How Expert Witness Directories Work

What this page does not do, and why

This page describes how the market for finding an expert is organized. It does not rank, compare or assess named directories, brokers or platforms, and that is a disclosed position rather than an omission: Hartzer Consulting is itself a listed expert in that market, and a comparison written by a participant is worth nothing to a buyer. Where a company is named below, it is because that company published something being cited, not as an endorsement or a criticism.

What follows is instead the thing a directory will not publish about itself: who pays, what the payment buys, and what the resulting entry does and does not establish. A buyer who understands the business model can read any listing correctly without needing anyone's ranking of the operators.

There is no licensing body for expert witnesses, no register, and no rule of procedure that governs how a party finds one. Every service in this market is a private business selling a marketing channel or a search service. None of them is a regulator, none of them adjudicates anything, and none of them is under any obligation to verify what appears in a profile.

The listing model: the expert pays

In the oldest and most common model, the expert buys an annual listing, often with paid upgrades for placement, category prominence or additional media. The attorney searches at no cost and contacts the expert directly. Reported listing costs run from several hundred to a thousand-plus dollars a year with premium placement extra — a figure drawn from a single interested source rather than from any survey, and useful only as an order of magnitude.

What the buyer is reading in that model is a self-authored claim of expertise, a CV supplied by its subject, contact details, and sometimes a self-reported testifying history and an indication of whether the expert works predominantly for plaintiffs or defendants. What the buyer is not reading is anything verified. Nothing in a listing has been checked against a registrar, a licensing board or a docket, and nothing has been adjudicated.

Presence in a directory is therefore evidence of exactly one thing: the expert markets for litigation work. That correlates with availability, with responsiveness, and with litigation-support volume. It does not correlate with methodological rigor in either direction, and reading it as a credential is the single most common error a buyer makes in this market.

The broker model: the law firm pays, indirectly

A referral or broker service takes a brief from the law firm, searches its own network and the open market, screens candidates against the brief, and presents a shortlist. It then bills the firm at a rate above what it pays the expert and keeps the difference. Sometimes the arrangement is disclosed as a markup; sometimes the firm sees only a single blended figure.

What that buys is real: a search run quickly by someone whose commercial relationship depends on the candidate showing up, being available, and not embarrassing the firm. For a case that needs an unusual discipline on a short timetable, that is worth paying for.

What it does not tell the buyer is the expert's own rate, in many arrangements; the size of the spread; and how wide the search actually was, as against how deep the broker's existing bench is. A commonly cited markup of 30 to 40 percent circulates in this market, but it comes from a single industry-consultant source with no survey behind it and should be treated as an order of magnitude rather than a finding. The practical question for a buyer is simply whether the quoted figure is the expert's own figure and, if it is not, what the spread is. It is a fair question and a broker's answer to it is informative either way.

The subscription model: the firm pays for the gate

In the third model the law firm pays an annual access fee, reported in the several-thousand-dollar range from the same single source, and experts are listed free or cheaply. The platform frequently does not expose individual experts until it makes an introduction, which is the point: the product is curation, and disintermediation is the thing the fee exists to prevent.

What the buyer gets is a filtered set, delivered faster than an open search would deliver it. What the buyer does not get is the basis for the filtering. Whether the curation reflects a review of prior testimony, verification of credentials, an internal quality rating, the platform's own margin on each expert, or simply who answered the phone that week is not ordinarily disclosed, and there is no standard that requires it to be.

None of these three models is disreputable. All three are marketing and search channels, and each solves a real problem for the party who pays. The error is category confusion: treating any of them as a vetting. A listing is a purchase. A broker's shortlist is a service. A curated introduction is a filter with an undisclosed basis. Vetting is a separate exercise that the buyer performs, or does not.

How the money moves, and why it shapes what a buyer sees

Following the payment explains most of what is otherwise puzzling about these services.

• Expert-pays listings optimize for expert acquisition and renewal. Coverage is broad in fields with many experts who market, and thin in fields where the leading practitioners have no need to advertise. The best person in a narrow discipline is frequently not listed anywhere, because they have never had a reason to buy a listing.
• Broker services optimize for placement volume and margin. That favors candidates who are known to the broker, quick to respond, and comfortable with the commercial terms — a set that overlaps with, but is not identical to, the set of best-qualified candidates.
• Subscription platforms optimize for firm renewal, which rewards speed and a high hit rate on introductions.

One consequence is worth stating for buyers who search only these channels: three of the seven routes to an expert are marketing channels, and the other four — bar and industry referral, the record of prior testimony, published work in the field, and academic and professional bodies — are records that nobody pays to appear in. Searches confined to the paid channels systematically miss the practitioner who does not market.

What a listing can be used for

Read as a lead source rather than as a verdict, a directory entry is genuinely useful. It produces names in a discipline, a starting point on availability, and a CV that can be checked. It is also a document worth preserving: a profile written by the candidate is a statement of qualifications made outside this case, and comparing it against the CV produced during the engagement is one of the cheapest integrity checks available. Divergence between a directory profile, an old conference program and the current CV is a red flag that costs nothing to look for.

Two things in a profile repay close reading. The first is a stated plaintiff or defense orientation, which many experts publish voluntarily. It is self-reported and it is checkable against the litigation record, and a divergence between the two is worth understanding before retention rather than after. The second is the claimed specialty. Profiles are written to be found, which pushes them toward breadth: an expert whose listing claims a dozen adjacent specialties has optimized for search rather than for cross-examination, and each claimed area is a line an opposing lawyer may ask the witness to justify.

The listing is also what the other side will find. Anything a candidate has written about themselves for marketing purposes — a claimed specialty, a success statistic, an implied alignment with one side of a recurring dispute — is available to opposing counsel and can be put to the witness. Counsel who retains an expert without reading that expert's own marketing has left a document in the case that only one side has read.

The question no directory answers

Across all three models, the one thing a buyer most needs is the one thing none of them supplies: whether this witness's methodology has ever survived a challenge under FRE 702. That fact exists only in the litigation record, it is scattered across opinions, dockets, filings and transcripts, and it is not for sale as a product.

Two corollaries follow. First, the search that answers it is a separate search, run against dockets and full-text filings rather than against any platform's index. Second, even that search returns an incomplete answer, because most rulings under Rule 702 are made from the bench or in unpublished orders and never become searchable text. Finding no adverse ruling establishes that the search was clean, not that the record is.

Which is why the useful sequence is the opposite of the intuitive one. The paid channels are efficient at producing candidates and useless at ranking them. The unpaid records — testimony history, published work, verifiable credentials, and what the candidate says on a first call about the limits of the available data — are inefficient at producing candidates and are the only material that bears on whether the eventual opinion holds up.

Frequently Asked Questions

Published 2026-08-12