What Drives an Expert’s Fee

A bill is two numbers, and only one of them is negotiated

Almost every conversation about what an expert costs is a conversation about the hourly figure, and almost every difference between one engagement and another is a difference in hours. Two experts at the same quoted rate, working the same question in the same field, routinely produce bills that differ by a factor of several — because one of them was given an organized record, a defined question and a workable schedule, and the other was given a hard drive and six weeks.

That is not an argument for ignoring the rate. It is an argument for understanding which of the two numbers a buyer can actually move. The rate is set by a labor market the buyer did not create and cannot negotiate very far. The hours are set by decisions counsel makes: what the expert is asked to answer, what the expert is given to answer it with, when the material arrives, how many times the question changes, and how much of the work is analysis rather than assembly.

What follows is about the second number, and about the small number of contractual terms that decide how the first one is applied. The published market data — who runs the surveys, what they sampled and what they report — is a separate subject, and it is worth understanding before reading any national figure as a benchmark.

What actually sets the rate, and what it is not evidence of

An expert’s rate reflects the value of the day being given up. That is why the highest figures in every published table sit with medical specialties: an afternoon of deposition is an afternoon of clinic or theater that cannot be rescheduled, and the rate has to clear that hurdle before the question of litigation work even arises. The same logic sets the floor for an engineer who bills client work, an economist who teaches, or a professional whose practice is seasonal.

Two further inputs move it: the market the expert works in, and how long the person has been marketing themselves for litigation work. A candidate whose practice is substantially litigation support has a rate shaped by that market rather than by an opportunity cost, which is a different thing and worth recognizing as such.

What the rate does not measure is quality. No study was located, in the research behind this site, correlating an expert’s hourly figure with admissibility outcomes, with surviving a challenge under FRE 702, or with any measure of testimony quality; so far as that research can establish, none exists. A high-billing surgeon and a moderately billing metallurgist are not two points on one scale — they are two labor markets. The most misleading thing a buyer can do with a national average is treat it as a floor and read anything below it as a discount, and the second most misleading is to treat a figure above it as a proxy for standing in the field.

The hours are set by the record, not by the question

The single largest driver of what an engagement costs is the state of the material. A question that takes two hours to answer on a clean, indexed, native-format production takes twenty on the same underlying facts delivered as printed images with no load file, no index and three overlapping copies of the same set.

The variables are unglamorous and they compound. Volume. Format, and whether the analysis has to begin by making the material usable. Completeness, and whether the decisive record has been produced at all. Sequence, because material that arrives after the analysis is built forces the analysis to be rebuilt rather than extended. And duplication, because an expert who cannot tell which of three versions is operative has to reconcile all three.

The second driver is definition. An expert asked to opine on what happened will read everything; an expert asked whether a specific method supports a specific conclusion on a specific date range will read what bears on that. The narrower question is not a cheaper question because the expert does less thinking. It is cheaper because it makes exclusion possible — the expert can put material aside knowing it is outside scope, and every hour of reading avoided is an hour not billed and, under FRCP 26(a)(2)(B)(ii), one less item in the list of facts or data considered.

Why testimony is priced above review even at one rate

Every published dataset shows testimony priced above file review, and the reasons are structural rather than a premium for saying the words out loud. A testimony day cannot be sold in parts: the calendar is blocked whether the deposition runs two hours or eight. Testimony dates slip, and they slip late. And appearance is the highest-exposure work the expert does, on a transcript that will be quoted back for years.

Those facts show up in the terms rather than in the rate. Minimums for deposition and trial days are common, in half-day and full-day forms. Cancellation policies that retain all or part of an appearance charge inside a stated notice window are common too, and they exist because an appearance cancelled two days out cannot be resold. A buyer reading a rate card without those terms is reading half of it.

There is one further asymmetry that nobody in the market states plainly. Under FRCP 26(b)(4)(E), the deposition of a testifying expert is time for which the party seeking the discovery must pay a reasonable fee unless manifest injustice would result — so the deposition rate is largely a charge to the opponent rather than to the retaining client. Courts have noticed. Where the deposition rate sits far above the rate the retaining party is charged, the difference invites a motion, and the reasonableness factors courts apply, drawn from Jochims v. Isuzu Motors, Ltd., 141 F.R.D. 493 (S.D. Iowa 1992), and restated in Hose v. Chicago & North Western Transportation Co., 154 F.R.D. 222 (S.D. Iowa 1994), include the fee actually being charged to the party who retained the expert.

The part the other side does not pay for

The belief that the other side pays for the expert is durable and wrong in a specific way that costs money. FRCP 26(b)(4)(E)(i) shifts a reasonable fee only for time spent responding to discovery under Rule 26(b)(4)(A) or (D) — the deposition of a testifying expert, and court-ordered discovery from a non-testifying consultant. It does not reach file review, analysis, the writing of the report, trial preparation or trial testimony. In most engagements those are the bulk of the bill.

