Audit methodology

How VisiGap Measures AI Search Visibility

The specific process behind every VisiGap AI Visibility Audit: what we scan, how we simulate AI queries, how we evaluate 83 parameters, and how we benchmark against local competitors — in 48 hours.

Methodology summary

Every VisiGap audit runs the same four-step process: citation source scan (48–52 sources), AI query simulation (5 queries × 3 engines = 15 responses), 83-parameter diagnostic (schema, content, local, trust, and conversion signals), and competitor benchmarking (top 3 local competitors, same framework). Output: a prioritized gap report with fixes ranked by business impact, delivered within 48 hours.

The four steps

How every audit runs

Citation source scan — 48 to 52 sources

AI engines do not generate local business recommendations from scratch. They ground recommendations in structured data from specific, verifiable sources. VisiGap scans every source in our category-specific library — checking for presence, accuracy, and consistency across each one.

This is the highest-weighted component of the AI Visibility Score (25 points) because it is the primary mechanism AI engines use to verify a business's identity, category, and location before citing it. A business absent from key sources is a business AI cannot verify. A business AI cannot verify is a business it will not cite.

What counts as a gap: Any source where the business is absent, where the business name differs by even one character from the primary record, or where the address or phone number does not match exactly. Partial presence is scored as a gap, not as coverage.

AI query simulation — 5 queries × 3 engines

VisiGap selects 5 queries for each business based on its primary service category and city. Each query is run live across ChatGPT, Google AI Overviews, and Perplexity — 15 total responses evaluated per audit. We record whether the business is cited, whether a named competitor is cited instead, and whether the business appears in any context relevant to the query.

The query set is not arbitrary. It mirrors the actual query patterns AI users ask for local services — derived from verified SERP behavior data across 237,000+ home service queries (WebFX 2025) and legal/healthcare query studies (Whitespark Q2 2025, BrightEdge Dec 2025).

Query structure: One direct service query ("best HVAC company in [city]") + one cost query ("how much does AC replacement cost in [city]") + one comparison query ("HVAC repair vs replacement") + one process query ("how to choose an HVAC contractor") + one local qualifier query ("[city] HVAC contractor reviews"). These five intent types together cover the full customer decision journey.

83-parameter diagnostic framework

Every audited business is evaluated across 83 parameters covering the five signal categories AI engines use to determine whether a local business is worth citing. Parameters are weighted by their impact on AI citation likelihood, not by traditional SEO value.

The framework separates parameters into two tiers: Preview (the 20 highest-impact parameters that drive the headline findings in the report) and Full Audit (all 83 parameters, fully scored and ranked by fix priority). Every $499 audit delivers the full 83-parameter evaluation.

What changed vs. traditional SEO audits: The 83-parameter framework explicitly excludes parameters with no AI citation impact (exact keyword density, noindex pages outside the main site) and includes parameters traditional SEO tools ignore: whether the schema type matches the specific service category (not just "LocalBusiness"), whether FAQ content is present and structured for extraction, whether AI crawl bots are explicitly allowed in robots.txt, and whether after-hours booking is available to capture AI-referred visitors who arrive at non-business hours.

Competitor benchmarking — top 3 local competitors

VisiGap identifies the top three local competitors for each audited business using the AI query simulation results: whichever businesses appear most frequently in AI-generated responses to the simulated queries are the primary competitive benchmark. We then evaluate each competitor against the same seven-component AI Visibility Score framework.

The competitor analysis answers one question: what specific signals does the most-cited competitor have that the audited business does not? That gap — not the absolute score — is what determines fix priority in the report.

Why this matters: A business in a low-competition market can have a moderate absolute score (42/100) and strong relative performance. A business in a high-competition market can have a higher absolute score (58/100) and still be the weakest competitor. Fix priority follows the competitive gap, not the absolute score.
Step 1 detail

The 48–52 citation sources

AI engines ground local business recommendations in structured data from specific sources. VisiGap's source library is organized into five categories. The exact count varies by industry because different service categories have different authoritative source sets.

4
Data aggregators

Neustar Localeze, Data Axle, Foursquare, Factual. These four aggregators supply business data to the majority of secondary directories and are the primary identity verification layer AI engines use.

