What the work actually is

A PAGA notice is not a lawsuit. It is a filing with the Labor and Workforce Development Agency that starts a clock, and the decisions made inside that clock determine the ceiling on exposure for the rest of the case. Most employers spend the first three weeks locating the right people internally. That is three weeks of a thirty-three day window.

The first task is calendaring, not analysis. Two cure deadlines can be running, and they belong to different tracks. An employer that employed fewer than one hundred employees during the period covered by the notice has thirty-three days from receipt to submit a confidential cure proposal (section 2699.3(c)(2)(A)); if a conference follows, the cure itself is due by the date agreed there, and no more than forty-five days after it. Separately, and whatever the employer’s size, an employer whose only cure is of a wage statement violation has thirty-three calendar days from the postmark of the notice (section 2699.3(c)(3)(A)). The sixty-day window governing the thirty percent penalty cap under section 2699(h) runs from receipt. The conservative calendar anchors everything to the postmark, because it is the earliest date the employer can prove — but the anchor is a practice habit, not the statute, and the tracks should be calendared separately so that missing one does not look like missing the other.

The second task is scoping what the notice actually alleges, as distinct from what it recites. Most notices plead a long list of Labor Code sections by boilerplate. The gap between the sections listed and the conduct factually alleged is the first defense lever, and it is also the first input into the exposure model — because a category that is pleaded but not supported is a category a plaintiff will have difficulty proving and may be willing to drop.

The third task is the model itself: penalties disaggregated by violation category, per-category violation rates drawn from the actual time and payroll data rather than assumed, categories PAGA does not authorize stripped out under ZB, N.A. and Kirby, and three scenarios produced — plaintiff maximum, data-driven realistic, and defense best case. That model is what a mediator, a carrier, and opposing counsel will all end up arguing about. It is better to be the party that built it.

After Adolph v. Uber Technologies, compelling individual claims to arbitration no longer disposes of the representative action; the plaintiff keeps standing and the case stays in court. The strategic question moved from elimination to containment, and containment is won on scope.

The primary tool is manageability. Estrada v. Royalty Carpet Mills held that trial courts may not strike a PAGA claim as unmanageable outright, but it also confirmed the constitutional principle underneath — a representative proceeding must still allow the defendant to present individualized defenses. The 2024 reforms then codified express authority in section 2699(p) to limit the evidence presented at trial or otherwise limit the scope of the claim. Where a single notice sweeps together multiple job classifications, multiple worksites, or multiple compensation structures, that provision is an invitation to narrow the case before discovery costs are incurred rather than after.

The second front is the data. Plaintiffs extrapolate. Duran v. U.S. Bank National Association sets the constitutional floor for doing that lawfully: the sample must be representative, the methodology defensible, and the defendant must retain the ability to contest individual claims. Controlling how the sample universe is defined, how a violation is defined within the sample, and whether premiums already paid are netted out routinely moves the violation rate by more than any argument made at mediation.

The third front is arbitration, which after Hohenshelt v. Superior Court is no longer a strict-liability trap. Late payment of arbitration fees forfeits the right to arbitrate only where the nonpayment was willful, grossly negligent, or fraudulent. Administrative failures that would have been fatal under the prior line of Court of Appeal authority are now defensible on the record.

And where the case resolves, a PAGA-only settlement still has to be approved. Moniz v. Adecco requires the settlement to be fair, reasonable, and adequate in light of PAGA’s purposes, and Kullar v. Foot Locker requires counsel to show the investigation actually done. Motions that recite those cases without applying them are the ones that draw questions from the bench.

A class action reaches back three years for wage claims, four under the unfair competition law, and it recovers wages. A PAGA claim reaches back one year and recovers penalties. The same facts therefore produce two very different exposure figures, and a single blended number will simultaneously understate the wage damages and overstate the penalties. Carriers setting reserves on a blended figure are setting them on a number that describes neither track.

