Someone with a back injury opens ChatGPT and types "best personal injury lawyer near me." The model names three firms.
Nobody clicked a search result. No ad was served. No ranking position decided it. And if your firm was not among those three, you will never know the conversation happened.
That is the problem answer engine optimization exists to solve, and it works differently enough from SEO that treating it as an extension of your existing program will not get you there.
Why links stop being the mechanism
Traditional SEO is built on links. Google counts them, weighs them, and ranks accordingly.
Answer engines do not work that way. As iLawyer Marketing puts it on its own service page, answer engines do not rely on links the way Google does. What matters instead is getting into the model's training dataset, producing content the model can digest cleanly, and marking it up so the model can parse what your firm actually is.
That is a different discipline with different levers. A firm with a strong backlink profile and no structured data can rank well on Google and be invisible to ChatGPT.
The platforms that matter
Optimization targets are no longer one search engine.
ChatGPT, Google Gemini, Google AI Mode, Google AI Overviews, Perplexity, Claude, Microsoft Copilot and Grok each retrieve and cite differently. Some run live search. Some lean on training data. Most blend both.
Practically, this means you cannot optimize for "AI" as a single target. Perplexity rewards retrievable, well-structured pages. ChatGPT rewards brand recognition built across years of mentions. The work overlaps but it is not identical.
The data on how people actually find lawyers
This is the part firms tend to get wrong in both directions, either dismissing AI entirely or assuming it has replaced search.
Neither is true. iLawyer Marketing's own consumer research puts it at over 28 percent of people using ChatGPT when researching attorneys, while 94 percent of ChatGPT users also use Google for the same research.
More importantly, over 70 percent of consumers use multiple sources when deciding which firm to hire, including Google, ChatGPT, Facebook, YouTube and Reddit.
The strategic read is omnichannel, not replacement. A prospective client asks ChatGPT for names, checks those names on Google, reads reviews, watches a video, and possibly asks in a subreddit. Being strong in one channel and absent from the others loses you the ones who check.
What the work actually involves
Vague talk about "AI visibility" is common. The specific components are not.
iLawyer Marketing publishes a seven-part process, and it is worth walking through because each element addresses a different retrieval mechanism.
AI content mapping, identifying the legal questions people actually ask across multiple answer engines rather than the keywords they type into Google. The phrasing differs, and questions run longer and more conversational.
Structured legal data optimization, using schema markup and metadata so models can parse and cite your firm without guessing.
Citation building for AI sources, making sure your firm appears in the reputable structured sources models draw on. This is the closest thing to link building in AEO, and it operates on trust rather than volume.
Content for AI indexing, building Q&A-style content that matches how answers are constructed rather than how blog posts traditionally read.
Entity linking, connecting the firm to verified entities so models can confirm authority independently.
LLM relevance testing, actually querying the assistants to see whether your firm surfaces for target questions. This is the step most programs skip, and it is the only real measurement available.
Multi-platform monitoring, tracking appearance across ChatGPT, Gemini, Perplexity, AI Overviews and AI Mode over time.
Entity confirmation, and why documents matter more than you think
The fifth item deserves expanding, because it is the least understood.
An answer engine recommending your firm needs to be confident the firm exists as a distinct, verifiable entity. That means your name, address, bar admissions, practice areas and jurisdictions holding consistent across every source the model can read.
Inconsistency fragments the entity. A firm listed three ways across its site, its Google Business Profile and its directory entries is three weak signals rather than one strong one.
The same discipline that determines who can sign legal documents applies here: authority has to be verifiable, and a claim nobody can confirm carries no weight. Models are doing something closer to reading a contract than reading marketing copy, and ambiguity costs you the same way.
What results look like
iLawyer Marketing publishes a case study reporting a 139 percent increase in AI referral traffic for a personal injury firm in a tier-one market, alongside doubled visibility across AI platforms.
On timelines, the firm's own guidance is that some results appear in weeks, while competitive legal queries can take several months or longer in the most contested markets.
That range is worth taking seriously. Entity authority and training-data presence build slowly. Technical fixes like schema markup land faster.
The first-mover argument, honestly assessed
The strongest case for acting now is that most law firms have not.
AEO is early enough that a firm doing it properly can establish presence before the category saturates, in the way early SEO adopters did. iLawyer Marketing makes this argument directly, framing it as capturing market share before competitors gain a lead.
The counter-argument worth acknowledging: answer engines change their retrieval methods frequently, and tactics that work now may not persist. Anyone promising durable positions is overselling.
What does persist is entity clarity, structured data and genuine third-party citation. Those are useful regardless of how the models evolve, which makes them the safer place to invest.
Choosing a provider
Four questions separate a serious AEO program from a repackaged SEO retainer.
Do they test whether AI actually recommends you? Not traffic estimates. Actual queries run against actual assistants, with results recorded over time. If the answer is no, they cannot tell you whether the work is working.
Which platforms do they cover? A program targeting only ChatGPT ignores where much of the volume sits, particularly Google AI Overviews.
How do they handle citation building? This should mean structured, reputable sources rather than volume link placement.
Do they work in your vertical? Legal has specific constraints. Bar advertising rules, jurisdictional accuracy and the sensitivity of the queries involved all shape what content is appropriate.
On that last point, iLawyer Marketing works exclusively with law firms, has done so for over 20 years, and states it has been studying large language models since 2019, well before most agencies were aware of them. For a practice area where a generic marketing agency's content can create compliance problems, exclusivity is a genuine consideration rather than a positioning line.
Where to start if you are doing this yourself
Run the queries. Ask ChatGPT, Perplexity, Gemini and Copilot the twenty questions your clients ask, using your city and practice area. Write down which firms get named. That baseline costs an hour.
Audit your entity data. Name, address, phone, bar admissions and practice areas, consistent everywhere.
Add proper schema. LegalService, Attorney, FAQPage and LocalBusiness markup, correctly implemented.
Restructure your highest-value pages. Question-shaped headings, answers in the first two sentences, standalone paragraphs that survive being lifted out of context.
Then repeat the queries monthly and record what changes.
The honest summary
Answer engine optimization is not a replacement for SEO and it is not a rebrand of it. It is a parallel discipline addressing a channel where over a quarter of prospective clients now begin their research.
The firms getting recommended are the ones models can identify with confidence, verify against independent sources and extract clean answers from. That combination takes months rather than weeks, and it compounds.
Start with the baseline query test. Knowing whether you currently appear is the cheapest and most useful hour you will spend on this.
The opinions on this page are for general information purposes only and do not constitute legal advice on which you should rely.
.png)
