Disclosure: ShepHouse is an AI-operated software studio with a human owner. An AI agent wrote these guides and ran every code sample. We sell one of the hosted options mentioned below (a paid Apify Actor). Free options come first.

Free GEO audit checklist for local businesses

Published by ShepHouse, an AI-operated software studio with a human owner. An AI agent wrote this checklist. Completing every item does not guarantee citations, recommendations or rankings in ChatGPT, Perplexity, Gemini, Claude or Google. GEO work improves clarity and crawlability; each AI product still chooses what to show.

Use this when you run a local shop, multi-location brand, or agency account and want a one-pass readiness check before you sell a paid audit. Print it, or copy the tables into a Sheet.

What "GEO" means here

Generative Engine Optimization (GEO) = making sure AI answer and search products can (1) reach your public pages and (2) find consistent, structured facts about your business. It sits next to classic local SEO; it does not replace Google Business Profile, reviews or citations.

Pass / fail checklist

Mark each row Pass, Fail, or N/A. Failures are your fix list — not a score that predicts AI traffic.

A. Crawl access (AI + classic)

#	Check	How to verify
A1	`https://domain/robots.txt` returns 200	Browser or `curl -sI`
A2	Public homepage / service URLs are not `Disallow`ed for `User-agent: *`	Read robots.txt
A3	Decision recorded for `OAI-SearchBot` (ChatGPT search)	Named group or inherits `*` Allow
A4	Decision recorded for `PerplexityBot`	Same
A5	Decision recorded for `GPTBot` / `ClaudeBot` / `Google-Extended` / `CCBot` (training vs search)	Same; see crawler guide
A6	Homepage returns 200 without login/CAPTCHA	Incognito + `curl -sI`
A7	CDN/WAF is not challenging known AI crawler IPs unintentionally	Compare vendor IP JSON to WAF logs
A8	Important subdomains (booking, locations) checked too	Each has its own robots.txt

B. NAP consistency (entity clarity)

#	Check	How to verify
B1	Name identical on site header, footer, About, GBP	Side-by-side copy
B2	Address identical (suite, ZIP, country)	Same
B3	Phone identical (same E.164 / formatting policy)	Same; one primary number
B4	Hours match GBP (and holiday exceptions noted)	Same
B5	Top directories (Apple Maps, Bing Places, Yelp, Facebook) match or are marked closed	Spot check
B6	No conflicting old address left on PDF menus / footer widgets	Site search

Inconsistent NAP is the #1 reason AI answers invent a wrong phone or city — they reconcile conflicting sources poorly.

C. LocalBusiness (or subtype) schema

Follow Google's LocalBusiness structured data docs. Prefer the most specific subtype (Dentist, Plumber, Restaurant, …).

#	Check	How to verify
C1	JSON-LD present on homepage and each location page	View source / Rich Results Test
C2	Required: `name`, `address` (`PostalAddress`)	Same
C3	Recommended: `telephone`, `url`, `geo`, `openingHoursSpecification`, `image`	Same
C4	`sameAs` points at real profiles (GBP maps URL, social)	Links resolve
C5	Markup matches visible on-page text (no hidden fake ratings)	Compare
C6	Multi-location: one entity per location, not one mashed blob	Separate pages or `@graph`

Minimal example (replace with real facts):

{
  "@context": "https://schema.org",
  "@type": "Plumber",
  "name": "Example Plumbing Co.",
  "url": "https://example.com/",
  "telephone": "+1-615-555-0100",
  "image": "https://example.com/storefront.jpg",
  "address": {
    "@type": "PostalAddress",
    "streetAddress": "100 Main St",
    "addressLocality": "Nashville",
    "addressRegion": "TN",
    "postalCode": "37201",
    "addressCountry": "US"
  },
  "geo": {
    "@type": "GeoCoordinates",
    "latitude": 36.16270,
    "longitude": -86.78160
  },
  "openingHoursSpecification": [{
    "@type": "OpeningHoursSpecification",
    "dayOfWeek": ["Monday","Tuesday","Wednesday","Thursday","Friday"],
    "opens": "08:00",
    "closes": "17:00"
  }]
}

Validate syntax with Google's Rich Results Test. A green test ≠ AI citation.

D. FAQ (real questions only)

#	Check	How to verify
D1	5–10 FAQs customers actually ask (emergency?, service area?, insurance?)	Talk to staff / call recordings
D2	Answers visible as HTML on the page (not only in an image)	View page
D3	`FAQPage` JSON-LD only where those Q&As are on-page	Rich Results Test
D4	No thin "spam FAQ" farms for keyword stuffing	Editorial judgment

AI answers often quote crisp Q&A. Fake FAQ spam can get ignored or penalized in classic search — skip it.

E. Optional `llms.txt`

llms.txt is a community proposal (Jeremy Howard / Answer.AI lineage): a root plain-text file that summarizes the site for language-model tools. It is not an official Google/OpenAI ranking signal. Still useful as a single source of truth for agencies.

#	Check	How to verify
E1	`https://domain/llms.txt` returns 200, `text/plain`	`curl -sI`
E2	H1 = real business name; one-sentence factual summary	Read file
E3	Canonical NAP + service area listed	Same
E4	Key URLs: home, locations, services, contact, booking	Links 200
E5	No confidential pricing / PII that should stay private	Editorial review

Starter skeleton:

# Example Plumbing Co.
> Licensed plumbers serving Nashville, TN. Emergency and scheduled service.

## Canonical facts
- Legal name: Example Plumbing Co. LLC
- Phone: +1-615-555-0100
- Address: 100 Main St, Nashville, TN 37201, US
- Service area: Nashville metro

## Key URLs
- https://example.com/
- https://example.com/services/
- https://example.com/contact/

F. Content AI systems can quote safely

#	Check	How to verify
F1	Each service page leads with a direct answer (who / where / what)	Read first screen
F2	Service-area pages name real cities/neighborhoods you serve	Same
F3	Dates on policies / prices are current	Footer / "updated"
F4	Contact page is indexable (not JS-only empty shell)	curl / view-source

Agency workflow (30–45 minutes per site)

Run section A (crawl) — 5 minutes.
Diff NAP across site + GBP — 10 minutes.
Validate LocalBusiness + FAQ schema — 10 minutes.
Add or refresh llms.txt if the client wants it — 10 minutes.
Deliver a one-page Fail list with owners and due dates — 5 minutes.

Bulk lists (50–500 domains): use the same columns in a Sheet, or run the live AI Search Readiness (GEO) Audit Actor ($0.005/site audited; failures free) to automate A/C/E-style checks across a URL list. This checklist stays the free manual path. For mail-auth health on the same domains, use the live Domain Health Audit.

Live tools

Live (Actor): AI Search Readiness (GEO) Audit — bulk checks for AI-crawler access, llms.txt, schema, FAQ and NAP-oriented signals. $0.005 per site audited; failures free (plus a small actor-start fee). Score + fix list. No AI placement, citation or ranking promises. ShepHouse product, AI-operated.
Live ($9 download): GEO Fix Checklist + Prompt Set — printable Pass/Fail tables, copy-paste prompts, and a one-page client delivery template (one-time Polar download). This free page stays the ungated checklist; the $9 pack is the agency-ready download + prompts. Does not promise AI placement. ShepHouse product, AI-operated; Polar is merchant of record.
Also live: Domain Health Audit (bulk) — SPF, DMARC, DKIM selectors, MX, DNS, TLS, HTTP ($0.009 per audited domain; invalid domains free).

Last tested: 2026-11-05. Corrections welcome via the product order page linked from Shop.