Introduction

Your clients are asking AI which law firm to hire. Do you know what it’s telling them?

That question is no longer hypothetical. Across every major AI platform — Perplexity, ChatGPT, Google AI Overviews — millions of potential legal clients are typing variations of “what is the best [practice area] law firm in [city]” and receiving confident, specific answers. Some firms appear consistently. Many don’t appear at all. And most firms have no idea which category they fall into.

The legal marketing world has responded to this uncertainty the way it usually responds to new technology — with speculation, extrapolation, and no shortage of advice that amounts to “publish more content” or “optimize your schema markup.” Some of that advice is useful. Most of it is untested. Almost none of it is grounded in data about how AI platforms actually behave when answering legal market queries in the real world.

We decided to find out.

Over the course of several days in late March 2026, Splat ran more than 120 queries across three AI platforms — Perplexity, ChatGPT, and Google AI Overviews — covering six practice areas in nine markets spanning three market tiers. We captured which firms appeared, which sources each platform cited, and how those patterns varied by market size, practice area, and platform. We ran every query cold, in fresh sessions, to eliminate the context contamination that would skew results. When the same query produced different results on back-to-back runs — which it sometimes did — we captured both versions and noted the variation.

What we found surprised us, and it will probably surprise you too.

AI platforms are not creating a new visibility hierarchy. They are reflecting the one that already exists.

That finding has significant implications — for how firms should think about AI visibility strategy, for which markets and practice areas represent the highest-leverage opportunities, and for why the advice to “just create more content” is incomplete at best and misleading at worst.

This report presents what we found, what it means, and what firms should do about it.

Methodology

Good research starts with good decisions about scope. Here is what we studied, why we made the choices we made, and what you need to know to interpret the findings correctly.

Markets and Market Tiers

We organized our nine markets into three tiers based on size and legal market depth. Large markets — New York, Chicago, and Los Angeles — represent the most competitive legal environments in the country, with deep directory coverage and significant BigLaw presence. Mid markets — Philadelphia, Atlanta, and Denver — are substantial regional markets with their own distinct ecosystems. Small markets — Boise, Providence, and Tucson — represent the thinner end of the coverage spectrum, where national directory presence becomes sparse and local signals take on greater weight.

The tier structure was not incidental to the research design. We expected market size to matter. What we did not expect was how cleanly and consistently it would predict source ecosystem — a finding we’ll return to in detail.

Practice Areas

We selected six practice areas designed to span the full spectrum from consumer-facing to institutional: personal injury, family law, criminal defense, bankruptcy, estate planning, and corporate law. The first three are primarily consumer-facing — individuals searching for representation. The last three have significant institutional dimensions — businesses, high-net-worth individuals, sophisticated buyers of legal services. Corporate law was added specifically as a B2B control, to test whether the patterns we observed in consumer practice areas held when the client profile changed.

Platforms

We ran queries on three platforms: Perplexity, ChatGPT, and Google AI Overviews. These three were selected because they represent meaningfully different approaches to answering AI queries — not three versions of the same thing.

Perplexity is retrieval-first. When you submit a query, Perplexity runs a live web search, retrieves current pages, and generates a response grounded in what it just fetched. Think of it as a search engine that writes prose summaries with footnotes. Its training data shapes how it interprets results, but the output is anchored in current web content.

ChatGPT is training-data-first. The model was trained on an enormous corpus of text assembled before a knowledge cutoff date. When you ask a question without explicitly invoking its browsing tool — which we did not do — it draws on what it learned during training. The citations it produces in follow-up responses are not live retrievals — they are the model’s best recollection of what sources informed its knowledge.

Google AI Overviews sits on top of Google’s existing search infrastructure rather than operating as a standalone model. Google runs its standard search ranking algorithm, identifies the most relevant pages for a query, and then passes those pages to its Gemini model, which synthesizes them into a prose summary. It is, in a meaningful sense, AI-powered Google Search rather than a genuinely independent AI system.

A Note on Non-Determinism

AI platforms do not produce identical results every time you run the same query. Temperature settings, index updates, and variations in the model’s sampling process all introduce variability. This is not a flaw — it is an inherent characteristic of how large language models work.

