Every Legal Directory, Ranked by AI Citation Weight.

Every legal directory sells the same promise: visibility. But when an AI engine recommends a lawyer, it cites the sources it actually trusts, and that list turns out to be short, lopsided, and very different from where firms spend their directory budgets. We measured which directories ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews cite when they name law firms, and ranked every major directory by that number.

What AI citation weight is and how we measured it

AI citation weight is the share of AI answers that cite a given source when recommending a lawyer, and it can be measured rather than guessed. Every ranking of legal directories you have read until now was built on domain authority scores, traffic estimates, or vendor claims. None of those measure the thing that now matters: whether the engines actually read the directory when they decide which firm to name.

The data comes from our AI Citation Index, a five-engine study of personal injury lawyer recommendations. Over a two-week window in June 2026 we logged 359 answers from ChatGPT, Perplexity, Claude, Gemini, and Google AI Overviews across five US metros, recording every firm named and every source cited.

For this ranking we counted, for each directory, the share of answers that cited it at least once. One caveat carried over from the study: Gemini returns opaque redirect links rather than real source URLs, so its sources could not be attributed and are excluded from the tally. The percentages below describe the four engines whose citations can be verified.

Google properties, Reddit, and YouTube also earned citations in the study, at 12, 8, and 5 percent respectively. They are not directories, so they sit outside this ranking; how they fit into the bigger sourcing picture is covered in where AI finds lawyers.

The ranking: every major legal directory

Two directories carry most of the weight, and several household names barely register. Here is every major legal directory we tracked, ranked by the share of answers that cited it.

Percentages are the share of the study's attributable answers that cited each domain at least once, personal injury queries, five metros, June 2026. A zero means the directory did not appear in a single logged answer in this sample, not that it can never be cited.

Read the top two lines again. Super Lawyers and Justia, at 53 percent combined, were cited more than every other legal directory in the table put together. The distribution is not a gentle slope; it is a cliff.

The three tiers of directory value

The measured weights sort every legal directory into one of three tiers, and each tier calls for a different decision. The tier matters more than the exact percentage, because the percentages will move between study runs while the structure has stayed stable across engines and metros.

Tier one: Super Lawyers and Justia

The two most-cited directories have almost nothing in common as businesses, which is exactly why their dominance is instructive. One is a selective peer-review list you cannot pay your way onto. The other gives every licensed attorney a free structured profile. What they share is what the engines reward: scale, structure, and a signal the engine can treat as independent.

Super Lawyers: the selection signal

Super Lawyers appeared in 33 percent of attributable answers, far ahead of everything else. Its selection process combines peer nominations, independent research, and peer evaluation, and names roughly the top 5 percent of lawyers per state; inclusion cannot be purchased, only the marketing of it can. For a language model deciding which firm is safe to name, a list that money cannot buy is close to the ideal citation.

The practical consequence: selection is earned years before it pays off. The inputs its researchers score, verdicts, credentials, bar activity, publications, are the same evidence trail that supports every other visibility surface a firm cares about.

Justia: the structure signal

Justia appeared in 20 percent of answers, and claiming a baseline profile in its lawyer directory costs nothing. Its weight comes from being one of the largest, cleanest, most crawlable legal datasets on the open web: consistent URL structure, complete fields, and profiles for nearly every licensed attorney sitting inside a portal of statutes and case law that engines already trust.

That makes Justia the cheapest meaningful move in legal GEO. A complete, accurate, fully filled-out free profile puts a firm inside a source the engines demonstrably read. An empty or inconsistent one wastes the citation the engine was prepared to give.

Tier two: the supporting cast

The middle tier earns single-digit weight individually, but its real function is corroboration. Engines grow confident when several independent sources describe the same firm the same way, so each additional trusted surface that confirms your firm's name, focus, and credentials compounds the ones above it.

