AI citations are how ChatGPT decides whether your business name comes up when a customer asks "who should I call near me" - and if you are a plumber, HVAC tech, garage door installer, locksmith, or chimney sweep, most of your competitors have no idea how they work. That is your opening.
AI citations are mentions of your business inside the answer an AI engine returns. When a user types "best garage door repair in Denver" into ChatGPT, the model produces a short list. The brands named in that list are AI citations. Traditional SEO is a ranking game, competing for position 1 on a Google results page. AI citations are a selection game: the model chooses 3 to 5 names for one short answer. SEMrush and Ahrefs show keyword volume from clickstream data, not what customers ask inside ChatGPT every day.
AI citations matter beyond their direct traffic because of the Trust Stack effect. Customers verify a ChatGPT recommendation on Google, and they verify a Google result on ChatGPT. When the same brand appears in both places, trust multiplies by 3 to 5 times. Showing up in only one place leaves that compounding effect on the table.
AI engines do not invent business names. They pull from a fixed set of sources at answer time. Perplexity lists sources right inside the response; ChatGPT and Gemini are less open, but the pattern is the same. They lean on directories, review aggregators, niche publications, Reddit threads, and YouTube. The selection logic favors recency, mention count across multiple independent sources, structured content that is easy to extract (Q&A blocks, stat blocks, clean tables), and strong EEAT signals with named authors and a visible updated date. If your business appears in 8 directories that ChatGPT trusts, you show up. In 1 or 2, you do not.
This is the technique I run for every client. No competitor I have audited does it. It costs zero dollars and takes about 90 minutes per client to set up. Understanding what citation building actually is makes the logic obvious: the more sources confirm your business exists, the more weight your name gets in the AI's answer.
Most "GEO experts" charge $5,000/mo and skip this step. They write blog posts and pray. The reverse-engineering process is the actual mechanism.
Or runs denvergaragedoor.com. He had 13 Google reviews when I started, against a competitor with 253. I asked ChatGPT 12 customer-style questions about garage doors in Denver and logged every cited source. ChatGPT was pulling from 9 directories plus 3 niche publications. Or was registered on 2 of them. Within 14 days I registered him on all 9. Within 4 weeks I submitted him to the 3 niche publications; two accepted. At week 6, Or appeared in 4 of 12 answers. By week 10, it was 7 of 12.
The Google side moved in parallel. The full case study covers the 7 unindexed pages I fixed and the keyword work. The core point: AI citations and Google rankings feed the same stack: clean schema, NAP consistency, and indexed pages move both.
Schema markup labels content for machines and removes guesswork for AI engines. None of the top 5 ranking pages for "ai citations" use any schema; that is a free win. The full schema-for-local-SEO playbook covers the syntax. The short version: Organization schema confirms the brand is a real entity; LocalBusiness schema is the spine of any citation play (NAP, hours, service area); Service schema matches user intent to your offering; FAQPage schema makes every Q&A machine-readable; HowTo schema gives AI engines step-by-step lists to extract into answers.
Across about 80 AI-cited home-service pages I have audited, a few patterns repeat: Q&A blocks with a single focused 60 to 100-word answer; stat blocks with specific numbers ("13 Google reviews beat 253 reviews in Denver" is more citable than any generic claim); tables with comparable rows for pricing or before-and-after metrics; named sources with real client URLs like americaschimneysweep.com; and a visible "Updated [month/year]" recency marker. AI engines downrank pages that haven't been updated; a visible recent date is one of the easiest signals to keep current.
Tracking is where most vendors fall short. Offshore teams often run a smaller prompt list and skip the hallucination check. I built an internal tool that automates the work, and every paying client gets a monthly AI citation report. The workflow: define 20 to 50 priority queries from sales calls (not SEMrush); run them through ChatGPT, Perplexity, Gemini, and Claude monthly in fresh sessions; score each answer for brand presence, position, and which cited source we built; compare month over month as citation rate climbs from 8% to 22% to 41%; feed gaps back by finding which sources ChatGPT used for a missed query and adding the client there. This is the only honest way to track your rankings in AI search. AI citation tracking shows whether the work is actually moving your rankings, or whether you are just paying for promises.
Google Business Profile feeds into every AI engine. A broken GBP cuts your citation potential at the source. Four mistakes I see weekly: hidden address with a bad SAB setup that loses local pack visibility; spammy service lists stuffed with keyword variations (good GBP optimization is selective, not exhaustive); wrong primary category that causes AI engines to misclassify the business; and a spam-sounding business name Google demotes. A clean, keyword-free business name that reads like a real company is one of the easiest trust signals to get right from the start.
Four phases built for home-service businesses that need AI citations and Google rankings to fire together.
I run the citation reverse-engineering process on the client's market, map every AI-cited directory and publication for their niche and city, pull the current GBP, and run a full technical audit. The audit checklist covers every element.
