I'm seeing keyword research for law firms shift from "what does Google want?" to "what does an anxious client ask out loud into a chatbot?" What's working best for my clients now starts with intake, not Ahrefs or SEMrush. We mine phone calls, website chats and email enquiries for exact phrases people use before they know the legal term: "Can I get out of my AVO?", "Do I lose my licence on the spot?", "What if I can't afford a lawyer?" That becomes the base keyword set. Then I map those into question flows that match how AI tools answer. Someone might go from "drink driving limit NSW" to "will I get a criminal record" to "do I need a lawyer for first offence" to "best drink driving lawyer near Parramatta Court". We build one deep content asset or cluster to cover that whole chain, instead of 10 thin blogs chasing slight keyword variations. To help with AI visibility, I lean hard on signals that large language models can parse: clear headings that mirror questions, detailed FAQs, and explicit jurisdiction cues (state law names, sections of Acts, local court names, suburbs). That's changed GEO tactics too: we now bake micro-location into scenarios (e.g. "what happens at Downing Centre Local Court") rather than just stuffing suburbs into a footer. I'm also seeing value in publishing "opinion plus process" content: explaining how a firm assesses a case, not just the law. AI systems seem more likely to summarise and cite firms that offer unique, procedural detail. What's not working now: generic "what is X offence in NSW" pages written around a single head term; city pages spun with only the suburb name swapped; and any content that dodges cost, timelines, or likely outcomes. Those all blend into the training data noise and rarely surface in AI answers or local packs.
As the director overseeing SEO, content strategy, and marketing at Express Legal Funding, I bring over a decade of experience in keyword research for lawyers and law firms. The SEO challenges we face at our company are also shared by personal injury firms, which in turn affect our content strategy. We target the same audiences, address high-intent queries, and respond to the urgent needs of potential clients. Meanwhile, search results remain saturated with legal service providers. Over the past year plus, however, I have observed a notable shift: an increasing number of leads are utilizing AI tools and generative search features such as Google AI Mode to seek information and select a provider. Consequently, keyword research now extends beyond ranking for specific phrases. It requires ensuring our content is perceived as the most trustworthy, comprehensive, and accessible source when AI systems generate responses. From a practical standpoint, instead of targeting keywords, we focus a significant portion of our original content creation efforts on what I like to refer to as primary FAQs. It helps us build topical authority. Additionally, since Google prioritizes ranking multiple sources and angles within its AI overview, it enables adjacent content, like ours, to still rank for relevant queries. In part, this largely means that keyword search volume is less of a data point we consider. We aim to post content that fully explains the process and the "why," not just the definition. We also take the time to structure pages so both AI and humans can handle and easily ascertain the topic we are covering. From what we've seen, that requires clear sections, straightforward language, and FAQ-style headers that answer specific questions.
I work with solicitors and law firms, and the biggest change I'm seeing is this: keyword research is no longer just "find a term and write a page". AI-driven search is rewarding clarity, topic coverage, and trust signals. What's working for us right now is combining classic intent research with "retrieval-ready" content. We still start with keywords, but we expand into the full set of questions behind them (definitions, scenarios, timeframes, costs, next steps). This tends to perform better in AI surfaces because the content reads like a complete answer, not a thin page built around a phrase. We're also leaning harder into entities and structure. That means consistent naming of practice areas, locations served, and legal concepts across the site, plus proper internal linking between hub pages and supporting guides so the site shows clear subject depth. One tactic that's becoming non-negotiable is JSON-LD schema. It helps search engines and AI systems interpret the site with less guesswork. For law firms, we typically implement (where relevant): LegalService/Organisation, LocalBusiness, Person (author), Article, FAQPage, BreadcrumbList, and Review/AggregateRating only where it's genuine and compliant. This improves machine understanding of who the firm is, what it offers, where it operates, and who is responsible for the content. Local visibility is also shifting. Strong Google Business Profile signals, consistent NAP citations, and location-specific service pages (written for humans, not templates) are performing better than generic "we serve X" pages. Where we've seen things fall flat is with mass-produced pages that all read the same. They're usually written to tick an SEO box, not to explain anything properly. Those pages don't get pulled into AI answers very often, and in the legal space they can cause headaches if they oversimplify issues or drift into implied outcomes. We're deliberately more cautious with content. It has to be correct, it has to explain the edges of an issue, and it has to stand up if a regulator or a client reads it closely. Keyword research still matters, but it's not the end goal anymore. For us, it's just the starting point before we build something that's properly structured, clearly written, and easy for both people and machines to understand.