Within a single query session, Perplexity returned two dramatically different responses to the same Atlanta bankruptcy question. The first was purely consumer-facing — no BigLaw firms, no Chambers citations, no corporate restructuring presence. The second returned a full corporate/consumer split, with Alston & Bird, Kilpatrick Townsend, and King & Spalding appearing at the top alongside Chambers citations. The corporate tier literally appeared and disappeared within the same session.

Throughout this report, we focus on patterns rather than treating any individual result as definitive.

Finding 1: AI Platforms Are Mirrors, Not Arbiters

The most important thing we can tell you about AI visibility in the legal market is also the most counterintuitive: the firms that show up most consistently in AI results are not the firms that figured out how to game a new algorithm. They are the firms that spent years — in some cases decades — building authority within the credentialing systems the legal industry already trusts.

Chambers Band 1. Best Law Firms Tier 1. Super Lawyers. Best Lawyers. Legal 500. These directories did not become relevant because of AI. They became relevant long before AI existed, because they represented the legal profession’s best attempt to answer a genuinely hard question: how do you evaluate the quality of a law firm when you can’t directly observe the work?

What AI platforms have done is inherit that answer.

When Perplexity retrieves a page to answer “what is the best corporate law firm in Philadelphia,” it is not making an independent judgment about legal quality. It is retrieving pages that rank highly in its index — and those pages are dominated by Chambers rankings, Best Law Firms tier lists, and similar credentialing content, because those sources have spent years building the domain authority and indexable content that AI retrieval systems reward.

When ChatGPT answers the same question, it is drawing on a training corpus assembled from authoritative sources across the internet — and the most authoritative sources for legal market information are, again, Chambers, Best Law Firms, Vault, Legal 500, and Lawdragon. The firms that appear in ChatGPT results are the firms that appeared in those publications consistently enough to be encoded in the model’s training data.

When Google AI Overviews synthesizes an answer, it is summarizing the pages that Google’s search ranking algorithm surfaced — and for institutionally complex practice areas like corporate law and estate planning, those pages are dominated by the same directory and credentialing sources.

The platforms are different. The underlying authority structure they reflect is largely the same.

This has a clarifying effect on how firms should think about AI visibility strategy. The question is not “how do I rank in AI?” The question is “how do I build the kind of authority that AI platforms recognize?” And that question has the same answer it has always had — peer recognition, directory presence, published work, and sustained visibility in the sources that the legal industry treats as credible.

AI visibility in the legal market is not a new game. It is the same game — building genuine authority within the systems the legal industry already uses to evaluate quality — played on new scoreboards.

There is one important caveat. The mirror is not perfect. Different platforms reflect different facets of the same authority structure. Perplexity reflects current web visibility. ChatGPT reflects historical depth. Google AI Overviews reflects search ranking signals — which skew toward SEO-optimized content and can be more susceptible to firms with heavy digital marketing investment than to firms with genuine peer-reviewed authority. We will return to these platform differences in Finding 5.

Firms that have invested in that authority will find AI working in their favor. Firms that have not will find AI reflecting that gap back at them, consistently, across every platform we tested.

Finding 2: Market Tier Is the Strongest Predictor of AI Visibility

If you want to know what drives AI visibility for a law firm, the single most useful piece of information is not the practice area, not the platform, and not the firm’s marketing budget. It is the market the firm operates in.

Across more than 120 queries on three platforms, market tier emerged as the clearest and most consistent predictor of which sources drive AI results — and therefore which investments are most likely to improve a firm’s visibility. Large markets, mid markets, and small markets each operate according to a distinct source ecosystem, and understanding which ecosystem your firm lives in is the prerequisite for any meaningful AI visibility strategy.

Large Markets: The Directory Bar Is High

In New York, Chicago, and Los Angeles, AI results are dominated by a small number of highly selective national credentialing systems. Chambers, Best Law Firms, Best Lawyers, Super Lawyers, and Legal 500 account for the vast majority of citations across practice areas and platforms. Firm websites are essentially invisible as direct citation sources — with rare exceptions, they do not appear in AI results for large market queries regardless of how well optimized they are.

