Lawyers Leave. The Firm's Memory Must Stay.

Every law firm strives to build institutional memory: the know-how and reasoning that accumulate across years of work. For most of the last century, that was easier, because partners stayed at one firm for their whole careers. That is no longer true.

In late September 2026, The Economist reported that "Big Law's Century-Old Business Model Is Being Transformed." Big Law, the largest and highest-earning American law firms, has started to look like a sports league. Star partners move like free agents, and many of their clients move with them. One New York partner summed up the shift: "You used to hire firms. Now you hire individuals."

That raises a hard question for every firm. If the talent can leave and the clients will follow, what makes one firm worth more than another?

The Old Model

For more than a century, large American law firms ran on a system built in 1903 by Paul Cravath, whose New York firm still carries his name. Firms hired top law graduates as associates, the junior lawyers who do much of the day-to-day work. They pushed them hard for several years and made the best of them partners, the senior lawyers who own the firm and bring in clients.

That system had an unintended benefit: knowledge stayed because people stayed.

Problem One: Partners Are Now Free Agents

That system is now buckling. Between January and mid-September 2026, more than 3,200 partners joined one of America's 200 highest-grossing law firms, more than in all of 2021, according to data firm SurePoint, as reported by The Economist. Star partners who switch can command $20 million or even $30 million in pay. Even Cravath, the firm that built the old model, hired a star dealmaking partner away from rival Weil Gotshal.

When a partner moves, the client relationship often moves too. However, research firm Decipher Investigative Intelligence estimates that more than 60 percent of lateral hires fail to bring in the business they promised, according to the same Economist article. That is a strong counterpoint. Taken at face value, that suggests clients are more loyal to firms than the free-agent story implies, and that a firm's client base is safer than it looks.

Both can be true, but the counterpoint proves less than it seems. Failing to bring in the business promised is not the same as bringing none. Partners negotiating eight-figure pay packages have every reason to overstate their books, so many moves fall short of a high bar. Some clients cannot follow because of a conflict of interest at the new firm. Others stay only because the old firm cuts its rates to keep them. So clients do follow their partners, just fewer than promised.

And the right comparison is with the past. Before the 1980s, partners rarely moved between large firms at all, according to research by legal scholar Robert Hillman. Even a modest share of clients following their partners today is a large shift.

Whatever the exact share, the direction is clear. Partner moves are rising, and firms show no sign of stopping: many managing partners, The Economist reports, "feel they must poach or die." A partner whose clients will follow has little reason to stay loyal to a firm. And even when a client stays, the partner's knowledge does not. The firm keeps the relationship but loses the person who knew the work best.

So for a firm to stay valuable and competitive, it must hold on to the one asset that does not have to walk out the door: its institutional memory.

Problem Two: Associates Are Leaving Too

A 2025 survey of more than 800 senior figures at UK and US law firms, by legal software company BigHand, found that 16 percent of junior associates had left private practice at law firms in the previous 12 months, up from 9 percent the year before, according to Legal Cheek.

Ravi Daparthi, chief executive of legal software company LawVyn, explains what that number hides in his essay "A Law Firm's Memory Has a Notice Period." When a junior lawyer leaves, nothing is technically lost. Yet the drafts that worked, the research behind them, and the reasons for each decision stop being findable, because the person who knew where they were is gone.

"Reasons are the most valuable memory and the least recorded," he writes, "because they were explained once, across a desk, and that was that."

Firms are losing knowledge through their partners and their associates at the same time. That is the unintended consequence of a more mobile profession: institutional memory leaves with the people who hold it.

The Insight: Without Memory, Firms Become Interchangeable

When a client hires a law firm, it is not only paying for a lawyer or a team of lawyers. It is paying for what the whole firm has learned: the precedents, the positions that held up, the lessons from similar cases. That institutional memory is what should set one firm apart from another.

New lawyers may be every bit as talented as the ones they replace. But perhaps talent is not the issue. Perhaps time is: institutional memory builds only when lawyers stay long enough to connect one case to the next. When they do not, the experience behind those cases ends up scattered across people and firms.

What remains is, to borrow Daparthi's phrase, "a group of individuals sharing a letterhead." If the knowledge walks out with the lawyers, the firm keeps its name and its offices, and little else that a rival cannot match.

The Biggest Firm Sees the Same Problem

The industry's highest-earning firm is spending heavily here. Kirkland & Ellis, the first law firm to top $10 billion in annual revenue, has set aside $500 million to build its own AI platform. The platform will draw on input from 250 of the firm's own lawyers, including 100 partners, who have shared details of how they work. Part of the aim, Bloomberg Law reported, is to harness the firm's "collective knowledge and judgment" and "put it at every Kirkland lawyer's fingertips." A Kirkland spokesperson said the goal is to "bring Kirkland's institutional experience into every engagement."

In other words, the most successful firm in the industry is paying to turn what its lawyers know into something the firm owns.

Most firms cannot spend $500 million. But they face the same question Kirkland is answering: how to make the firm's knowledge belong to the firm, not to whoever happens to hold it.

What Lasting Institutional Memory Looks Like

The fix is not to stop lawyers from moving. It is to keep the history of each deal in a form that does not depend on any one person staying. That means a record that:

• Links each decision to the contract language it affected
• Notes the reason for a concession when it is made, not years later
• Lets the lessons of one deal inform the next, whoever is staffed on it

Where Aracor Fits

Aracor was built by people who have sat in these seats, a team that has closed multibillion-dollar transactions and worked in more than 20 countries. Among them: a former co-chairman of one of the world's leading law firms, named a Dealmaker of the Year by The American Lawyer; a former co-head of private equity at another global firm; the founder of more than two dozen technology companies, with more than 350 patents to his name; and a chief executive who is a practicing attorney and has been on every side of the table, as a law firm partner, chief legal officer, and public company director.

They know this problem from the inside out, and built the solution they would want themselves: one platform built for the deal, designed to preserve the institutional memory that firms lose when people move.

It keeps a continuously updated record of how each deal evolves: its documents, analysis, findings, risks, requests, and decisions, along with the evidence and context behind each conclusion. A firm can save its own review methods and apply them to every new deal, so that knowledge stays with the firm, not with the lawyer who developed it.

When a lawyer leaves, the record stays with the firm.