How to Get ChatGPT to Recommend Your Law Firm

A growing number of people now ask an AI system a question before they ever call a lawyer. "My ex won't let me see the kids, what are my options in Queensland?" "Who are the best estate litigation lawyers in Sydney?" "Is it worth getting a lawyer for a workers compensation claim?"

The answer they get back often names firms. Some firms are named again and again. Most are never mentioned at all.

Partners are now asking us, quite reasonably, how to become one of the firms that gets named. This article is our answer. It draws on the best recent research available, including a large 2026 study of how AI systems recommend software products, and translates it into what a law firm should actually do, in order.

The short version: there's no trick. The firms AI systems recommend tend to be the firms that already rank well in Google, are talked about across the web, have a clear and consistent identity, and publish genuinely useful answers to the questions clients ask. The work is familiar. The emphasis has changed.

How many clients actually use AI to find a lawyer?

Enough to matter, not enough to panic.

In iLawyerMarketing's 2026 survey of 1,110 US consumers, 42% said they would use ChatGPT to research which lawyer to hire, up from 9% in 2023. Half said they would use at least one AI answer engine. Google was still the most-used source, at 72% (iLawyerMarketing, 2026).
Clio's 2025 Legal Trends Report found that among consumers who had asked AI a legal question, 28% were told to contact a lawyer (Canadian Lawyer, reporting Clio's 2025 Legal Trends Report).
Pew Research Center found that when Google showed an AI summary, users clicked a traditional search result on 8% of visits, against 15% when no summary appeared. They clicked a link inside the summary only about 1% of the time (Pew Research Center, 2025).

Read those numbers carefully. The iLawyerMarketing figures are stated intentions from US consumers, published by a legal marketing firm, not observed hiring behaviour, and nothing equivalent has been published for Australia yet. But the direction is consistent across every source we've seen: AI is becoming one of the places clients form a shortlist, and the answer they read increasingly replaces the click they used to make.

AI doesn't need to replace Google to matter. It only needs to influence a meaningful share of how clients choose. For a firm whose growth depends on being considered, that's already the case.

First, stop thinking about "ranking in ChatGPT"

Most of the advice circulating about AI search borrows the language of SEO: rank, position, number one. That model doesn't fit.

Ask ChatGPT the same question twice and you may get a different list of firms. Ask Gemini, Claude and Perplexity and you'll almost certainly get different lists again. In the 2026 Minuttia and Convex Works study of 24,067 prompts across Google's AI Overviews, ChatGPT, Gemini, Claude and Perplexity, about 54% of brands recommended in answers to "alternatives to X" prompts were named by only one engine, and only 8% were named by all five (Minuttia AI Visibility Data Study, 2026).

As I've argued on Law By Dan, if the same question rarely produces the same list, a single list isn't a ranking (Why "ranking in ChatGPT" may be the wrong mental model). The more useful questions are:

How often does your firm appear when clients ask the questions that matter to your practice?
How strongly is it presented? Merely mentioned, suggested as an option, explicitly recommended, or only cited as a source in the background? (That's the classification FirmRanker uses, and the distinction matters; see what actually counts as an AI recommendation.)
How consistently across different AI systems and over time?
What does it say about you, and is it accurate?

Treat any agency that sends you a single screenshot as proof of "ranking number one in ChatGPT" with caution.

What the research says drives AI recommendations

The most rigorous large-scale research so far comes from outside legal services, mostly from studies of how AI systems recommend software products. The products are different, but the mechanics of how AI systems choose what to name are the same, so the findings transfer reasonably well. Here's what stands out.

1. Google visibility still matters a great deal

In the Minuttia study, brands ranking in Google's organic top 10 for a prompt were 6.8 to 11.6 times as likely to be named by an AI engine, depending on the engine. In Google's AI Overviews, 38.5% of cited domains also appeared in that search's organic top 10 (Minuttia, 2026).

That makes sense once you know how these systems find information. ChatGPT search, for example, uses third-party search providers, including Bing, to find web pages before writing its answer. If your firm can't be found by conventional search, it's much less likely to be found by the AI systems built on top of it.

The researchers are careful to say these are associations, not proven causes. But the practical implication is clear: AI visibility isn't a replacement for SEO. It's built on it.

2. Being talked about matters more than links

Ahrefs analysed 75,000 brands and found that the number of times a brand was mentioned across the web had the strongest correlation with appearing in Google's AI Overviews (0.664), far stronger than backlinks (0.218). Branded search volume also correlated positively (0.392) (Ahrefs, 2025).

AI systems form a view of who your firm is from everything said about it: your website, directories, news coverage, other firms' references, community and industry sites, and reviews. A firm that's discussed widely and consistently is easier for a model to recognise and recommend.

3. The number of reviews matters more than the star rating

This finding from the Minuttia study is the one law firms should pay most attention to. Within a software category, brands in the top quarter for review volume on G2 were 7.6 times as likely to appear in AI Overviews for category prompts as those in the bottom quarter. Star rating showed no comparable association (Minuttia, 2026).