Fix GBP errors. Submit to every tier-1 directory from the audit. Deploy the schema stack. Fix unindexed pages (Or had 7). Rewrite weak service pages with Q&A blocks, stat blocks, and tables.
Niche-specific blog posts, local case studies, EEAT signals, and outreach to industry publications AI engines pull from. Our local SEO for home services guide goes deeper on what ships in this phase.
Monthly AI citation report. Monthly GBP report. Quarterly content refresh based on citation gaps.
The tool I trust most is one I built: an internal AI citation tracker that runs four engines on a fixed schedule and emails monthly reports to clients. For supporting work: Perplexity Pro for reverse-engineering ($20/mo); Google Search Console for what AI Overview pulls from (free); BrightLocal for directory submissions at scale (around $39/mo per client); Screaming Frog for site crawl ($200/year). I do not use SEMrush or Ahrefs for AI citation work; their AI search coverage is thin. Surfer and Clearscope optimize for keyword density, not for the specific numbers and named examples that AI engines quote. Pages built to their briefs tend to look identical across agencies. AI engines quote pages with specific numbers and named examples; generic overview content rarely makes it into the cited set.
Pricing is public. Three tiers:
| Tier | Price | Best for | What you get |
|---|---|---|---|
| Starter | $750/mo | Solo operator, 1 service area | Reverse-engineering audit, 15 tier-1 directory submissions, schema stack deploy, 2 service pages rewritten, monthly report on 20 queries |
| Pro | $1,500/mo | Established business, 2-4 cities | Everything in Starter plus 30 directories, neighborhood pages, 2 blog posts/mo, report on 35 queries, GBP optimization |
| Custom | $3,000+/mo | Multi-location, multi-state | Everything in Pro plus 50 directories, niche publication outreach, video content, weekly reporting |
See full package details for what is included at each tier. Most clients land on Pro. Month-to-month; no lock-in.
If you used ChatGPT or another LLM as a source in a paper, APA 7 recommends citing it as software output with the model name, version, prompt date, and developer. MLA suggests citing the prompt and the AI tool. Chicago treats it as personal communication. The format is evolving fast; check your style guide's most recent update before submitting.
Customers research purchases inside ChatGPT, Perplexity, and Gemini before they open Google. If your business is invisible in those answers, you are invisible to a growing share of buyers. Tracking shows which queries cite you, which competitors are taking your share, and where to fix the gaps.
It starts with reverse-engineering the sources AI engines actually use, not generic "GEO best practices." It includes directory submissions to the specific sites cited in your niche, schema markup tuned for AI extraction, content patterns AI engines prefer, and monthly query-level tracking across 4 engines. The typical agency skips both the reverse-engineering and the tracking; those two steps are the difference between guessing and proving.
Generative AI is a class of models that produce new content based on training data. ChatGPT, Claude, Gemini, and Perplexity are all generative AI engines. For local businesses, the practical consequence is one synthesized recommendation instead of 10 ranked links. Earning a slot in that one answer is the whole AI-citation game.
For most home-service clients, first AI citations appear within 30 to 45 days of registering on the right directories. By month 3, citation rate typically moves from under 10% to 30-50%. Or's Denver account was cited by ChatGPT at week 6. Tomer at Pine Garage Doors took about 8 weeks because his niche was more competitive. Speed depends on niche, city density, and how many tier-1 directories the client was already on.
Yes and no. The infrastructure overlaps: clean site, fast load, schema, EEAT, NAP consistency, GBP optimization. The AI layer adds reverse-engineering of cited sources, content patterns AI engines prefer, and query-level tracking across chat engines. The full GEO vs traditional SEO comparison breaks it down line by line.
I run the reverse-engineering process during the free audit. You see the exact directories ChatGPT cites in your niche, which ones you are on, and which ones your competitors are using. You get the source list and the gap, not a sales deck. Book your free AI visibility audit here and I will run the full reverse-engineering on your top 10 queries before our call.
Not every business wants to show up in AI Overviews. If that is your call, here is how to turn off Google's AI Overview for your pages.
I specialize in home services SEO – taking websites that sit invisible on page three and turning them into the business Google and ChatGPT recommend first. I started on the developer side, writing software and doing SEO on the side, until I saw how much home-service owners were overpaying for work that quietly hurt them. So I built a method that fixes the broken technical work and the outdated thinking behind it.
From garage door companies to plumbers, roofers, locksmiths and cleaning services, the playbook is the same: rank where your customers actually search, earn real reviews, and back it with a fast site that books the job. No PBNs, no bought reviews, no directory spam – only work that survives Google’s next five updates. See exactly how it’s priced on the pricing page.
Send me your site and I’ll send back a free audit: what’s broken, what it’s costing you in calls, and the first three fixes.
No spam and no sales pitch. Just a clear look at what’s leaking leads.
