or a paid engagement to be useful. RAD may later consume the machine-readable output; it does not rewrite the score.

What it refuses to do

• Claim that a visibility gap caused a specific lost job.
• Invent a revenue impact from a ranking change.
• Treat a directory as a contractor competitor.
• Treat a branded search as evidence of non-branded discovery.
• Reward raw backlink quantity.
• Turn a provider outage into a low business score.
• Publish a real client's query set, grid coordinates, prompts, or report.

The goal is not more checks. The goal is a score another analyst can reproduce from the same evidence and explain to the business owner without hand-waving.

---

Scorecard family boundary

This page covers discovery. The GBP Scorecard diagnoses profile trust and action; the Website Scorecard diagnoses website conversion. Read The Scorecard Family for the ownership matrix and the no-duplicate-scoring rule.