The implication is uncomfortable but important: Chambers Band 1 for corporate law in New York is not something a firm earns with a content marketing campaign. It is earned through doing distinguished work, cultivating client relationships, and building a reputation that survives peer scrutiny over multiple annual research cycles. Firms in large markets that are not yet in these directories should treat directory submissions as their highest-leverage marketing investment — not because of AI, but because AI reflects what those directories say.

Mid Markets: National Directories Plus Local Editorial

Philadelphia, Atlanta, and Denver show a more accessible ecosystem. The national directories — Best Law Firms, Super Lawyers, Best Lawyers — remain important, but they are joined by a category of sources that does not appear in large market results: local editorial publications. The Philadelphia Inquirer, Attorney at Law Magazine, 5280 Magazine in Denver, ColoradoBiz, and the Atlanta Magazine Super Lawyers special section all appeared as citation sources in mid-market results. These are not vanity placements. They are functioning trust signals that AI platforms surface alongside national directory citations.

This is significant for mid-market firms because these local editorial sources are genuinely accessible in a way that Chambers Band 1 is not. Getting recognized in Attorney at Law Magazine or earning a mention in 5280’s Top Lawyers feature requires effort and some degree of peer standing, but it does not require the multi-year institutional investment that Chambers demands. For mid-market firms, local editorial presence is the highest-leverage gap to close.

Small Markets: Websites as Trust Signals

In Boise, Providence, and Tucson, the ecosystem shifts again — and this time the shift is dramatic. National directory coverage is thin. Chambers publishes rankings in some practice areas but not others. In their place, a source category appears that is essentially invisible in large and mid markets: firm websites.

In small markets, a well-structured, well-indexed law firm website functions as a direct trust signal for AI platforms. We observed firm websites appearing as primary citations — not as background context, but as the specific sources driving firm recommendations — across multiple practice areas in Boise, Providence, and Tucson. This does not happen in Philadelphia or Chicago. It happens in Boise because Perplexity, when it cannot find authoritative directory content, reaches for whatever credible content is available — and in thin-coverage markets, the firm’s own website is sometimes the most substantive source in the index.

In New York, a great website is table stakes. In Boise, it may be your most powerful AI visibility asset.

For the law firm seeking greater visibility, this condition offers real opportunity. Small-market firms that invest in well-structured, content-rich websites — practice area pages with genuine depth, attorney biographies that establish credentials, clear geographic signals — are positioning themselves to compete directly for AI visibility in a way that their large-market counterparts cannot.

One finding cuts across all three market tiers and deserves its own note. In bankruptcy and criminal defense — the two practice areas with the most distinct institutional and consumer dimensions — AI results consistently show two separate tiers. The corporate/institutional tier is dominated by Chambers, Legal 500, and Best Law Firms. The consumer tier is dominated by Avvo, Justia, Super Lawyers, and review platforms. In small markets, the split largely disappears because Chambers doesn’t publish rankings for most practice areas there.

Finding 3: Estate Planning — The Clearest Pattern in the Data

If you are looking for the single most defensible finding in this study, this is it.

Across eight of nine markets tested on Perplexity, the firm ranked first for estate planning was either Band 1 in the Chambers High Net Worth Guide for Private Wealth Law or directly associated with a Band 1 firm. New York, Chicago, Los Angeles, Philadelphia, Atlanta, Denver, Boise, and Providence — eight markets across all three tiers — produced the same result. Where Chambers publishes a private wealth ranking, it dominates the top position. Not usually. Not mostly. Every time.

The one exception is Tucson. Chambers does not publish private wealth rankings for the Tucson market. In the absence of that signal, the ecosystem reverts to the small market pattern — boutique firms with strong local visibility and client review presence rising to the top. The exception proves the rule: Chambers dominates estate planning AI results everywhere it has coverage, and its absence is immediately visible in the results.

This pattern held on ChatGPT as well, with one notable divergence. For Denver estate planning, ChatGPT surfaced Evans Case LLP rather than Holland & Hart — which Perplexity identified as Chambers Band 1. The divergence reflects the platform difference: Perplexity retrieved the current Chambers rankings page; ChatGPT’s training data encoded a different set of prominent Denver estate planning firms. The Chambers signal was present on both platforms — it expressed itself differently depending on how each platform accesses information.