Attorney at Law Magazine, at 7 percent, is the surprise of the tier: a trade publication cited ahead of several directories firms pay far more for. The engines appear to treat its editorial coverage and directory as a distinct voice rather than another listing, which is consistent with how they weight earned coverage generally.

Best Law Firms at 6 percent and Best Lawyers at 4 percent are two properties of the same peer-review company, which has run the Best Law Firms rankings independently since its split from U.S. News. Notably, the U.S. News lawyer directory itself did not appear in our sample; the citation weight followed the rankings to their new domain rather than staying with the bigger brand.

Avvo, at 5 percent, is the strongest of the older consumer directories, and its profiles are free to claim. Yelp at 4 percent and Expertise.com at 3 percent round out the tier, with the study's broader finding worth repeating here: Reddit, at 8 percent, outweighed all three.

Tier three: the famous names the engines skip

The directories with the biggest brands among lawyers carried almost no weight with the engines. FindLaw was cited in roughly 1 percent of answers. Martindale-Hubbell, the oldest name in legal directories, appeared in under 1 percent. Lawyers.com, Nolo, and HG.org did not appear in a single logged answer.

This is the gap between two kinds of reputation. Lawyers know these names because they have been sold to lawyers for decades, in Martindale's case since the 1800s. Engines do not care about that history; they retrieve, parse, and corroborate pages at answer time, and they are evidently finding what they need elsewhere.

The honest caveats: this is one practice area, five metros, one window, and a zero in our sample is not a zero forever. Some of these companies also sell things a listing does not measure, such as websites, intake tools, or lead programs, which deserve their own cost-benefit math. But if the line item in your budget says visibility, the measured answer is that these pages are not where the engines look.

The consolidation nobody noticed

One company now owns both ends of this ranking, and most firms buying directory listings have no idea. In December 2024, Thomson Reuters closed the sale of its FindLaw business, including Super Lawyers and LawInfo, to Internet Brands, the KKR-backed company that already owned Avvo, Martindale-Hubbell, Nolo, and Lawyers.com.

That puts the most-cited legal directory and four of the least-cited under the same roof. It is the clearest evidence in the whole dataset that citation weight belongs to domains, not owners. A shared parent company did not lift FindLaw or Nolo an inch, because each domain earns or loses the engines' trust on its own crawlability, structure, and independence.

For firms, the practical implication is vigilance. When one owner controls most of the industry's listing surfaces, bundles become tempting: a package that pairs the profile you need with several you measurably do not. Buy the domain, not the brochure.

Where to put your directory budget

The measured weights convert directly into a sequence: claim the free tier everywhere, earn what cannot be bought, and make every profile agree.

The same discipline applies at the level of individual answers. Wherever an engine names a competitor and cites a directory you are absent from, that is a gap with a known fix, and the per-vertical playbooks such as how personal injury firms get cited cover the on-site half of the equation. The NAP consistency work behind card three is the same project that drives Local Pack rankings, detailed in the Map Pack playbook.

Where directory spending sits in the larger budget is its own question. Listings are a small, corroborating slice of the rent-versus-equity split we lay out in what lawyer marketing budgets should look like; they support the compounding channels, they do not replace them.

What citation weight does not tell you

Citation weight measures one thing: where the engines look when they recommend lawyers. It is not a verdict on any directory's whole business. A directory can carry little AI weight and still send profitable human referrals, and a heavily cited one can send visitors who never convert. The two measurements answer different questions and belong in the same spreadsheet.

The sample has edges worth restating. It covers personal injury, five metros, and a two-week June 2026 window, logged from clean sessions. Engines change retrieval sources often, so expect movement between runs; we re-run the study quarterly and will update this ranking when the data changes. And none of this measures the quality of any directory's editorial work or of any firm listed on it.

What the ranking does settle is the burden of proof. Directory spending has run on brand familiarity for decades. Now that citation weight can be measured, the question for every renewal is no longer whether lawyers have heard of the directory. It is whether the engines read it.