8–10
Primary directories

Google Business Profile, Apple Maps, Bing Places, Facebook, Yelp, BBB, Nextdoor, Yellow Pages. These are the highest-authority citation sources for local businesses across all categories.

12–18
Industry-specific directories

Avvo and FindLaw for legal; Healthgrades and Zocdoc for healthcare; Houzz and Angi for home services; Psychology Today for behavioral health. These vary by category and carry the highest weight for AI citations in their respective verticals.

16–20
Secondary citation sources

MapQuest, Citysearch, MerchantCircle, Superpages, CityLocal, EZlocal, and approximately 12 others. These reinforce the primary record and increase AI citation confidence through source volume.

6–8
Review platforms

Google Reviews, Yelp, Facebook Reviews, Trustpilot, Birdeye, and 1–3 industry-specific review sources. Review count, recency, and rating all influence AI citation confidence — particularly for healthcare and legal categories.

Step 2 detail

How query simulation works

Example query set for an HVAC contractor in Chicago. Every query set is generated specifically for the business's service category and city. Each query runs live across all three AI engines at audit time — not from a cached database.

Query simulation — HVAC contractor, Chicago IL 15 total responses evaluated
1
"best HVAC company in Chicago" Direct service ChatGPT AI Overviews Perplexity
2
"how much does AC replacement cost in Chicago" Cost / pricing ChatGPT AI Overviews Perplexity
3
"HVAC repair vs replacement — when to replace" Comparison ChatGPT AI Overviews Perplexity
4
"how to choose an HVAC contractor" Process / how-to ChatGPT AI Overviews Perplexity
5
"Chicago HVAC contractor reviews 2025" Local qualifier ChatGPT AI Overviews Perplexity
Why query selection matters

Query 1 ("best HVAC company in Chicago") triggers AI Overviews only 12% of the time — local pack still dominates. Queries 2, 3, and 4 trigger AI Overviews 37–41% of the time. This is the actual risk: AI is now answering the mid-funnel questions your website used to answer before a customer decided to call. If your business is not cited in those AI responses, the customer who was researching before calling you has already moved on to a competitor that was cited.

Step 3 detail

The 83-parameter framework

Five parameter categories, each weighted by AI citation impact. This is not a traditional SEO audit — parameters are weighted by how much they influence AI engine citation decisions, not by how much they affect Google rankings.

Schema & entity signals 26 params
AI citation weight: 30%

The primary mechanism by which AI systems identify, classify, and cite a local business. Without correct schema type, AI cannot determine what service the business provides. Without NAP in structured data, AI cannot verify location. Without entity-specific schema, AI cites the wrong category or not at all.

Key parameters: schema type match (HVACBusiness vs LocalBusiness), service schema presence, NAP in JSON-LD, geo coordinates, opening hours markup, practitioner credentials schema (healthcare/legal)
Content depth & extractability 19 params
AI citation weight: 25%

AI systems cite pages that answer questions well. Thin service pages, absent FAQ content, and missing educational context are the most common reasons a business is skipped in AI answers. FAQ presence alone is a near-decisive signal — 78% of home service businesses have none.

Key parameters: FAQ schema presence, service page depth (word count + specificity), pricing context, educational content, city named in H1 and page title, evidence of real outcomes and case specifics
Local relevance signals 10 params
AI citation weight: 20%

AI local answers depend on the system's ability to confirm that a business serves a specific location. City name in page title and H1 are non-negotiable. Businesses that omit their primary city from key content positions are effectively invisible to location-qualified AI queries.

Key parameters: city in page title, city in H1, city in meta description, primary market explicitly named, service area structured data, local landmarks or neighborhood references
Trust & E-E-A-T signals 13 params
AI citation weight: 8%

E-E-A-T (Experience, Expertise, Authoritativeness, Trust) is an explicit factor in AI quality assessment. For healthcare and legal businesses, named practitioners with visible credentials are required for AI systems to treat the site as authoritative. This weight rises to 15%+ for healthcare and legal categories.

Key parameters: named practitioner with credentials, license number visible, professional association membership, review count and recency vs. top competitor, BBB accreditation, HTTPS and security signals
AI crawlability & booking flow 15 params
AI citation weight: 8% (booking conversion: 30%)

AI crawlers face the same access barriers as Google — a page blocked in robots.txt cannot be cited. Mobile speed and a working booking flow determine whether an AI-generated referral converts to an actual customer contact. A significant share of AI-referred visitors arrive outside business hours.