On Google AI Overviews, the pattern broke down entirely for small markets. Boise estate planning on Google AI Overviews returned consumer-facing boutiques with strong SEO — Holland & Hart, Chambers Band 1 and the consensus top firm on both Perplexity and ChatGPT, did not appear at all.

For firms that already hold Chambers Band 1 or Band 2 private wealth rankings, the finding is validating — that investment is paying dividends in AI visibility across multiple platforms. For firms that do not hold Chambers rankings but aspire to compete in the estate planning market, the finding is clarifying. The path to AI visibility in this practice area runs directly through Chambers. There is no shortcut visible in the data.

Finding 4: Small Market Websites as Trust Signals

In large markets, a law firm’s website is background. It may confirm what a prospective client already found in a directory. It is rarely, if ever, the reason a firm appears in an AI result.

In small markets, that changes.

Across our Boise, Providence, and Tucson queries on Perplexity, firm websites appeared repeatedly as primary citation sources — not as supplementary context, but as the specific pages driving firm recommendations. This is not something we observed in New York or Chicago or Los Angeles. It is a small market phenomenon, and it has a straightforward explanation.

AI retrieval systems are looking for authoritative content about a query. In large markets, that content is abundant — Chambers rankings pages, Best Law Firms tier lists, Super Lawyers directories, Legal 500 editorial commentary. The supply of credentialing content is so deep that firm websites rarely need to do that work. In small markets, that supply thins dramatically. When the authoritative directory content is thin, AI retrieval systems reach for whatever substantive, well-indexed content is available — and in thin-coverage markets, that content is often the firm’s own website.

This creates a genuine opportunity that does not exist at the same scale in larger markets. A small-market law firm with well-structured practice area pages, substantive attorney biographies, clear geographic signals, and regularly updated content is not just improving its traditional SEO. It is positioning itself as a primary source for AI retrieval in a market where the competition for that position is relatively limited.

The contrast with large markets is worth sitting with for a moment. A Chicago personal injury firm competing for AI visibility is effectively competing with Chambers rankings, Best Law Firms tier lists, and decades of accumulated directory authority. No amount of website optimization closes that gap in the short term. A Boise personal injury firm competing for AI visibility is competing with a much thinner field — and its website content is a meaningful part of that competition.

This does not mean small market firms should abandon directory investment. Super Lawyers recognition, Best Lawyers listings, and Best Law Firms tier placement all appeared in our small market results and contribute meaningfully to AI visibility. But those investments take time. Website content can be improved in weeks.

One practical note on what “well-structured” means in this context. The firm websites that appeared as AI citation sources in our small market results were not sophisticated content marketing operations. They were firms with clear practice area pages that described what the firm does, attorney pages that established credentials and experience, and enough geographic specificity that AI retrieval systems could confidently associate the firm with the relevant market. The bar is not high. But it must be cleared.

Finding 5: Three Platforms, Three Different Answers

Run the same query on Perplexity, ChatGPT, and Google AI Overviews. You will not always get the same answer.

This is not a bug. It is the natural consequence of three platforms that approach the same question through fundamentally different lenses. Understanding those differences is not academic — it has direct implications for where firms should focus their visibility investment and what results they should realistically expect from each platform.

The clearest single example in our dataset is New York criminal defense. On Perplexity, the top results were dominated by Chambers-ranked white-collar boutiques. On Google AI Overviews, none of those firms appeared. Instead, the results were led by Spodek Law Group — a firm with heavy SEO investment and a well-optimized website, whose domain appeared multiple times in the source cards driving the AI Overview. On ChatGPT, the results reflected a hybrid — drawing from both the peer-reviewed prestige ecosystem and the media prominence ecosystem.

Same query. Same market. Same practice area. Three meaningfully different answers.

Boise estate planning shows the pattern most starkly. On Perplexity, Holland & Hart appeared at #1, driven by its Chambers Band 1 private wealth ranking for Idaho. On ChatGPT, Holland & Hart also appeared, driven by the same Chambers signal encoded in its training data. On Google AI Overviews, Holland & Hart did not appear at all. Consumer-facing boutiques with strong local SEO filled the result set entirely. The firm with the strongest peer-reviewed credentials in the market was invisible on the platform most susceptible to search ranking signals.

