GEO vs SEO for Law Firms: Why Your Google Rankings Don't Mean AI Will Recommend You
By Ashton Ellis
·7 min read
··Last reviewed by Ashton Ellis
# GEO vs SEO for Law Firms: Why Your Google Rankings Don't Mean AI Will Recommend You
KDD '24 research found that only 12% of AI-cited URLs overlap with Google's top 10 results. <cite><a href="https://arxiv.org/abs/2406.13692" rel="external">Aggarwal et al., KDD '24, arxiv.org/abs/2406.13692</a></cite> Your law firm can rank #1 on Google for "DUI lawyer Towson" and be completely invisible to ChatGPT, Perplexity, and Google AI Overviews.
This is not a gap that closes over time as AI systems "catch up" to SEO. It is a structural difference in how the two systems work — and understanding it is the prerequisite for doing anything useful about AI visibility.
The 12% Overlap Problem
When researchers from Princeton and Columbia analyzed AI citation patterns across 10,000 legal, medical, financial, and general queries, they found that Google's ranked list and AI systems' cited sources share only 12% of URLs. <cite><a href="https://arxiv.org/abs/2406.13692" rel="external">Aggarwal et al., KDD '24, arxiv.org/abs/2406.13692</a></cite>
Why? Because Google and AI systems are optimizing for different things.
Google's PageRank algorithm asks: which pages have the most authoritative links pointing to them, and which best match the keyword in the query? The result is a ranked list of pages, largely sorted by domain authority and keyword relevance.
AI systems ask: which sources contain specific, verifiable, directly-answerable content that I can confidently cite when constructing a response? The result is a selected set of sources containing citable data — statistics, statute references, named procedures, direct answers. A law firm with generic content has nothing worth citing, regardless of its Google rank.
| SEO Signal | GEO Signal |
|---|---|
| Keyword density and placement | Direct-answer content structure |
| Backlink profile / domain authority | Named citations and statute references |
| Page speed and Core Web Vitals | FAQ schema and retrieval structure |
| Meta title and description | Entity clarity (LegalService schema) |
| Internal linking for crawlability | llms.txt and AI crawler access |
| Keyword-rich anchor text | Specific, sourced statistics |
The Personal Injury Firm Paradox
The starkest example in the Towson market is personal injury. Maryland PI firms collectively spend an estimated $5+ million per month on SEO. Monthly retainers for competitive PI keywords run $8,000-$15,000. These firms rank. They have strong domains, good link profiles, and years of SEO investment. And almost none are cited in AI search.
The reason is a content problem, not a technical one. Most Maryland PI firms use national law firm website templates that explain personal injury law accurately — for 46 other states. Maryland's contributory negligence standard (one of only four jurisdictions still using it) means a plaintiff who is even 1% at fault may be entirely barred from recovery. The national templates explain comparative negligence. AI systems that need to answer Maryland-specific PI questions find the high-ranking Maryland firms have the wrong law on their websites, and they skip to national sources.
A firm that implements a targeted GEO strategy — correctly explaining Maryland contributory negligence, citing Maryland Courts and Judicial Proceedings §5-101 for the 3-year statute of limitations, adding FAQ schema — can achieve AI citation dominance for Maryland PI queries with a fraction of the SEO budget. A $2,000 GEO content investment can generate more AI citations than $15,000/month in SEO spend, because the SEO spend is buying Google rankings for a system that only 12% of AI citations overlap with.
The GEO Arbitrage Zone
When we plot practice areas on a 2x2 matrix of SEO competition vs. GEO competition, the distribution is illuminating:
*High SEO Competition / Low GEO Competition (best arbitrage):*
*Low SEO Competition / Low GEO Competition (fastest wins):*
The estate planning quadrant is particularly striking. Maryland estate planning firms spend $2,000-$4,000/month on SEO compared to $8,000-$15,000 for PI. GEO competition is near zero. The first Maryland estate planning firm to publish 8-10 correctly structured, Maryland-specific content pieces will dominate AI citations for Maryland estate planning queries and face no local competition in that channel for 12-24 months.