Whether a potential client finds you through a search engine or an AI prompt, they are looking for trust. Both platforms value content written by qualified experts. A detailed author biography for your content and clear credentials for who has reviewed the content is essential. People speak to AI differently than they type into a search bar. They ask full, detailed questions about their specific situation. We advise our legal clients to focus on these long phrases. Do not limit your research to 'medical negligence solicitor'. Look for the specific questions clients ask in your initial meetings, such as 'what is the time limit for a medical claim in the UK?'. Creating content that directly answers these specific questions is the best way to become the source the AI chooses, and you can often add these as structured FAQs to existing pages. Getting your firm mentioned on respected third party websites has always been important for SEO because it builds authority. It also helps AI as LLMs scan data for consensus to decide who to trust. If independent industry sites reference your brand, it acts as a signal that both Google and AI engines can use. With all the noise about AI it's important to remember that the vast majority of people still look for a lawyer using a standard Google search. The great news is that SEO fundamentals work for search engines and AI tools.
I run digital marketing campaigns for a couple large accident attorneys, and we've been having success taking a more personal approach on social channels. Instead of keyword research taking place strictly on Google or using search tools, we've expanded it to more conversational spaces like Reddit, Quora, TikTok, and Instagram to see how those conversions are taking place, what people are tagging, and how they are asking questions. From there, we can quickly create highly targeted content based on issues people are talking about. That content is not only posted to social media, but also linked back to FAQs or Media sections on the website with full transcriptions, citations, and extra resources, providing more detailed information that help rankings in LLM results.
At EMILY Revolutionary Marketing Group(r), we help law firms and attorneys grow their visibility by combining traditional SEO best practices with emerging AI search optimization (AEO / GEO) strategies. With more people turning to generative AI to find legal answers and providers, the way law firms approach keyword research and content creation must evolve. What's Working Right Now: 1. Shifting from "exact-match" keywords to "natural language queries." People don't search for "personal injury attorney Columbia SC" anymore—they ask, "What should I do if I was hit by a commercial truck in South Carolina?" We build out content that answers these questions directly, aligning more closely with how AI engines generate responses. 2. Creating structured, scannable content with schema markup. We use LegalService schema, FAQ schema, and LocalBusiness JSON-LD to support both traditional Google visibility and AI result integration. This helps AI tools understand the practice's core services, areas served, and niche focus. 3. Supporting content clusters instead of one-off pages. Rather than creating dozens of isolated blog posts, we group them into intent-based topic clusters—for example, a "Workers' Comp" hub that includes FAQs, case timelines, state-specific laws, and attorney Q&A. This improves authority, dwell time, and AEO alignment. 4. Prioritizing local and voice search factors. Since many legal services are hyper-local, we integrate geo-anchored keywords, optimize Google Business Profiles, and use tools like Microsoft Clarity to watch user behavior. This ensures that pages not only rank but convert. What's Not Working Anymore - Keyword stuffing or legacy SEO tactics. Pages written for bots, not people, get ignored by both AI and users. - Thin "services" pages that lack depth, examples, or answers to real user intent. - Generic blog posts that don't differentiate a firm's unique positioning or local experience. Bottom Line Keyword research for lawyers now requires thinking beyond search volume. We look at conversational intent, AI training patterns, and schema structure. The goal isn't just to rank—it's to be chosen by the AI tool that delivers the answer. Let EMILY(r) create a strategy that drives traffic and builds trust. Visit us at www.ermarketinggroup.com. EMILY(r) — Powered by Data, Driven by Results.