Perplexity measures current web authority. Firms that appear prominently in directories, with current and well-indexed listings, will perform well on Perplexity. A firm that has allowed its directory presence to lapse or has never invested in it will not.

ChatGPT measures historical depth. Its training corpus was assembled from authoritative sources across the internet before a knowledge cutoff, and the firms that appear most consistently in its results are the firms that appeared most consistently in the publications that shaped that corpus — Chambers, Vault, Lawdragon, Legal 500, the American Lawyer, and similar trade publications. A firm cannot directly influence what ChatGPT learned. But consistent presence in authoritative publications over time means representation in future training cycles as models are updated.

Google AI Overviews measures search ranking signals. It is, in a meaningful sense, AI-powered Google Search — and the firms that perform well in it are the firms that perform well in traditional organic search. For consumer-facing practice areas, that means firms with SEO-optimized websites, high review volume, and strong local search presence.

A firm that optimizes for one platform is not necessarily optimizing for the others. The most defensible multi-platform strategy combines both: sustained investment in peer-reviewed credentialing systems for Perplexity and ChatGPT visibility, combined with the fundamentals of local SEO and review management for Google AI Overviews visibility. Neither alone is sufficient. Together, they address the full spectrum of how prospective clients are likely to encounter AI-generated legal market recommendations.

Finding 6: Google AI Overviews and the SEO Factor

Google AI Overviews is the most widely encountered AI platform in the legal market. It requires no account, no deliberate choice to use an AI tool, and no behavioral change from a prospective client. It appears automatically in Google Search results — the place where most people still begin their search for a law firm — and it synthesizes an answer before the user has clicked a single link. For that reason alone, it deserves careful attention.

Because Google AI Overviews synthesizes content from pages that Google’s search algorithm has already ranked highly, it inherits Google’s search ranking biases in full. For consumer-facing practice areas — personal injury, family law, criminal defense, bankruptcy — those biases favor firms with heavy SEO investment, high review volume, and well-optimized websites. The result is a set of AI recommendations that often looks more like a Yelp search than a Chambers ranking.

The most striking examples from our data involve firms that appeared at or near the top of Google AI Overviews results despite appearing on neither Perplexity nor ChatGPT. Cordell & Cordell — a national franchise firm focused exclusively on representing men in divorce cases — appeared at #1 for Philadelphia family law and in Boise family law results on Google AI Overviews. It did not appear in either market on Perplexity or ChatGPT. Its appearance is not a reflection of peer-reviewed standing or directory authority. It is a reflection of SEO investment.

Tatum Wysocki Law appeared at #1 for Boise criminal defense on Google AI Overviews, citing recognition as “Best Defense Law Firm in Idaho” by a local consumer poll. It appeared on neither Perplexity nor ChatGPT for the same query. Nevin Benjamin & McKay — the most consistently peer-validated criminal defense firm in the Boise market across both Perplexity and ChatGPT — did not appear on Google AI Overviews at all.

These are not isolated anomalies. They are the predictable output of a system that inherits search ranking signals, and search ranking signals in consumer legal services are heavily influenced by marketing investment rather than legal quality.

The clearest systematic expression of this pattern appeared in our bankruptcy queries. Both New York and Philadelphia bankruptcy queries on Google AI Overviews returned exclusively consumer-facing attorneys. No BigLaw restructuring firms appeared. No Chambers-ranked practices appeared. Kirkland & Ellis — the consensus top firm for corporate bankruptcy across both Perplexity and ChatGPT in New York — was entirely absent.

This is not because Google AI Overviews failed to understand the query. It is because the pages that rank highest in Google Search for “best bankruptcy law firm in New York” are dominated by consumer-facing content. BigLaw restructuring firms do not invest in consumer SEO. They do not accumulate Yelp reviews. And so they do not appear in the Google results that feed Google AI Overviews.

One practice area behaves differently on Google AI Overviews, and it is worth acknowledging. Corporate law queries produced results that closely matched Perplexity and ChatGPT — Chambers-ranked firms leading, Vault prestige rankings present, the same elite firms appearing across all three platforms. The reason is straightforward: Chambers and Vault have invested heavily in SEO for corporate law content, and their pages rank highly in Google Search for corporate law queries. For corporate law practices, the three platforms are pulling in the same direction.

