AI search for legal.
“I got rear-ended and the offer feels low.” “We need a prenup, where do we start?” Clients now tell AI the story and let it name the firm. This guide explains what the engines actually weigh when they answer, and which signals you control.
AI refers legal clients by classifying the story into an area of law, then matching firms whose evidence fits: practice clarity, jurisdiction, results, fee model. It retrieves from bar records, directories, your practice pages, and review language; weighs trust signals with high-stakes caution; then names two or three firms with reasoning. Firms win by publishing story-level practice content and keeping every record verifiable and consistent.
¶AI classifies stories, not categories
Clients don’t know practice areas. They know what happened: “the insurance offer feels low.” “My landlord kept the deposit.” “My co-founder wants out.”
The model does the classification, then matches firms to the story’s specifics: case type, jurisdiction, stakes, urgency. The mental shift for legal marketers: “full-service firm” matches nothing. The firm whose pages speak in story language, low offers, kept deposits, departing partners, gets matched to the stories being told.
1Stage one: what AI reads about your firm
Across presence sweeps in legal moments, the sources cluster into five groups:
- Bar and court records. License status, admissions, disciplinary history. The quiet floor under every referral.
- Legal directories and rankings. Structured, vetted profiles that models lean on for practice and jurisdiction matching.
- Your practice-area pages. Story-level content in crawlable HTML: who you help, with what, where, on what fee model.
- Results and review language. “Settled at 3x the first offer” and “returned every call” are retrievable referral facts.
- Press and case coverage. Notable outcomes and legal-press citations that validate the expertise claimed.
2Stage two: how AI weighs what it reads
| Signal | What the model is checking | Influence |
|---|---|---|
| Practice clarity | Can the model match your pages to the story being told: case type, jurisdiction, client type? | High |
| Record consistency | Do attorneys, practice areas, and locations agree across bar records, directories, and your site? | High |
| Results evidence | Published outcomes and matter-specific reviews that prove the claim, within advertising rules. | High |
| Fee transparency | Contingency, flat, hourly, stated plainly. The most-retrieved legal logistic. | Medium |
| Responsiveness signals | Reviews describing communication speed; urgency stories route to firms that answer. | Medium |
Largely absent: TV budgets, billboard counts, firm size. A three-attorney shop with sharp story pages, clean records, and specific reviews regularly out-refers household names. The courthouse rewards reputation; the referral rewards verification.
3Stage three: the qualified referral
Legal answers arrive pre-qualified: the model assesses the case before naming firms. Three consequences:
- The client arrives convinced. AI validated the case and set expectations before your phone rang. The consult starts at yes.
- Vague firms get filtered, not ranked low. Without story-level evidence, you’re not fourth on the list; you’re absent from it.
- Story wins compound. The firm that owns “low settlement offers” gets every adjacent story: denied claims, delay tactics, bad-faith disputes.
Clients tell stories; models match them to published, verifiable fit. Write your practice pages in story language and keep every record clean, and the referrals find you.
§The trust bar
Legal outcomes are life-changing, so models apply YMYL caution: verifiable records, conservative claims, compliance-safe language. Outcome promises and superlatives get filtered, which keeps bar-rule-compliant presence building as the only kind that works.
The bar protects the diligent: firms with clean, structured, verifiable presence compete inside answers against a fraction of the market.
✓The signals you control
- Publish story-level practice pages in crawlable HTML: the situations clients describe, not the categories you file under.
- Reconcile bar, directory, and site records. Names, areas, and jurisdictions must agree everywhere.
- State your fee model plainly. Fee clarity is the most-asked legal logistic.
- Publish results within advertising rules. Verifiable outcomes are the strongest referral signal you control.
- Cultivate matter-specific reviews that name the case type and the experience.
- Track presence by story, monthly. “Low offer” and “DUI first offense” are separate referrals with separate fixes.
Items one through five are what the Action Engine generates and ships. Item six is the presence map itself.
See your firm’s presence map
30 minutes. Your market live, every moment, every gap.
More for legal teams
Be the firm in the referral.
Find out whether you're in the answer. 30 minutes. your market live across the platforms that matter.