Every PERM application names the law firm that prepared it, and the Department of Labor publishes that name when it publishes the decision. Add the names up and you get a ranking of who files the most. The ranking is real. What people read into it usually is not.
What DOL actually records about a firm
Form ETA-9089 has a block for the attorney or agent representing the employer. That is the only thing being counted here. It is not a measure of who is a good lawyer, who took the case first, or who the worker talked to.
It also arrives late. DOL names the firm in its quarterly disclosure files, which only carry cases that have already been decided. So a firm's totals describe work that is finished, and nothing waiting in the queue can be attributed to a firm at all.
What high volume actually tells you
It tells you a practice does this work at scale and does it repeatedly. That is worth something. A firm filing tens of thousands of applications has met most of what DOL does, including the parts that only show up a few hundred times a year.
It also tells you who the clients are. Volume at the top of this list is concentrated in a handful of large employers. The busiest firm's six biggest clients account for about a quarter of everything it files, and about a third of its cases are for software developers. That is a description of a client book, not of skill.
Do approval rates separate one firm from another?
Barely, at the top. Across firms with enough decided cases to carry a meaningful rate, approval clusters above 99%, and hundreds of firms have no denials at all. The field as a whole approves 96.8% of decided cases, and the denial rate is 3.2%.
The spread only opens up at the other end, and when it does, look at the caseload before you look at the firm.
| Firm | Filings | Denied | Approval | Median days |
|---|---|---|---|---|
| Fragomen, Del Rey, Bernsen & Loewy | 48,322 | 351 | 99.2% | 459 |
| Berry Appleman & Leiden | 19,139 | 188 | 99.0% | 450 |
| American International & Immigration Law | 2,972 | 260 | 91.0% | 478 |
| Dentons Cohen & Grigsby | 1,386 | 178 | 85.9% | 493 |
| Worldwide HealthStaff Solutions | 1,763 | 313 | 81.9% | 406 |
The firm at the bottom of that table files nursing assistant positions almost exclusively. That occupation is 68% of its work and is approved 73.9% of the time within its own filings. The firm's overall rate is mostly that one number, wearing a firm's name.
This is the pattern everywhere in the data. Case attributes move outcomes far harder than firm identity does. A position that is not full time is denied more than half the time, roughly seventeen times the field rate. The denial rate page publishes those factors separately and deliberately refuses to fold them into a single score, because they are not independent of each other and one number would read as more precision than the files support.
Why median days says almost nothing about a firm
DOL works one national queue, oldest first, regardless of who filed. So a firm's median time to decision is mostly a statement about when its cases were filed.
The busiest firm's median is 459 days against a field median of 469. Ten days, over 46,526 decided cases. Where a firm does land well clear of the field, the usual explanation is a run of filings that happened to sit in a faster stretch of the queue, not anything the firm did after filing.
One practice, several spellings
DOL prints whatever was typed on the form. The busiest firm in the country appears in the files under more than a dozen spellings, including a transposed name, a missing space, and two office names.
Close variants get pooled, so a punctuation difference or a single mistyped letter counts once. Spellings that differ by more than that are left alone, because merging them would mean guessing. For that one firm, 152 filings sit on other pages rather than on its own. Ranks are therefore approximate, and the gap matters most in the middle of the list where a few hundred cases move a firm several places. Each firm page lists the near-miss spellings so you can see what was and was not merged. The methodology has the rule.
You probably did not choose the firm
This is the part that surprises applicants. The employer is the client. The employer pays, the employer picks counsel, and the application is the employer's, not the worker's. Plenty of people go through the whole process without ever speaking to the attorney whose name is on the form.
Which means the ranking is not a shopping list for most readers. It is more useful in reverse: you can see the kind of work a firm does, which employers it files for, and roughly how its cases have gone.
You also do not need the firm to follow your own case. The case status lookup takes your DOL number and reads DOL's own record. No account, and it works whether or not anyone has told you anything.
What this cannot tell you
- Whether a firm is good. These are filings DOL approved or denied. Advice, cost, responsiveness, and how a firm handles an audit are not in the files.
- How your case will go. A firm's approval rate is a property of the cases it has taken. Yours is one case, and the things DOL denies for are attributes of the job and the recruitment.
- How long your case will take. A firm's median is history. Your own wait comes from your filing month against where DOL is now, and even that estimate can miss; the estimate scorecard shows by how much.
- Anything about pending work. No firm's current queue can be broken out, because the firm name is only published with the decision. Recently filed cases are searchable by employer on the case search.
- Why a case was denied. DOL publishes the outcome, not the reasoning.
Related
- Every PERM law firm, ranked, with volume, approval and median days
- What gets denied, the factors measured one at a time
- Where DOL is working now, which sets everyone's wait