We've worked with a UK-based crypto law firm and a crypto accountant, and one of the biggest shifts we've seen in 2026 is that traditional keyword research needs to evolve. To stay visible in AI-generated answers, we're now structuring legal content around real user intent, regulatory context, and specificity. For instance, instead of chasing terms like "crypto solicitor London," we create targeted content hubs around questions such as "is staking taxable in the UK" or "how to register a crypto asset business with the FCA." These detailed, localised queries are far more likely to surface in AI-generated answers. What no longer works is publishing broad, generic articles like "how crypto is regulated." AI engines ignore them in favour of expert-written, structured content. What's working now is author-led insights, strong schema markup, and language that reflects how users actually phrase legal concerns in the UK. We track AI Overviews and use those phrasings directly in headings and FAQs. GEO in 2026 is about trust signals and structured answers, not just ranking for the obvious keywords.
For law firms, keyword research is shifting from volume chasing to intent proof. What's working is mapping queries to specific legal outcomes and jurisdictions, then reinforcing them with first hand expertise signals like attorney authored insights, case context, and plain language explanations. AI engines seem to reward clarity and credibility over clever phrasing. What isn't working is generic blog content rewritten at scale or pages stuffed with loosely related terms. AI surfaces brands it trusts, and that's pushing legal SEO toward fewer pages, stronger authority, and tighter topical focus rather than broader reach.
1) Instead of standard keyword density, it's now about machine-readable authority. For law firms, that means deep Schema markup (especially LegalService and Attorney types) so that AI knows to verify the jurisdiction and expertise being conveyed. We're still breaking content into bites through Q&A but trying to match natural prompts: statista + open ai data integration shows 50%+ of legal search queries as voice/AI by 2026, so if you're not answering a specific, spoken question: you don't exist to these models. 2) Biggest paradigm change is from "ranking" to "recommendation." Old SEO was about being one of 10 blue links, GEO is about being the one source that AI calls out in its summary. So our keyword research went from hunting for high-vol head terms like "personal injury lawyer," to "intent clusters." Let's focus on the lifecycle of the legal problem. Create a content hub, tell them the "why" of the case and the "how" of the case, and you give AI the context it needs to choose your firm as a trusted entity. 3) What's working now is attorney-in-the-loop content--eyeball deep, specific articles with citations, local case references, etc. AI is getting much better at spotting trash "slop." What's not working is the high volume low quality blog that many firms employed for years: if it's thin, looks like an old LLM which has no expert oversight, it's ignored by the engine trying to reach. You can't out-AI the AI... you'll lose, but you can give it the human juice it currently lacks that you possess. And if that's all overwhelming, it's all still about trust! It's another way for a client to find a trusted advocate... and if you open the door with visible expertise, the tech does the rest of the work to introduce you.
Personal Injury Attorney and Founder at Loyd J Bourgeois Injury & Accident Lawyer
Answered 3 months ago
Our marketing team is focusing on local community involvement and digital PR efforts surrounding that involvement to build up our local authority in our service areas.
Traditional keyword research for law firms focused on "personal injury lawyer Chicago" variations. That game has fundamentally changed. What's working now is answering the questions AI engines pull from. When someone asks ChatGPT or Perplexity about handling a wrongful termination case, the AI synthesizes content that directly addresses process, timelines, and what to expect. Our clients ranking in AI results are the ones creating exhaustive FAQ content that mirrors how real people actually ask legal questions conversational, worried, specific. The shift: we've moved from targeting "car accident attorney" to targeting "what happens if the other driver's insurance won't pay." Same practice area, completely different content strategy. What's not working is thin location pages. Law firms used to create dozens of city-specific pages with swapped location names. AI engines see through that instantly. They reward depth over geographic keyword stuffing. The firms winning in generative search are publishing genuine thought leadership on niche legal scenarios. One employment law client now ranks in AI overviews specifically because they wrote detailed content about non-compete enforceability by state something AI engines constantly reference.Claude is AI and can make mistakes. Please double-check responses.