← All writing

The staffing math stopped working

Four things that are true about plaintiff-side practice in 2026, and why none of them are solved by better recruiting.

A hiring problem has a hiring solution. Firms that describe their review capacity as a hiring problem have usually already tried the solution, twice, and are now describing the symptom in the language of the thing that did not fix it.

Four facts sit underneath most of what plaintiff-side firms are dealing with this year. None of them are about recruiting, and they compose in a direction that does not reverse.

The pool is not growing

The Bureau of Labor Statistics projects little or no change in paralegal and legal assistant employment between 2024 and 2034. Roughly 376,200 people are in the occupation and about 39,300 openings are projected annually — most arising because people transfer out of the occupation or leave the workforce entirely, not because firms are creating seats.

The median wage was $61,010 as of May 2024, before payroll tax, benefits, space, equipment, and the ramp period during which a new hire consumes capacity rather than adding it. A four-attorney boutique posting a role is not hiring into a growth market. It is competing for churn, against firms that pay more.

The files got heavier

NCCI's 2026 State of the Line reports that workers' compensation lost-time claim frequency declined about 2 percent in 2025, while medical claim severity and indemnity claim severity each rose roughly 4 percent. Fewer claims, each carrying more medical documentation and more cost. California's WCIRB found in June 2026 that total costs for cumulative trauma claims have more than doubled since 2020.

The same direction shows up outside comp. National Center for State Courts data reported in 2026 shows probate and estate filings rose roughly 32 percent between 2020 and 2024 across 39 reporting states. Fewer, larger, more document-intensive files is the pattern across injury practice generally.

The volume concentrated

Judicial Panel on Multidistrict Litigation statistics as of March 2, 2026 show 67,115 pending actions in the Johnson and Johnson talcum powder MDL, 23,695 in Bard polypropylene hernia mesh, 15,220 in AFFF, and 11,440 in hair relaxer. Roughly 158 active MDL dockets carried close to 198,000 pending actions entering 2026.

At that scale, plaintiff fact sheets, census filings, and records authorizations stop being paperwork and become an operations discipline. A dismissal for a documentation default costs a firm exactly what losing on the merits costs.

The day did not get longer

Clio's benchmark data puts the average lawyer at roughly 3.0 billable hours out of an eight-hour day — a 38 percent utilization rate. That number has been stable for years, and it is stable for a reason: the other five hours are not idle. They are absorbed by work that has to happen and has no better owner.

What composes

Read those four together. Volume per matter is rising. Documentation per matter is rising. The local hiring pool is flat. The working day is fixed.

Every one of those is a supply constraint on the same scarce input — the attention of somebody trained enough to make a judgment about a case. And the constraint is structural, which means the usual responses are all forms of paying more for the same shortage.

The only lever that has any room left in it is what that attention is spent on.

In most firms, a reviewer's hour is dominated by finding and assembling: chasing a custodian, splitting a compiled PDF, reconstructing a treatment timeline, working out which of four hundred pages actually bears on causation. That is real work and it gates everything downstream. It is also not judgment work, and it is not what the firm is short of.

The firms that got out from under this in the last eighteen months did not out-recruit the market. They separated the assembly from the judgment, gave the assembly to a system, and kept every judgment exactly where it was.

That is a structural answer to a structural constraint, which is the only kind that holds.

Field notes

Three of thirty scenarios in this series. All are composites, built from patterns we see repeatedly across the practices we work with rather than from single named engagements. Firm details are illustrative; every figure cited traces to a named source.

01 · The boutique that kept losing paralegals

Personal injury and medical malpractice · 4 attorneys · review owned by the surviving senior paralegal · deployed with Specter review

In-house tenureendedendedendedReviewed capacity · continuousCoverage stopped tracking headcount
01Three departures, one continuous standard

Three in-house paralegals in eighteen months. Each departure meant weeks of unfilled capacity followed by a new hire who needed months to reach file fluency, and the one senior paralegal who stayed absorbed every gap — which is the fastest way to lose her too.

What changed. The firm stopped hiring against churn. Discovery management, deposition support and medical record review moved onto the pipeline, with review capacity added rather than headcount. File standards were written down across every active matter, so institutional knowledge stopped living in one person's memory. The senior paralegal moved to setting the standard and reviewing exceptions.

Outcome. Coverage held through two trial periods without the usual scramble. She stopped being the shock absorber for every vacancy, which is why she is still there. Attorney contact: two to three a week, all on case strategy.

02 · The filing backlog she did not create

Probate · solo practitioner, three counties · review owned by her legal secretary

Probate filings, 2020–2024+32%Paralegal employment, 2024–2034little or no change~39,300 openings a year, mostly replacement
02Rising filings against a flat hiring pool

Filing volume rose steadily without any change in marketing. The work simply kept arriving, and the legal secretary was drafting petitions between answering the phone.

What changed. This is a demographic trend, not a marketing outcome — NCSC data shows probate and estate filings up roughly 32 percent between 2020 and 2024 across 39 reporting states, against a hiring pool the BLS projects flat through 2034. The estate lifecycle was standardised into defined stages, county-specific form variation was encoded once rather than rediscovered per matter, and statutory notice periods moved onto a calendar with dual reminders.

Outcome. The practice absorbed materially higher filing volume without adding local headcount, and the legal secretary stopped drafting between phone calls.

03 · Forty files to a hundred and twenty

Personal injury · solo practitioner · review owned by her paralegal, who became team lead

Utilisation by case stageIntakeRecordsTreatmentDemandNegotiationThree stages constrained · none of them judgment work
03Capacity modelled by stage, not by headcount

She wanted to grow and had run the numbers on a second local hire twice. Against contingency cash flow they did not work: the BLS median paralegal wage of $61,010 as of May 2024, plus payroll tax, benefits, space, equipment and a ramp period during which the hire consumes capacity rather than adding it, is a fixed cost against lumpy revenue.

What changed. Capacity was modelled by case stage rather than by headcount, which made the constrained stages visible — intake, records, demand assembly, file maintenance. Those stages moved onto the pipeline. Her paralegal moved from doing the work to setting the standard and reviewing it.

Outcome. The caseload roughly tripled with no change in local headcount. Clio puts billable revenue per responsible lawyer at roughly $83,219 for solo firms against $156,963 at firms with two to four employees; that gap is mostly capacity, and capacity turned out to be purchasable without a hire.

Keep reading

See Specter on your caseload.

A twenty-minute walkthrough on your torts, your jurisdictions, your intake volume.

No commitment · 20 minutes