Certification is where the class track is won or lost, and the argument is almost always about variation: whether the policy was uniform, whether the practice was uniform, and whether determining liability requires looking at individual employees, supervisors, or locations. The evidentiary work is building that record early — classification-by-classification, location-by-location — rather than asserting it in opposition briefing.

The regular rate is where the money usually hides. Overtime and premium calculations that use the base hourly rate instead of the regular rate produce systematic underpayments that repeat every pay period for every employee earning any non-discretionary compensation. Ferra v. Loews Hollywood Hotel requires meal and rest premiums at the regular rate. Alvarado v. Dart Container requires flat-sum bonuses to be divided by non-overtime hours rather than total hours. Most payroll systems do neither correctly, and neither side tends to quantify the gap precisely.

Complaints involving executives are structurally different from ordinary personnel matters. The reporting lines that normally handle a complaint often run through the person complained about, the population of witnesses is small enough that confidentiality is fragile, and the eventual audience for the report may be a board, an acquirer, or a court rather than a human resources file.

The work is scoping the allegations precisely, identifying and sequencing witnesses so that accounts are not contaminated, documenting what each witness was asked and what they said, evaluating the evidence against the applicable standard rather than against instinct, and writing findings that state what is supported, what is not, and what could not be determined — in language that survives being read aloud.

Independence matters to the weight the report carries. An investigation conducted by outside counsel with no stake in the outcome, on a defined scope, with a written methodology, is materially harder to attack than one run internally by people who report to the participants.

Section 2699(g) reduces penalties to fifteen percent for an employer that took all reasonable steps to comply before receiving a PAGA notice. Section 2699(h) offers thirty percent for reasonable steps taken within sixty days after. No published appellate decision had construed the reform's cap standards, manageability provision or cure procedures as of August 2026. The question of what qualifies is therefore answered by the documentation an employer can produce, and by how early it was created.

A pre-notice audit is therefore two exercises at once. The first is finding the actual exposure: meal and rest practice as it operates rather than as written, regular rate calculations across every compensation component, wage statement compliance against the nine elements of section 226(a), alternative workweek elections and whether the underlying process was valid, and off-the-clock patterns visible in clock-in data.

The second is building the record: written policies distributed and acknowledged, supervisor training with attendance documented, periodic payroll audits with findings and corrections, and a timekeeping system capable of showing that breaks were provided rather than merely scheduled. Each of these is a remediation item and simultaneously an exhibit for the cap.

The economics generally favor doing this early. Remediation of a defective practice is measured in months of operational change; the penalty exposure it prevents is measured per employee, per pay period, for every period the practice continues.

The most common failure in carrier reporting is a status report that describes activity rather than exposure. Depositions taken and motions filed do not tell a claims professional what the file is worth. A carrier-ready report states the current exposure range, the assumptions driving it, what would move it in either direction, and what the recommended settlement authority is at this stage.

On PAGA and wage-and-hour matters that means separating the tracks: class damages on their own lookback, PAGA penalties on theirs, derivative penalties identified rather than folded silently into a total, and the effect of the penalty caps modeled rather than asserted. Where the reforms apply, the anti-stacking provision in section 2699(i) materially changes the derivative arithmetic, and a report that has not accounted for it is reporting the pre-reform number.

Reporting cadence is set to the carrier’s requirements. The default here is an initial assessment within the first days of assignment and a written status report with the initial exposure model inside the first two weeks, updated at each milestone that changes the number.

PAGA defense has become a specialized practice in a short period of time. The 2024 reforms changed the penalty structure, the cure mechanics, and the scope tools all at once, and the appellate law interpreting them is still being written. A general commercial or employment practice can litigate the case competently and still leave substantial reductions on the table simply because the penalty architecture is unfamiliar.

The most common arrangement is discrete: the exposure model and the penalty analysis, produced as a work product the referring firm uses in its own briefing and mediation. Others are fuller co-counsel arrangements covering the wage-and-hour claims while referring counsel handles the balance of the matter. Conflicts are checked before anything substantive is discussed.