Nor is the shifted amount whatever the expert charges. The rule says a reasonable fee, and courts have repeatedly cut the shifted rate toward what the retaining party itself was paying. Whether the expert’s preparation time for the deposition shifts at all is one of the genuinely unsettled questions in federal practice: there are four identifiable approaches among district courts, districts in the Ninth Circuit consistently decline, districts in the Second Circuit consistently award, and where preparation is shifted courts have converged on something like a 3:1 ratio of preparation to deposition hours as the ceiling in complex cases — a convention, not a rule, and one no appellate court has adopted.

Two budgeting consequences follow. Preparation time is real money that may or may not be recoverable, depending on the district. And a party planning around recovery it has not confirmed is planning around a number that changes with the courthouse.

Where the money reliably gets spent twice

Waste in an expert engagement is rarely extravagance. It is repetition, and it comes from a short list of recognizable causes.

• Moving assumptions. An analysis built on a date, a population or a figure that later changes is an analysis rebuilt, not adjusted.
• Material arriving in waves. Each late production reopens work already done, and the reopening is more expensive than the original pass because it also requires reconciliation.
• Scope drift. A question that widens by email produces work nobody has budgeted and, often, work nobody uses.
• The expert doing assembly. Collating, converting, de-duplicating and indexing are billable at an expert’s rate and are frequently not expert work.
• Rework as rebuttal. An opinion that belonged in the opening report and appears in a rebuttal has been paid for twice if it is struck — and rebuttal that carries the disclosing party’s affirmative case is routinely struck.
• Cancelled appearances. Inside the notice window, a cancelled deposition or trial day is often billed in full, and it is the one line item that buys nothing at all.

Every item on that list is a scheduling or scoping decision made by counsel rather than a pricing decision made by the expert. That is the practical answer to the question of what a buyer controls.

Compression is the most expensive thing a buyer can purchase

A schedule that leaves an expert three weeks for six weeks of work does not produce a compressed version of the same engagement. It produces a different and more expensive one: parallel workstreams that duplicate effort, analysis begun before the record is complete and revised when it is, review at the pace at which mistakes are made, and — most costly of all — an opinion that has to be defended on a record the expert did not have time to test.

The pattern is visible in the sanctions case law from the other end. Severity under FRCP 37(c)(1) correlates strongly with proximity to trial: a defect found early is commonly cured with a supplemental deposition and some cost-shifting, while the same defect found on the eve of trial produces exclusion. Time is the resource that converts a problem into an inconvenience, and it is bought at the start of an engagement or not at all.

Working backward from the disclosure deadline is the whole discipline. The report needs a complete record; the record often needs a production or a deposition; both need requesting; and requests have their own clocks. Counsel who run that arithmetic in the first week usually discover that the schedule was decided before the expert was chosen.

The terms that decide the bill more than the rate does

Four clauses in an engagement letter do more to determine what an engagement costs than any negotiation over the hourly figure.

Invoicing cadence. Monthly billing is the recommended default in the ABA’s litigation materials for one reason: it makes spend visible while it is still possible to act on it. An engagement invoiced at the end is an engagement whose budget was never managed, only reported.

The retainer, and whether it replenishes. A one-time deposit applied against early invoices and an evergreen retainer topped back up to a floor are different instruments with different cash-flow consequences. Model engagement letters in the field offer several alternative formulations precisely because the choice is a drafting decision rather than a market default.

Minimums, cancellation and travel. Half-day and full-day minimums, the notice window inside which a cancelled appearance is still billed, and whether travel and waiting time bill at full or reduced rate. There is no national convention on travel; in federal fee-shifting practice some courts apply a half-rate presumption and others allow the full rate.

Who is liable for payment. Counsel or client, and what happens if the client stops paying. Courts split on whether a retaining attorney is personally liable in the absence of an express disclaimer, and the clause is cheaper than the authority.

What no fee arrangement can buy

Compensation for a testifying expert cannot be contingent on the outcome. ABA Model Rule 3.4(b) prohibits offering a witness an inducement prohibited by law, and Comment [3] states that the common law rule in most jurisdictions is that it is improper to pay an expert witness a contingent fee. Accrued Financial Services, Inc. v. Prime Retail, Inc., 298 F.3d 291 (4th Cir. 2002), held that a structure in which prospective expert witnesses shared in the recovery amounted to offering expert testimony for a contingent fee and would violate public policy. District courts have excluded testimony on that basis; Tagatz v. Marquette University, 861 F.2d 1040 (7th Cir. 1988), instead treats the arrangement as a credibility question for the jury. The ethics position is clear in most jurisdictions and the evidentiary consequence is not, which is a reason to keep well away from the line.

Nor is any of it confidential. FRCP 26(a)(2)(B)(vi) requires the report to state the compensation to be paid for the study and testimony in the case, and FRCP 26(b)(4)(C)(i) excludes communications relating to compensation from the protection given to attorney–expert communications. Some courts go further into the expert’s economics: Behler v. Hanlon, 199 F.R.D. 553 (D. Md. 2001), ordered disclosure of the share of the expert’s gross income attributable to expert work over five years, a five-year case list and a ten-year list of retaining insurers.

What money buys is time and attention. It does not buy a conclusion, and an arrangement built as though it might is an arrangement that will be described to a jury in exactly those terms.

Frequently Asked Questions

Published 2026-08-12