Five sessions across two engines and two states have now returned the same Dayton law firm’s blog post when we asked how a person can tell whether a firing was illegal. By most published checklists that page should not be winning anything.

At Lexicon Legal Content we spent August testing which employment law firms answer engines actually cite, and the pattern in the pages that keep surfacing is not the one the industry keeps recommending.

The Firms Winning Employment Answers Are Ones You Have Not Heard Of

Duwel Law, a Dayton practice, has been cited five times across Google AI Mode and Perplexity, from sessions run in Colorado and Illinois, on the question of how to recognize wrongful termination. Nisar Law in New York came back on hostile work environment questions in run after run, and again on damages. Bell Law Group and Amber Boyd Law both reappeared weeks apart on different engines.

Not one of them is a national brand. There is no Am Law firm in the set, and the only large firm we saw all month showed up on a directory-flavored question rather than a client one.

What the four have in common is narrow and repeatable. Each is a practice focused on employment work rather than a general firm with an employment page. Each publishes explanatory blog content on the questions clients ask, not service copy describing what the firm offers. And each states its geography plainly enough that an engine never has to guess where the advice applies.

That is not an accident of our sample. InterCore Research, which ran 540 high-intent legal queries across 30 US metros in July 2026, found firm citations split across 254 different websites, with the most-cited single firm appearing 36 times out of 18,900 citations. Nobody owns this category, which means nobody has to be displaced.

What a Cited Employment Page Actually Looks Like

We went back and read the post engines keep quoting. It is a blog post from March 2026, roughly 1,200 words, and it breaks most of the rules our industry publishes.

There is no named attorney byline, only the firm name. It carries no external citations, no FAQ section, and it opens on how losing a job feels rather than answering the question in its own title.

What it has is a page a machine can take apart. Short labelled sections, bolded key terms, three scannable lists, a five-step numbered sequence at the end, and its city and five surrounding counties stated without ambiguity.

Duwel Law hits one of the five signals on our own list of what engines check, and gets cited anyway. That is not the list failing. It is the list mattering less on a question where the reader wants an explanation than on one where they are choosing who to call.

Extractability wins the informational questions. Author and credibility signals carry the commercial ones, where a person is deciding who to trust with a claim.

They Win the Questions Directories Cannot Answer

InterCore found that 77.8% of answers cited a directory first, asking variations of “top employment law firms in Chicago.” Across the sixteen answers we collected asking client-phrased questions, a directory led exactly none of them.

Both findings are correct, and the gap between them is where these small firms live. Justia and Super Lawyers were built to answer “who should I hire in my city” and they win it decisively. They have almost nothing to say about whether your boss can fire you for no reason, what counts as a hostile work environment, or what your case might be worth.

Duwel Law is not out-competing Justia. It is answering a question Justia never tried to answer. That distinction is worth more than any single tactic in answer engine optimization.

The practical version for a firm owner: stop writing pages that describe your employment practice and start writing pages that answer the questions a person has before they know they need a lawyer. The firms getting cited did not out-spend anyone. They wrote the explanation first.

Being Named Is Not the Same as Being Cited

There is a version of this that looks like a win and is not. The firm appears in the answer, but the firm’s own website is nowhere in the sources.

InterCore found 34.2% of firms named in an answer had no page of their own among the cited sources. The engine knew those firms only from directory listings, which means the description a prospective client reads was written by Avvo, and there is no link back to anything the firm owns.

The firms in the first section are not in that position. They are cited because a page they wrote is doing the work.

Why You Probably Cannot See Any of This From Your Own Account

Here is the part that makes this hard to verify for yourself. We asked Perplexity six employment law questions from a Chicago connection, then asked again after turning off the account history that had been saving our searches.

On “do I need a lawyer for a wrongful termination case,” both runs returned eight law firms, and exactly one firm appeared in both answers. A third run, with conditions held steady, came back identical to the second on two of four questions. These answers are reproducible, which means the difference between the first two runs was us.

A firm checking its visibility while signed in to the account it uses for competitor research is reading its own history back to itself. Location matters less than most people assume. From that Chicago connection the same question returned Ohio firms once and California firms twice, and never an Illinois practice.

None of this is stable, and any firm treating a single good result as an achievement should know that. Duwel Law sat at position 2 in one of our runs and position 13 in another, on the same question, thirty minutes apart. Consistent citation across sessions is the thing worth chasing. A single appearance is closer to weather.

Where to Put the Next Page

In every run we did, usa.gov or the EEOC held the first citation slot on definitional and procedural questions. That ground is not available, and no amount of content investment will take it.

What sits unclaimed is everything the agency does not publish. What a charge actually costs, how long an EEOC process runs in practice, whether a particular set of facts is worth filing on, what happens to your job while a charge is pending, and what your state adds on top of federal law.

Build those as guides rather than short posts and connect them, which is what topic cluster strategy is for. Otterly.ai, analyzing more than a million cited URLs in 2026, found guide pages averaged 2.7 citations against 2.0 for blog posts.

What Not to Spend On

One in five employment law sites carries no structured data, according to a March 2026 audit of 299 employment law websites by AISO Studio, a company that sells AI visibility auditing. Duwel Law’s cited page has no FAQ schema either, and it is cited anyway.

Google’s own guidance states that structured data “isn’t required for generative AI search, and there’s no special schema.org markup you need to add.” Schema still earns rich results in classic search, which is a separate and legitimate argument.

Nor does a top ten ranking carry a citation the way it once did. Ahrefs put AI Overview citations coming from the top ten at 37.9% in March 2026, down from 76% the previous July, though the same authors note they improved their citation parsing in between, so part of that drop is method rather than behavior.

How to Check Whether You Are in the Set

Sign out first, or use a private window. Ask the six questions your clients actually ask, in their words rather than a directory’s, and read the source list under each answer.

Do it twice on different days and write down the date and whether you were signed in. Six questions takes twenty minutes, and one reading on its own tells you very little.

Frequently Asked Questions

Which employment law firms are getting cited by AI search?

Small and mid-size practices focused on employment work. In our testing Duwel Law, Nisar Law, Bell Law Group and Amber Boyd Law recurred across engines and sessions, while national brands appeared only on directory-style questions.

Do I need a big budget to get cited by answer engines?

No. Citations split across 254 firm websites in InterCore’s study, with the most-cited firm appearing 36 times out of 18,900. The category has no incumbent, so the barrier is content structure rather than spend.

Why does being signed in change which law firms an engine cites?

Engines personalize to account activity. A firm that has spent months researching competitors from one account will see that history reflected back, which makes its own visibility check the least reliable one available to it.

Do legal directories dominate AI answers or not?

It depends entirely on the question asked. Directories led 77.8% of answers to “top firms in city” phrasing in InterCore’s testing, and none of the sixteen client-phrased questions that we ran ourselves.

The Harder Half

Twenty minutes gets you the reading. Knowing which gaps are worth a page and which belong to a federal agency you will not outrank is the harder half, and it is where we start. Every piece we deliver is reviewed against the statute it cites before it reaches you, which matters more in employment law than in most practice areas, because the wrong sentence about a filing deadline is not a marketing problem.

Call 877-486-8123 and tell us which questions your clients ask most often, or see the legal content writing itself first. You can also contact us with what you are up against.

---

David Arato, JD is the founder of Lexicon Legal Content, an attorney-owned legal content marketing agency serving law firms since 2012. He has spent the past year rebuilding how Lexicon measures AI visibility, after finding that most published methods produce numbers a firm cannot act on. David is a frequent contributor to Attorney at Law Magazine and Attorney at Work and a recurring guest on legal marketing podcasts.