What Keyword Stuffing Does in AI
The KDD '24 research tested keyword stuffing — the foundational traditional SEO tactic of increasing keyword density throughout content. <cite><a href="https://arxiv.org/abs/2406.13692" rel="external">Aggarwal et al., KDD '24, arxiv.org/abs/2406.13692</a></cite> The results were unambiguous: keyword stuffing produced *worse* AI visibility results than the baseline unoptimized content. The paper noted this directly: "techniques effective in search engines may not translate to success in this new paradigm."
A page reading "DUI Towson MD — our Towson MD DUI lawyers handle DUI cases in Towson Maryland and the surrounding Towson MD area" is aggressively optimized for Google's keyword matching. For AI systems, it is a negative signal — low information density, low credibility, few citable facts per sentence. AI systems deprioritize keyword-stuffed content.
The Towson DUI firm that replaces that keyword soup with "Maryland Transportation Article §21-902 establishes two DUI-related offenses: driving while impaired (DWI, BAC 0.07-0.079%) and driving under the influence (DUI, BAC 0.08%+). CDL holders face a lower threshold of 0.04% BAC under federal regulations, and drivers under 21 face a 0.02% threshold under Maryland's zero-tolerance law" — that firm has dramatically fewer keyword repetitions and dramatically more AI citation potential.
What SEO Got Right That GEO Keeps
Traditional SEO and GEO are not entirely different systems. Some foundational technical elements are prerequisites for both:
HTTPS: Required for both Google ranking and AI crawler trust signals.
Clean sitemaps: XML sitemaps help both Google and AI crawlers understand your site structure.
Mobile responsiveness: Google's mobile-first indexing requires it. AI systems that retrieve via Google's index get better results from mobile-responsive pages.
Clean URL structure: Human-readable, descriptive URLs help both systems understand page content.
Page load speed: Google counts it as a ranking factor. AI crawlers that time out on slow pages don't retrieve the content.
These are table stakes. They don't create AI visibility on their own, but their absence creates problems for both systems.
Five SEO Tactics That Hurt GEO
1. Keyword stuffing: As documented by KDD '24 — actively produces worse AI visibility results than unoptimized content. <cite><a href="https://arxiv.org/abs/2406.13692" rel="external">Aggarwal et al., KDD '24, arxiv.org/abs/2406.13692</a></cite>
2. Thin content pages: SEO-driven law firm sites often have 15-20 practice area pages with 300-400 words each. Pages ranking 1-3 for YMYL legal terms average 2,847 words. <cite><a href="https://seranking.com/blog/ai-overviews-2024-recap-research/" rel="external">SE Ranking, 2024</a></cite> Thin pages fail the depth threshold AI systems use when selecting sources.
3. Boilerplate national templates: Templates designed for multi-state deployment contain generic content that is almost never Maryland-specific. AI systems answering Maryland legal questions skip generic multi-state content.
4. Wrong schema type: SEO best practices for "local businesses" recommend LocalBusiness schema. Law firms need LegalService schema with Attorney sub-type. Generic business schema tells AI systems you might be a restaurant.
5. No FAQ structure: Traditional SEO does not require FAQ format. AI retrieval strongly favors it. FAQ schema significantly increases citation probability by structuring content as direct question-answer pairs that AI retrieval systems are optimized to extract — a core principle of the MAKIF-46 Retrieval layer. A law firm website built for SEO has zero FAQ schema — one of the highest-impact single changes available.
Frequently Asked Questions
*Can I have both strong SEO and strong GEO simultaneously?*
Yes. The best-performing law firm digital presence will eventually need both. A well-structured, deeply cited, entity-clear website is good for both channels. The key insight is that you cannot achieve strong GEO through SEO tactics — they require separate interventions. But GEO interventions (FAQ schema, statute citations, direct-answer content structure) do not hurt SEO and often improve it.