Law firms don't have a G2. The closest equivalents are Google reviews and the legal directories in your jurisdiction. We'd expect the same pattern to apply, a steady volume of genuine, recent reviews mattering more than a perfect average, but that's our inference, not a tested finding for legal services.

4. AI systems cite comparison content, and much of it is self-serving

In Minuttia's follow-up study, vendors listing their own product accounted for between 34% (ChatGPT) and 47% (Google Search and Gemini) of the listicles cited for "alternatives to X" questions (Minuttia, 2026). In other words, a lot of "best of" content that AI systems rely on is written by the companies being ranked.

5. Clear, answer-first content is easier to use

AI systems don't read a page the way a person does. They pull out the passages that answer the question. Pages that answer the question immediately, are easy to scan, and are written so that each paragraph makes sense on its own are easier for an AI system to use and cite. That's consistent with what we see on law firm sites.

What this means for your law firm: the practical plan

Here's how we'd sequence the work, using Practice Proof's GET FOUND → GET CHOSEN → RUN BETTER → MORE & BETTER MATTERS framework. The order matters. Most firms that come to us about AI search have a problem further up the list.

Step 1: Find out what AI currently says about you

Before changing anything, look.

Write down 20 to 40 questions your best clients actually ask before they hire a lawyer. Use their words, not marketing language. Your intake team and recent enquiry forms are the best source. Include:

Category questions: "Who are the best family lawyers in Parramatta?"
Problem questions: "My business partner is taking money from the company, what can I do in Victoria?"
Evaluation questions: "Is [your firm] experienced in contested estates?" "What are [your firm]'s fees like?"
Comparison questions: "[Your firm] or [competitor], who's better for workers compensation?"

Evaluation questions are especially useful. They show you not just whether you appear, but what AI systems believe about your firm, and where that belief differs from the truth.

Then ask these questions across ChatGPT, Gemini, Google's AI Mode and Perplexity, and repeat them over several weeks. One snapshot tells you very little. Weekly is enough.

You can do this manually, or use FirmRanker's free AI check to see what the major AI systems say about your firm. Either way, the goal is a baseline: where you appear, how strongly, in which systems, and what they say.

Step 2: Get found in conventional search first

If your firm doesn't rank in Google for its priority practice areas and locations, fix that first. It's the strongest single association in the research, and it compounds: better Google visibility also means more traffic, enquiries and opportunities to be mentioned elsewhere.

For most firms that means:

Practice pages with real depth. One strong page per genuine service and location, answering the questions clients ask, with jurisdiction-specific information and your lawyers' expertise visible. See law firm SEO and local SEO for law firms.
Topic breadth and depth. Cover each practice area from multiple angles: the overview, the process, costs, timeframes, common mistakes, and the specific situations clients face.
A crawlable, fast website. Make sure your site isn't blocking the crawlers AI systems use, such as OpenAI's OAI-SearchBot, and that your important content is in the page's HTML, not hidden behind scripts or forms. Many older law firm sites fail here. See technical SEO for law firms.
Google Business Profile. For local practices, a complete and active profile is foundational. See Google Business Profile for law firms.

Step 3: Write content an AI system can actually use

AI systems favour content that answers clearly. For law firms, that means:

Answer first. Put the direct answer in the opening lines, not after three paragraphs about your firm's history.
Make each section stand alone. An AI system may lift one paragraph. It should still make sense and still identify the jurisdiction.
State the jurisdiction every time it matters. "In NSW, a family provision claim must usually be made within 12 months of death" is useful. "You have 12 months" is not, and may be wrong elsewhere. Legal answers without jurisdiction are a liability.
Show who wrote it. Named lawyer authors, with genuine credentials and accreditations, help both clients and machines judge credibility.
Keep it current. Laws change. Content that's out of date isn't just ineffective, it's risky. Build a review schedule, especially for pages about limitation periods, fees and procedural rules.
Write things search volume won't tell you to write. Buyer guides and "how to choose" content can earn AI visibility even where keyword tools show no search volume. For a firm, that might be "How to choose a family lawyer in Brisbane" or "What to ask before hiring a personal injury lawyer".

Practice Proof's AEO for law firms guide covers this in more detail. Law By Dan's AEO for law firms: evidence versus hype separates what's supported by evidence from what's being oversold.

Step 4: Make your firm's identity unambiguous

AI systems have to work out who you are. Make that easy:

Use the same firm name, address, phone number and description everywhere: website, Google Business Profile, directories, Law Society listings, LinkedIn.
Make your practice areas, locations and lawyers explicit. Every lawyer should have a proper profile page listing their genuine admissions, accreditations and areas of work.
Add structured data (schema markup) that truthfully describes your firm, its lawyers and services. Never mark up reviews, ratings or credentials that aren't real and visible on the page.
Decide what you want to be known for. A firm that says it does everything for everyone gives an AI system little reason to recommend it for anything in particular. See strategic positioning for law firms.

Step 5: Get chosen by building your reputation beyond your website

This is where most firms have the biggest gap, and it's where the research points most strongly.