Strategic Recommendations

The findings in this study point toward a set of recommendations that are more differentiated than most AI visibility advice currently circulating in the legal market. The standard guidance — publish more content, optimize your schema markup, claim your Google Business Profile — is not wrong. It is incomplete. What the data shows is that the right investments depend heavily on where your firm operates, what you practice, and which platforms your prospective clients are most likely to use.

Here is what the data actually supports.

Start with an Honest Assessment of Your Market Tier

Before making any AI visibility investment, firms need to understand which ecosystem they are operating in. The three tiers we identified are not just descriptive — they are prescriptive. The investments that move the needle in Boise are not the same investments that move the needle in New York.

Large market firms — New York, Chicago, Los Angeles, and comparable markets — operate in an ecosystem where national directory authority is the primary driver of AI visibility across all platforms except Google AI Overviews. If your firm is not in Chambers, Best Law Firms, or Best Lawyers for your primary practice areas, those submissions represent your highest-leverage AI visibility investment. Not because of AI specifically, but because AI reflects what those directories say. The path runs through the directories, not around them.

Mid-market firms — Philadelphia, Atlanta, Denver, and comparable markets — have a more accessible opportunity set. National directory investment remains important, but local editorial presence is a meaningful and achievable gap to close. Attorney at Law Magazine, city magazine Top Lawyers features, regional Super Lawyers recognition, and similar local editorial placements appeared consistently in our mid-market results. These are not vanity placements — they are functioning trust signals that AI platforms surface.

Small market firms — Boise, Providence, Tucson, and comparable markets — have the most immediate opportunity available to any tier. Website content investment is the highest-leverage action a small market firm can take. Practice area pages with genuine depth, attorney biographies that establish credentials clearly, geographic specificity, and regularly updated content all contribute to the kind of well-indexed, substantive website presence that AI retrieval systems surface when directory content is thin. This is achievable in weeks rather than years, and the data shows it works.

Differentiate Your Strategy by Platform

Beyond market tier, firms should think about which platforms their prospective clients are most likely to use — and calibrate investments accordingly.

For Perplexity visibility, the investment is directory authority. Chambers, Best Law Firms, Best Lawyers, Super Lawyers, Legal 500, and similar peer-reviewed directories are the primary sources Perplexity retrieves. Firms should ensure their directory profiles are claimed, complete, and current across all relevant platforms for their practice areas and markets.

For ChatGPT visibility, the investment is authoritative publication presence over time. ChatGPT’s training data encoded the legal industry’s most authoritative publications — Chambers, Vault, Lawdragon, Legal 500, the American Lawyer, and regional legal trade publications. A firm cannot directly influence what ChatGPT learned from its training corpus, but consistent presence in those publications over time means representation in future training cycles as models are updated. Treat media relations and directory submissions as long-term brand investments, not short-term tactics.

For Google AI Overviews visibility, the investment is traditional local SEO. Google Business Profile maintenance, review management, well-structured and regularly updated website content, and the fundamentals of local search optimization all drive Google AI Overviews results for consumer-facing practice areas.

Match Your Practice Area to Your Investment Priority

The data also supports practice-area specific recommendations that cut across market tiers.

For estate planning practices, Chambers High Net Worth Guide submission is the single most important AI visibility investment available regardless of market size. The consistency of the Chambers signal across eight of nine markets on Perplexity and ChatGPT is unambiguous. Firms that are not yet in Chambers should treat submission as a multi-year project and begin immediately. Firms already in Chambers should ensure their profile is current and their ranked attorneys are prominently featured in their web presence.

For corporate law practices, the investment in Chambers, Vault presence, and Best Law Firms tier rankings pays dividends across all three platforms simultaneously. Corporate law is the one practice area where the three platforms converge — the same firms appear consistently regardless of which platform a prospective client uses. Directory investment for corporate law has the broadest reach of any practice area investment in the study.

For consumer-facing practice areas — personal injury, family law, criminal defense, and consumer bankruptcy — the platform split matters most. Perplexity and ChatGPT reward peer-reviewed directory standing. Google AI Overviews rewards SEO investment and review volume. Consumer clients searching for a personal injury attorney on their phone are more likely to encounter a Google AI Overview than a Perplexity result.

