Data Intelligence · Methodology

Methodology

What we count, over which windows, what we leave out, and how to tell us when a number is wrong.

Hire legal talent Partner recruiting
01 Reading the data

What the data show

Sources: public law-firm announcements, firm websites and professional profiles, checked monthly. We publish analysis, never rows: totals, shares, medians, ranks and bands.

Windows: each edition covers the 12 months to the end of the last complete month (November 2025 to October 2026 in the November 2026 edition). Year-on-year changes compare October 2025 to September 2026 with October 2024 to September 2025.

A lateral move: a lawyer leaves one law firm and joins another; the start date at the new firm is known to the month.

Why the newest month reads low: the newest month reads low because start dates are confirmed with a delay, so year-on-year changes use 12-month windows that end one month earlier.

Reading a change: a change under 5% in either direction is flat: older years miss the lawyers who have since left private practice, which lifts recent years by a few points.

Moving up a level: share of lateral moves where the lawyer joined at a higher title than the one left (associate or counsel to partner, associate to counsel).

Changing city: share of lateral moves where the lawyer also changed city, among the moves where both cities are known (so it cannot be multiplied by the total of moves).

Group moves: share of lateral moves that were part of a team joining the same firm in the same quarter.

Promotions per lateral partner: lawyers promoted to partner for each lateral partner hired, same 12 months.

Time to partner: years from the first year of practice to the promotion; city = where the lawyer practices today.

Leaving rate: share of a firm's lawyers who left it, measured on firm websites since April 2026 and annualized.

Net partner flow: lateral partners hired against partners who left for another firm, 12 months; associates are left out because firms that train their own associates lose them sideways by design.

Index against the market: 100 = the same share as in the whole market; 150 = one and a half times the market share.

Practice and industry shares: share of moves (or deals) involving the practice or industry: one lawyer or deal can carry several, so these shares add to more than 100 and are never a split of the market.

Team moves and office openings: team moves and office openings are public announcements: counted as they are, firms named when the announcement is public.

Moves out of private practice: tracked in full from May 2026; no earlier comparison is published.

Bands: a firm row carries a range instead of a count: lawyers hired (lateral lists), partners promoted (promotion lists), deals advised (deal lists), people (group moves).

Promotion windows: promotion figures stop at May 2026, the last month that is complete: "12 months" of promotions = June 2025 to May 2026, and promotions per lateral partner sets them against the lateral partner hires of the same 12 months. Later months are not published.

Where leavers go: lateral = another law firm; in-house = a company; government; academia; out = left private practice with no new employer on record.

Moves to companies: a lawyer leaves a firm for a company: a legal department in most cases, but companies that sell technology to lawyers also hire them for product roles, so the employer list is "companies that hired lawyers out of law firms", not "legal departments".

Shares on small counts: a count under 20 is not printed, but its share is when it rests on 5 or more cases: such a share is a rough size, not a ranking tool.

Deal shares: shares of deal types, practices and industries are shares of all dated deals; about one deal in six carries no industry and one in ten no type, so the shares can add to less than 100.

Firm rankings: firm lists, bands and ranks count individual lateral hires; lawyers who arrived as one team in the same quarter are in the market totals but not in a firm's rank.

Open roles: job openings live on law-firm career pages on the data date: a stock on one day, not the number posted in a period; an opening leaves the count when the firm takes it down.

Where the work is done: share of the industry's deals (those with a known lawyer location) that had lawyers working on them from the city; one deal can count in several cities, so this is where the work is done, not where the lawyers are counted.

Deal counts: deals include disputes taken from public court and trademark-office records, which are many in intellectual property: a count is matters on record for a practice, not a measure of revenue, and practices are not ranked against each other by it.

Team size: lawyers identified in the team move; it can be lower than the size the firms announced.

Thresholds: a cell with fewer than 20 moves is not published and the page shows nothing in its place. A firm is named only with at least 10 moves in the window; firms appear only in top-10 lists, as a rank and a band. No lawyer is named and no client is tied to a firm.

Editions: monitors are rebuilt from each new edition. A monthly edition keeps its own month.

Corrections: if a number on these pages is wrong, write to [email protected] with the page address. We check it against the source and correct the next edition.

Edition November 2026. Window: November 2025 to October 2026. Source: Sartori & Partners, from public law-firm announcements, firm websites and professional profiles, checked monthly. Methodology. Spot an error? [email protected].

For law firm leaders

Planning a partner hire in this market?

We run retained partner searches using the same data behind this page.