When a prospect asks ChatGPT to recommend a business like yours, one of two things happens: your name comes up, or it doesn't. The citation score exists to tell you which one — and more importantly, exactly why.
It's a single number from 0 to 100 that measures your AI visibility across the major answer engines. But unlike a vanity metric, every point on the scale is traceable to real behavior — actual AI queries, actual responses, actual citations. Here's exactly how it's calculated and what your number means in practice.
How the citation score is calculated
SwiftGeo runs your business against a battery of queries across ChatGPT, Perplexity, Gemini, and Copilot. These aren't generic queries — they're generated based on your specific business type, location, and industry, matching the actual language your prospects use when asking AI engines for recommendations.
The final score is a weighted composite of four dimensions:
(Mention Rate × 0.35)
+ (Prominence × 0.30)
+ (Sentiment × 0.20)
+ (Cross-Engine Consistency × 0.15)
// All components normalized to 0–100 before weighting
Mention Rate (35%)
The percentage of relevant queries where your business is named at all. If you run 20 queries and appear in 12 responses, your mention rate is 60%. This is the foundation — you can't score well on the other dimensions if you're not being mentioned in the first place.
Prominence (30%)
Not all mentions are equal. Being the first name in a response — introduced as the primary recommendation — carries far more weight than appearing fifth in a list. Prominence measures where in the response you appear and how confidently the AI presents you. A score of 7+ out of 10 on prominence typically correlates with conversion-quality traffic from AI search.
Sentiment (20%)
When AI engines mention your business, what language do they use? Neutral mentions ("Smith Law is one option in Manhattan") score lower than confident endorsements ("Smith Law is widely regarded as one of the leading personal injury firms in New York, with over $150M recovered for clients"). Sentiment captures the quality of the citation, not just its existence.
Cross-Engine Consistency (15%)
ChatGPT, Perplexity, Gemini, and Copilot don't share information — they draw from different training data and retrieval systems. A business that appears consistently across all four engines has a fundamentally stronger AI presence than one that appears frequently on one engine and is absent from the others. Consistency also signals authority: it means multiple independent systems have high-confidence information about your business.
Mention rate is weighted highest because it's the gateway metric. No amount of prominence or sentiment quality matters if you're not being cited. The weights reflect the sequential logic of AI citation: first you get mentioned, then you get mentioned well, then you get mentioned consistently everywhere.
What a good score looks like by industry
Context matters. A score of 45 means something different for a local med spa in a mid-size market than for a national SaaS company competing against category leaders. Here are the benchmarks we see across our core verticals:
| Industry | Avg. First Scan | Competitive | Strong |
|---|---|---|---|
| Personal Injury Law | 18–28 | 45+ | 65+ |
| Med Spa / Aesthetics | 12–22 | 40+ | 60+ |
| E-Commerce (niche) | 22–35 | 50+ | 68+ |
| Financial Services (SMB) | 15–25 | 42+ | 62+ |
| SaaS / Software (SMB) | 28–42 | 55+ | 72+ |
| Marketing Agency | 20–32 | 48+ | 65+ |
Two observations from this data: first, most businesses score significantly lower than they expect on their first scan. The average visibility gap on a first scan across our user base is approximately 72% — meaning the typical business is missing from roughly 7 out of 10 relevant AI queries. Second, the gap between a first scan score and a "strong" score is almost always closeable within 90 days with systematic GEO work.
What each score range means in practice
0–25 (Invisible): You're absent from the overwhelming majority of relevant AI queries. AI engines either don't have reliable information about your business, or what they have isn't structured in a way they can cite with confidence. This is the starting point for most businesses that haven't done any GEO work.
26–45 (Emerging): You're appearing in some queries — probably ones where your business name is directly searched, or in low-competition categories. You're not yet showing up for the high-intent recommendation queries where your best prospects are. This is where most businesses land after basic optimization work.
46–65 (Competitive): You're regularly cited across at least two major AI engines for relevant queries in your category. A prospect asking for a recommendation in your space has a reasonable chance of hearing your name. This is the first tier where AI search starts generating meaningful referral activity.
66–80 (Strong): You're appearing consistently across most major engines, often as a primary recommendation. At this level, AI search is a measurable traffic and lead source for your business. You have a clear advantage over the majority of competitors in your space.
81–100 (Dominant): Reserved for businesses with extensive authoritative citation networks, near-perfect structured data, and strong cross-platform consistency. Category leaders in well-established industries typically land here.
What the gap means in revenue terms
The question we get most often: what is a 10-point improvement in citation score actually worth?
The honest answer is that it depends on query volume in your category and your average deal size — but the directional math is consistent. AI search now influences a meaningful percentage of high-intent recommendation queries in most service categories. A business moving from a score of 20 to 50 typically sees a 3–5x increase in AI-referral leads, because it's moving from appearing in roughly 1 in 5 queries to appearing in roughly 3 in 5.
For a personal injury law firm handling cases worth $50,000+ on average, showing up in 2 additional AI queries per week that convert to consultations at a 20% rate represents significant incremental revenue — from a one-time optimization investment.
Unlike paid ads, AI citation presence compounds. A well-structured LLMs.txt file and Schema markup don't expire. Every AI training update that incorporates your authoritative data makes you progressively harder to displace. The businesses building strong citation scores now are creating a durable advantage that will be increasingly expensive for competitors to replicate.
Your citation score is a real measurement of a real gap. Most businesses are missing from 70%+ of the AI queries where they should appear. The good news: it's entirely fixable, and it's fixable systematically. The scan shows you where you are. The platform shows you exactly what to do next.
Find out your citation score in minutes.
Free scan across ChatGPT, Perplexity, Gemini, and Copilot — see exactly where you stand and what's holding you back.
Get Your Free Audit →