Key parameters: AI bots permitted in robots.txt (OAI-SearchBot, PerplexityBot, Google-Extended), mobile PageSpeed score, booking flow completable on mobile, after-hours booking availability, form submission working
What you receive

The audit report — delivered in 48 hours

Every VisiGap audit delivers the same set of outputs, formatted for immediate use: you hand it to whoever you already work with, or implement the top three fixes yourself. No follow-on pitch. No retainer required.

AI Visibility Score — 0 to 100

Your composite score across all seven components, with individual component scores shown separately. Compared directly to the top three local competitors in your category.

Citation source gap report

Every source scanned. Status for each: present and accurate, present with errors, absent. Errors show the exact discrepancy — what your record says vs. what it should say.

AI query simulation results

Results from all 15 query-engine combinations. Which responses cited your business. Which cited a competitor. The exact competitor name and why it was cited instead.

83-parameter scorecard

Full parameter-by-parameter evaluation with weighted scores, sorted by fix impact. Each failing parameter shows exactly what is wrong and exactly what to change.

Competitor gap analysis

Side-by-side comparison with your top three local competitors across all seven AI Visibility Score components. Shows the specific signals your most-cited competitor has that you do not.

Prioritized fix list

Three to five specific fixes, ranked by impact on your AI Visibility Score, with implementation steps. Designed to hand directly to a web developer, marketing manager, or agency without additional translation.

Common questions

Methodology — FAQ

VisiGap scans 48 to 52 citation sources in every audit. These include the four primary data aggregators (Neustar Localeze, Data Axle, Foursquare, Factual), 8–10 primary directories (Google Business Profile, Apple Maps, Bing Places, Facebook, Yelp, BBB, Nextdoor, Yellow Pages), 12–18 industry-specific directories relevant to the business's service category, 16–20 secondary citation sources, and 6–8 review platforms. The exact count varies by industry because different service categories have different authoritative source ecosystems.
VisiGap selects 5 queries for each business based on its primary service category and city, covering five intent types: direct service, cost/pricing, comparison, process/how-to, and local qualifier. Each query is run live across ChatGPT, Google AI Overviews, and Perplexity at audit time — 15 total responses evaluated. We record whether the business is cited, whether a named competitor appears instead, and whether the business appears in any context relevant to the query intent.
VisiGap's 83-parameter framework evaluates every technical and content signal that determines whether an AI engine can identify, classify, retrieve, and cite a local service business. Parameters are organized into five categories weighted by AI citation impact: schema and entity signals (30%), content depth and extractability (25%), local relevance signals (20%), trust and E-E-A-T signals (8%), and AI crawlability and booking flow signals (8%). The remaining 9% is accounted for by the Competitor Citation Gap and Fan-Out Coverage components of the AI Visibility Score.
VisiGap identifies the top three local competitors from the AI query simulation results: whichever businesses appear most frequently in AI-generated responses to the five simulated queries. Those competitors are evaluated against the same seven-component AI Visibility Score framework. The gap between the audited business and the most-cited local competitor — broken down by component — determines which fixes are prioritized in the report.
An SEO audit measures your website's ability to rank in traditional Google search results. A VisiGap AI Visibility Audit measures your business's ability to be cited in AI-generated answers. The parameters are different, the benchmarks are different, and the fixes are different. A business can score well on a traditional SEO audit and still have an AI Visibility Score below 35 — because the signals AI engines use to decide what to cite are fundamentally different from the signals Google uses to decide what to rank. Most SEO tools do not measure AI citation signals at all.

Get your business's AI Visibility Score

VisiGap runs the full four-step methodology on your business: 48–52 sources scanned, 5 queries simulated, 83 parameters evaluated, 3 local competitors benchmarked. Prioritized fix report in 48 hours.

Order My Audit — $499
Delivered within 48 hours One-time fee, no subscription View sample report
Related
Learn
What Is an AI Visibility Score?
The 7-component framework and what each component measures — from citation coverage to entity recognition.
Data
AI Visibility Rankings: Chicago
See how the methodology scores real businesses across 9 categories in Chicago — HVAC, law, roofing, plumbing, and more.