Law Firm Hierarchy: 521 of 1,008 Firms Publish No Rank at All

What we measured

Search for law firm hierarchy and you get a ladder. Equity partner, non-equity partner, counsel, senior associate, associate, paralegal, with a chart. It is accurate about the firms it describes, which are large commercial firms, and the pages describing it are written for people who want a job at one.

Nobody on that search result has looked at a law firm's website. The question a client actually has is narrower and nobody answers it: when I open this firm's team page, will it tell me who these people are?

So we counted. For every firm in the directory we read the pages where it names its own people, and recorded two things about each name: the title the firm printed beside it, in the firm's own wording, and where on the site that title came from. 1,008 of the 1,388 published firms name somebody. 380 name nobody at all.

Those two fields are separate on purpose, and the second one is load bearing. An earlier version of our crawler wrote the word "Attorney" into the role whenever a firm printed nothing, which meant the record asserted a credential the firm had never claimed. The field recording that the firm had said nothing is the only thing that tells the two apart, so every count on this page tests it rather than the title string.

One name goes in one of three places, and the split is the same one each firm profile publishes.

3,964 of the 10,010 names carry either a lawyer's title or a registration we matched in a state register. 578 carry a title that is not a lawyer's: paralegal, case manager, intake, legal assistant. The largest group is the middle one, and it is larger than the other two together.

That matters more than it sounds. 421 of the 1,008 firms with a roster name people and give us nothing that identifies any of them as a lawyer, which means the page a client would use to check who is handling the case does not do that job.

What a team page says

Split by firm rather than by person, the shape is sharper. The question is what share of its own people a firm bothers to rank.

521 firms print no rank at all and 137 print one beside every name. Between them sit 350 firms that rank some of their people and not others.

Here is the problem with reading that straight. A title our reader does not recognise reads to us as no title, and the pattern list it uses is short by design. So part of that first bar is our crawler rather than the firm.

The control is to throw away every firm we never got a title from. 487 firms gave us at least one, which proves our reader could see titles on those sites. Inside that group, 3,785 of the 6,378 named people carry one and 2,593 do not. Those 2,593 names sit on pages where the firm demonstrably knew how to print a title and did not print one beside them.

The habit also moves by market.

The ladder, and who publishes one

Now the ladder itself. Of the 487 firms that print any title at all, here is how many print each rung at least once.

The most common thing a law firm prints beside a name is the bare word "Attorney", which tells a reader that the person is a lawyer and nothing about where they sit. The ladder that every explainer draws is on 51 websites in this directory. That is 51 of the 487 firms that print any title, and 51 of the 1,008 that publish a roster at all.

Of counsel appears on 76 of these sites, beside 188 people. Shareholder, which is what a professional corporation calls a partner, appears on 8.

Two firms show the two shapes as clearly as anything on this page, and both were picked by the size of their roster rather than by hand.

McCready Law names 76 people and prints a job title beside none of them. Every one of those names is somebody the firm chose to publish, and a reader has no way to tell the founding partner from the intake coordinator.

Singleton Schreiber names 127 and prints a title beside 118 of them, partners and associates among them. It is the same kind of page, the same size of firm, and it answers the question.

Those firms sit in Chicago and San Diego, and the ranked lists for those markets are built from the same records.

Whether a rank predicts anything

A page that publishes ranks looks like a more careful page. The question is whether it goes with anything a client would notice, and there is one measure here that does not come from the firm's own website: the rating its clients left on Google.

Median certification score runs 57 against 51. Mean Google rating runs 4.85 against 4.82, a difference of 0.03 of a star across 487 and 521 firms. Our scale cares a great deal and the clients do not appear to care at all.

Before trusting that, hold the headcount still. A firm naming thirty people has more chances to print a title than a firm naming two, and large firms are not small firms in any other respect either.

The two ratings stay within a tenth of a star in 4 of the 4 headcount bands, and they are closest among the largest firms. The falsification survives: printing a rank beside a name does not go with a better rated firm.

The same test by market goes the other way for our own score, which is the honest half of it. In the 12 markets holding at least eight firms on each side, the firms printing ranks carry the higher median score in 10. So the score gap is real and it is not a trick of which cities the two groups sit in.

