Strong firms with twenty-year track records and four-figure review counts are routinely invisible in ChatGPT, Perplexity, Gemini, and Google AI Overviews. The reasons are specific. The fixes are concrete. Seven of them, ranked by frequency.
The model cannot tell which firm you are. Your firm name appears in multiple variations across the web — "Smith & Smith P.A.," "Smith Smith Law," "The Smith Firm," "Smith Brothers Personal Injury" — and the model treats them as different (or worse, low-confidence) entities. Entity ambiguity is the single most common cause of invisibility and the most underdiagnosed.
Fix: Pick one canonical firm name. Audit every directory, profile, citation, and review platform. Reconcile to the canonical name with consistent NAP data (name, address, phone). This is the foundation that everything else builds on.
The firm's website emits no schema markup, or emits broken schema. AI engines parse schema directly when retrieving content. Without it, the firm reads as flat undifferentiated text — just words, no metadata, no entity hooks. The competitor with clean Attorney + LegalService + Review schema gets surfaced because the model can read them.
Fix: Implement Attorney, LegalService, Organization, FAQPage, and Review schema across the relevant pages. Validate with Google's Rich Results Tester. This is among the highest-leverage AEO moves available — and one of the cheapest.
Every page on the firm's site is a thousand-word "we handle personal injury cases including car accidents, truck accidents, motorcycle accidents..." block. None of them answer a specific consumer question in depth. AI engines have no narrow page to retrieve when the user asks a narrow question, so the firm doesn't surface.
Fix: Build narrow pages that answer narrow questions. "How to file a UM claim in Florida" is a page. "What is the average settlement for a rear-end accident with delayed-onset symptoms" is a page. Generic "personal injury" pages can stay, but they are not what wins AI answers.
The firm is barely mentioned anywhere outside its own website. Two listings in legal directories. No news coverage. No association presence. No expert quotes in third-party content. AI engines reward firms that show up across the legal web; firms that exist only on their own domain look untrusted by comparison.
Fix: Citation engineering. Targeted directory submissions on authority sources (Justia, Avvo, Super Lawyers, state bar profiles). Expert source pitches to legal publications. Association memberships with linked profiles. The goal is not volume — it's authority and consistency.
The firm has plenty of five-star reviews. Almost none of them say anything specific. "Great firm, would recommend" appears two hundred times. AI engines cannot quote that. They can quote "she walked me through the UM coverage limits and got my claim approved in three weeks" — and that is what gets pulled into answers.
Fix: Adjust review acquisition to encourage specificity. Ask satisfied clients about practice area, outcome, and attorney name. The substance becomes the substrate AI engines extract from.
The firm exists in isolation. The attorneys do not appear as expert sources on other sites. The firm name does not appear in case-result articles, association rosters, awards mentions, or news coverage. Without an entity surround, the model has no contextual signal that this firm is part of the conversation in its practice area.
Fix: Build topical authority outside the firm's own website. PR placements, expert-source outreach, contributed articles in legal publications, podcast appearances, conference inclusion. The web around the firm matters as much as the firm's own site.
The agency is doing more SEO work — link building, keyword optimization, on-page tuning — and assuming it will eventually translate to AI visibility. It won't. The correlation between Google SEO rank and AI visibility is 0.076%. Doubling SEO investment does not double AEO visibility. It usually doesn't change AEO visibility at all.
Fix: Treat AEO as a separate discipline. Fund it separately. Measure it separately. Hold it to a different metric (engine appearance rate, not search rank). The work has structural overlap with SEO but distinct deliverables.
Across roughly 200 PI firm diagnostics LawShift has run, the average firm has three or more of these seven causes active simultaneously. The single most common combination: entity ambiguity (01) plus no structured data (02) plus generic practice pages (03). Fixing those three covers 70% of invisible firms.
The firms that are visible now are accumulating citation density. Their entity surround compounds. Their schema gets richer. Each month of inaction widens the gap. In a year, the dominant AI-visible firm in a metro will have three to five times the citation surface of an invisible competitor. Closing that gap retroactively takes longer and costs more than starting now.
The asymmetry is real. First movers in AEO are locking in market position in a way that traditional SEO never allowed. The leader board is being written in 2026. It will be far harder to rewrite in 2028.
60 seconds. Four engines tested. Per-engine breakdown of what's working and what isn't.
Run the Diagnostic →