The Compounding Effect

One final observation worth making explicit: the investments that drive AI visibility are not AI-specific investments. Directory submissions, peer recognition, local editorial presence, website content, and review management are all things firms should be doing regardless of AI. What AI has done is make the consequences of neglecting those investments more visible — and more immediate. A firm that has underinvested in directory presence for years will find that underinvestment reflected back at it every time a prospective client asks an AI platform for a recommendation.

The good news is that the data points toward concrete, achievable actions at every market tier. Large market firms can audit their directory standing and prioritize submissions for the practice areas where gaps exist. Mid-market firms can identify the local editorial outlets that appeared in our results for their markets and develop a plan to earn recognition in them. Small market firms — and this finding deserves emphasis — have perhaps the most immediately actionable opportunity of all: a well-structured, content-rich website is not a years-long project. It is something that can be meaningfully improved in weeks, and the data shows it produces direct, measurable AI visibility gains in thin-coverage markets.

The firms that will benefit most from AI visibility are not the ones waiting for the landscape to stabilize or for a definitive playbook to emerge. They are the ones that recognize what the data already shows — that the path to AI visibility runs through the same fundamentals that have always driven legal market authority — and start walking it now.

A Note on the Directory Ecosystem

Throughout this report we have referenced a range of directories and credentialing systems — Chambers, Best Law Firms, Super Lawyers, Vault, Lawdragon, Legal 500, and others — without fully explaining how each works, what it measures, or how firms can pursue recognition in each. That explanation matters, particularly for firms that are early in their directory investment journey and need to understand what they are actually buying into before they commit resources.

We will be addressing that in a forthcoming companion piece — The Legal Directory Primer for Law Firms — which covers the most influential directories that appeared in our research: their methodologies, their submission processes, their relative authority, and their strategic value for firms at different market tiers.

Appendix — Methodology Notes

Dataset

This study captured results from more than 120 queries run across three AI platforms in late March 2026. Perplexity and ChatGPT were run across all nine markets and six practice areas — 54 queries each. Google AI Overviews was run across three representative markets (New York, Philadelphia, and Boise) covering all six practice areas — 18 queries. For each ChatGPT query, a follow-up attribution prompt was run in the same session, producing source-level disclosure for every firm recommendation.

Run Types

Perplexity queries were run cold — in fresh browser sessions with no prior conversation history — to eliminate context contamination. Early New York and Philadelphia Perplexity queries were initially run in a contextualized single thread and later re-run cold. Where both versions exist in the dataset, cold run data is treated as primary. ChatGPT queries were run in fresh conversations without invoking the browsing tool. Google AI Overviews queries were run in standard Chrome browser sessions.

Non-Determinism

AI platforms do not produce identical results on every run. Temperature settings, index updates, and model sampling variation all introduce result variability that is independent of methodology. Where a query produced materially different results between runs — most notably Atlanta bankruptcy, which produced a pure consumer result in one run and a full corporate/consumer split in another — both versions were captured and noted. Throughout the report, findings are stated as patterns observed across multiple queries rather than conclusions drawn from individual results.

Platform Limitations

Perplexity citations reflect live web retrieval at the time of the query and may not be reproducible at a later date as index content changes. ChatGPT citations are training-data-generated references, not verified live retrievals — they reflect what the model learned from its training corpus rather than what any source currently says. We cannot confirm with absolute certainty that ChatGPT did not supplement any response with live retrieval, though no browsing indicator appeared in any session. Google AI Overviews results reflect Google’s search ranking algorithm at the time of the query and are subject to the same variability as organic search results.

Google AI Overviews Sample

Google AI Overviews data covers three of nine markets. Findings from this platform are therefore directional rather than comprehensive. The patterns observed across New York, Philadelphia, and Boise were consistent enough to support the conclusions we draw, but we note the smaller sample size and encourage readers to treat Google AI Overviews findings as indicative rather than definitive.

Scope

This study focuses exclusively on the legal industry. Findings should not be extrapolated to other professional services verticals without independent research. Market tier classifications (large, mid, small) reflect legal market depth and directory coverage density rather than population size alone.