Reviews, steadily. Ask every satisfied client, at the right moment, for a review on Google and on any legal directory that matters in your market. Volume and recency appear to matter more than chasing a perfect 5.0. Stay within your professional rules on testimonials. See how your law firm can get more Google reviews.
Mentions in local and legal publications. AI systems answering local questions draw on local sources. For a law firm, that means local news, community organisations, industry associations, and legal media, through genuine commentary on changes in the law, not paid "best lawyer" placements.
Referrer visibility. Accountants, financial planners, other firms and community services refer clients and talk about firms. Content, events and resources that keep you visible to them build the kind of web presence AI systems pick up.
Video and expert commentary. Lawyers explaining real issues clearly on YouTube, podcasts and in the media create widely discussed, attributable expertise. See video marketing for law firms.

This is brand building, and it's slow. That's exactly why it works: it's hard for a competitor to copy quickly. We explain why it matters beyond AI in why performance marketing fails law firms.

Step 6: Measure the right things

Track three things, at the right time horizons:

AI visibility: how often, how strongly and how accurately your firm appears for your priority questions, across several systems, monthly. Not a single "position".
Leading signals: branded search volume, direct enquiries, and the share of new clients who say they "looked you up" or "asked ChatGPT".
Commercial outcomes: qualified enquiries, consultations and opened matters in priority practice areas, and what they're worth.

Add "Did you use ChatGPT or another AI tool when looking for a lawyer?" to your intake questions now. It costs nothing and it's the most direct evidence most firms will ever have. See how to track law firm marketing to file opens, not clicks.

What not to do

Some tactics that appear to work in other industries are a bad idea for law firms.

Don't fake reviews, mentions or Reddit discussions. Astroturfing breaches the conduct rules and consumer law, and it's increasingly detectable. If your firm participates in online communities, do it openly as yourself: a clearly identified lawyer giving general information, never personalised advice in public.
Don't mass-produce AI-written content. Low-quality AI content has damaged many sites' search performance. For law firms the risk is worse, because inaccurate legal information can mislead clients. Use AI to help, never to publish unreviewed.
Don't buy "AI SEO" as a standalone product. AI search is one layer of how clients find lawyers. The firms that win are the ones with strong foundations, reputation and intake, not a separate AI campaign bolted on.
Don't panic over one bad result. AI answers vary. Act on patterns that repeat over weeks and across systems, not on one screenshot.
Don't ignore what happens after the recommendation. A client who finds you through ChatGPT still calls, and if no one answers, the recommendation was wasted. See law firm client intake.

How confident can we be about any of this?

Honestly: moderately.

The best large-scale research comes from other industries. It shows associations, not proven causes. AI systems change their behaviour frequently, and their answers vary from one request to the next. There's no published, rigorous study yet of what drives AI recommendations for Australian law firms specifically.

That's why we recommend this order. Every step that seems to improve AI visibility, ranking in Google, building genuine reputation, publishing clear and accurate answers, making your firm easy to understand, also improves conventional search, referral strength and conversion. None of it is wasted if AI search develops differently from what we expect.

That's also why FirmRanker exists: to measure how AI systems actually discover and recommend law firms over time, rather than relying on assumptions borrowed from software. Law By Dan explains what the research means. Practice Proof diagnoses where your firm stands and does the work.

Where to start

If you do three things this month:

Ask the questions. Put 20 real client questions to ChatGPT, Gemini and Google's AI Mode, record what they say about your firm and your competitors, and repeat in four weeks.
Add one intake question. Ask every new enquiry whether they used an AI tool when looking for a lawyer.
Fix your foundations. Check that your most valuable practice pages rank in Google, answer clients' questions clearly, name the jurisdiction, and show who wrote them.

If you'd like help working out where your firm stands, get a free audit and we'll show you where your enquiries come from, what AI systems say about your firm, and what to fix first.

Want to know what AI says about your firm?Our free audit shows where your enquiries come from, how visible you are in Google and AI search, and what to fix first.

Get a Free Audit →

Sources

Written by

Dan Toombs

Lawyer & Founder, Practice Proof

Dan Toombs is a lawyer, law reformer and founder of Practice Proof. Admitted in 1997, he built his first career inside the legal profession — practising law, leading legal organisations and driving law reform — before building the marketing, technology and AI practice he has now run exclusively for law firms for 20 years.

In that time, Dan and his team have helped law firms grow from single offices into multi-office and national practices, ASX-listed brands and acquisition targets. That experience gives Practice Proof an unusually deep understanding of both sides of law firm growth: what happens inside a legal practice, and what it takes commercially to make one grow.

Thomson Reuters featured his law-firm marketing advice as early as 2014, on positioning firms to win business online, websites and social media. He is also the founder of FirmRanker, which researches how AI systems discover and recommend law firms. Dan is a Winston Churchill Fellow and Australia Leadership Award recipient, with national recognition for his work in law and human rights.

If you want to know what 20 years of growing law firms could bring to yours, talk to Dan and the Practice Proof team.

Book a call with Dan →