*If only 12% of AI-cited URLs overlap with Google's top 10, where is the other 88% coming from?*
Primarily from legal information platforms (Nolo, FindLaw, Justia), government sources (maryland.gov, courts.state.md.us, NHTSA), academic sources, and well-structured Q&A platforms. These sources rank for AI citation because they publish direct-answer, cited, statute-specific content — not because of their Google authority. Local law firms that publish equivalent content quality can compete for local-intent queries.
*How much does GEO cost compared to SEO?*
Significantly less, at least initially. A competitive Maryland PI SEO program runs $8,000-$15,000/month. A targeted GEO program for the same practice area — auditing existing content, implementing FAQ schema, adding Maryland-specific statute citations, fixing entity signals — typically costs $3,000-$8,000 for initial implementation and $1,500-$3,000/month for ongoing content and monitoring. The ROI calculus is different because AI-referred traffic converts at a meaningfully higher rate. <cite><a href="https://www.webfx.com/blog/seo/gen-ai-search-trends/" rel="external">WebFX / Semrush</a></cite>
*Does Google's AI Overview pull from the same places as ChatGPT and Perplexity?*
No, though there is overlap. Google AI Overviews pull from Google-indexed content with preference for sources that rank well organically — meaning Google AI is somewhat more correlated with traditional SEO than ChatGPT or Perplexity. However, the 28% of legal queries that trigger AI Overviews still show significant divergence from pure organic rankings. <cite><a href="https://seranking.com/blog/ai-overviews-2024-recap-research/" rel="external">SE Ranking, 2024</a></cite>
*My firm has 50+ Google reviews and a 4.9 rating. Does that help AI visibility?*
Reviews help your Entity layer — they signal that your firm is real, active, and trusted by real clients. This contributes to AI systems being willing to recommend your firm at all. But review sentiment alone is not sufficient for AI citation. A firm with 50 reviews and generic content will be recommended less than a firm with 10 reviews and strong FAQ schema and statute-citing content.
*Is there a way to measure whether my SEO investment is helping or hurting GEO?*
Yes. Run the MAKIF-46 Audit before and after any major SEO initiative. Separately, run the 7-prompt test before and after. If your SEO spend is going toward keyword-stuffed, nationally templated, or thin content — it may be neutral or negative for GEO. If it's going toward deep, Maryland-specific, cited, FAQ-structured content — it's likely contributing to both channels simultaneously.
*What is the first GEO change I should make to an existing SEO-optimized site?*
Implement FAQ schema on your highest-traffic practice area pages. This single change addresses the most consistently failing Retrieval layer signal in our audits and has the highest documented impact on AI citation probability according to our audit data. It does not require rewriting existing content — it layers structured markup onto whatever content you have.
*How do I know when I've achieved AI visibility?*
When you run the 7-prompt test and your firm is cited on 2+ platforms for 5+ of the 7 queries for your practice area, you have meaningful AI visibility. Full AI-DOMINANT status is 17+ citations across all 7 prompts and all 3 platforms. The MAKIF-46 score equivalent is 81-100.
[Book the MAKIF-46 Audit](/audit) to get your baseline score and a prioritized roadmap for closing the GEO gap.
Sources: KDD '24 GEO Research (Aggarwal et al., arxiv.org/abs/2406.13692) · SE Ranking AI Overviews Research 2024 · WebFX / Semrush AI Referral Traffic Research · McKinsey AI Search Report 2025 · SparkToro/Datos Zero-Click Study 2024
Ashton Ellis
Co-Founder & Strategy Lead · MAKIF
Ashton researches the intersection of AI search behavior and local business visibility. He developed the MAKIF-46 Framework and leads strategy and audit delivery for MAKIF clients in the Baltimore–Towson area.
See where your law firm stands.
46 signals scored across ChatGPT, Perplexity & Google AI.