What else moves with it is reach. Median monthly organic search traffic runs 806 visits against 365, and median Google review count 305 against 186. That is the same finding we reached from the other direction when we sorted the whole directory by search visibility: a firm that publishes more of everything is a firm with a marketing operation, and visibility brings more reviews rather than better ones.

The firms with no roster at all make the point hardest. Their median certification score is 40, the lowest of the three groups by a distance, and their mean Google rating is 4.88, the highest of the three. A firm can name nobody on its website and still be the firm its clients rate best.

What our own score does with it

Two of our sub-factors read a roster. A2 is experience, worth 10 points, computed as the mean years since admission across the people a firm presents as lawyers. A6 is roster verifiability, worth 9, the share of that roster we matched to a state attorney register. Together that is 19 of the hundred, and the full method is on the methodology page.

Neither can be scored without a register to read, and a register is the state's to publish. A2 is scored for 305 of the 1,388 firms here and A6 for 328. 908 firms sit in a state that publishes no attorney register we may query, so those 19 points leave the scale entirely rather than being charged to the firm. In 11 of the 13 states this directory covers, neither sub-factor can be read for anybody, which we have written up state by state in what you can and cannot check about a lawyer.

The other branch is the firm's. 152 firms have no roster for A2 to average and 381 have none for A6 to verify. Those points leave the scale too, because a firm must not lose points to the reach of our own crawl. What a firm is actually asked for is a published roster, and that is gate G5 rather than a sub-factor.

Now the part that is ours to answer for. Our reader is not as good as this page has implied. 328 of the 3,785 titles we did read are ones our own classifier cannot act on, so those people fall through to "named without a role" exactly as if the firm had printed nothing. They sit across 156 firms.

Some of that pile is our own name splitter leaving a suffix behind where a person is a third or a fourth of their name, and some of it is a practice area lifted off a heading. Some of it is a real title in the wrong language: 10 of those people are titled with the Spanish word for lawyer, which our list of lawyer's titles does not contain, on a directory where we have counted how many of these firms publish in Spanish. None of it is the firm's fault, and all of it is counted above as the firm's silence.

The asymmetry worth naming is that our scale pays a firm nothing for printing a rank. A6 pays for the share of a roster we can match to a register and for bar numbers published on bios; neither reads the word "Partner". Yet the two groups sit 6 median points apart, which means the gap is coming from everything else a thorough firm publishes rather than from the ranks themselves. We found the same shape when we counted which pages a law firm website carries.

What this does not measure

Four limits, in the order they would bite.

An absence here is partly our reading. The pattern list our crawler uses to recognise a title beside a name is deliberately short, because a word not on it is a word we will not assert is somebody's job. That is the right trade for a record and the wrong one for a denominator, which is why the central figure is reported twice and the control cohort of 487 firms carries the claim.

Nothing here says anybody is not a lawyer. 5,468 people are named by their firm without a title we could read. That is all the record says, it is all their profiles say, and it is all this page says. A person who is not matched to a register is not a person the register rejected.

The roster read carries no date. Every other measurement family in this directory records when it ran, and this one does not: no field on an attorney record says when the team page was last fetched. So this page prints no measurement date rather than borrowing one from a different reading, and a page a firm rewrote last week is a page we may not have.

This is a directory of consumer firms, not of the firms the ladder describes. 1,113 of the 1,388 firms here lead with personal injury. The explainers that rank for this question describe large commercial practices, and the honest reading of everything above is that the two are different worlds rather than that one of them is doing it wrong. If you are hiring a lawyer for an injury, a deportation or a divorce, the ladder in those articles is not the firm you are about to meet.

Every firm can see its own roster, its own sub-scores and the evidence behind each one on its profile, and a firm that publishes a roster we never found can tell us where it is. The ranked lists these figures sit behind are at the personal injury hub and the city pages under it.

Titles and names are read from each firm's own attorney, team and bio pages. Bar numbers and admission years come from the New York and California attorney registers. Review counts and ratings come from Google Business Profile listings verified to the firm. Search traffic comes from Ahrefs. A firm whose page we could not read keeps the sub-factor pending rather than scoring zero.