The firm hired between eight and twelve lateral partners a year and had done so for six years. It could describe every one of those hires in detail: practice mix, the book as presented at offer stage, the guarantee, the partner who had championed the hire.

What it could not describe was which of them were in trouble. Departures arrived in the third year. The decision behind them was usually months older than that, and by the time the firm could see it the guarantee was spent.

The question that started the engagement was how early the firm could tell.

What the firm had been measuring

The firm measured originations, quarterly, against the book presented at offer stage.

Originations are the thing the firm bought, and the partnership is entitled to know whether they arrived. The problem is timing. Originations move last, after referral patterns, after the sponsor relationship, after the credit conversation. A lateral whose integration failed in month five shows up as an originations problem in month fourteen, and the firm is then nine months into a situation it could have changed.

The firm also ran a check-in at six months, conducted by the partner who had championed the hire. Nobody involved thought this produced candid information. The champion had staked their own judgment on the hire, and the lateral knew it. What came back was the answer that kept both parties comfortable.

What the program changed

The program changed three things, and none of them was the question set.

Custody moved outside the firm. Check-ins were administered by SRA, with responses held externally. Laterals were told who held the data and what the firm would receive.

The cadence moved forward. Thirty days, ninety days, one hundred and eighty days, in place of a single conversation at six months. The failure signals are time-ordered, and an instrument that arrives once at month six misses the two earliest.

The sponsor was asked separately. The sponsor's account of a given three months and the lateral's rarely matched, and where they diverged was what the firm acted on.

What the check-ins surfaced

The signals appear in a consistent order. These are the median points at which each becomes detectable across SRA client engagements, not this firm's own timings, and the firm used them as the frame for reading its results:

When each integration failure signal becomes detectable, from SRA instrumentation across client engagements.
SignalTypically detectableWhat it predicts
Client introductions promised but not madeMonth 3Referral flow never starts
Sponsor has disengagedMonth 5No internal advocate when credit is allocated
Credit and compensation frictionMonth 9The economic case for staying weakens
Book fails to followMonth 14The stated rationale for the hire is gone
DepartureMonth 30 and beyondRecorded as a lateral that did not work

Two things about that sequence mattered to the firm more than the individual rows.

The first is that the two earliest signals are behavioral rather than financial. Neither appears in a billing report. A partner who has not made the introductions they committed to at offer stage is visible in month three to the lateral and to nobody else.

The second is that those two are also the cheapest to fix. At month three an introduction can still be made. At month five a sponsor can still be replaced. By month fourteen the firm is managing a departure it has already lost the ability to prevent.

The sponsor problem

The finding that changed the most at this firm was not about laterals at all.

Sponsorship had been committed at the point of hire and assumed thereafter. There was no date in the calendar on which anyone was asked whether they had actually done it, and the partner who championed a hire in a competitive process had, by month four, moved on to championing the next one.

Asking the sponsor and the lateral separately made that visible for the first time. In a meaningful number of cases the sponsor reported the integration as going well and the lateral reported having had no substantive contact in six weeks. Neither was lying. The sponsor had heard nothing to worry about, which is not the same as having been in contact.

Put both in the same room and they settle on one version before anyone writes anything down.

What the firm changed

The changes the firm made after the first round, with the owner of each.
ChangeOwnerWhat it replaced
Sponsor named in the offer memo, with the commitment written downPractice group leaderAn assumed sponsor, usually the champion
Check-ins at 30, 90 and 180 days, administered externallySRAOne conversation at six months, run by the champion
Sponsor interviewed separately at each intervalSRANobody asked the sponsor anything
Each round closes with a decision: intervene, adjust, or exitPractice group leaderA status update that recommended nothing
Integration carried into the sponsor's own reviewManagement committeeSponsorship with no consequence attached

The last row was the contested one, and it took most of a year. The objection was not that sponsorship is unimportant. It was that a partner should not carry a mark for a hire that failed for reasons outside their control, which is fair.

What made it workable was narrowing what the sponsor is accountable for. Not whether the book followed, which depends on the lateral's former firm, their clients and the market. Only whether the introductions, the internal advocacy and the contact they committed to at offer stage actually happened. Those are things a sponsor controls completely, and the check-ins establish whether they occurred.

What happened

Across two hiring cohorts run under the program, third-year lateral partner departures fell 41 percent against the firm's own baseline before SRA took over administration.

Two cohorts is not a long series, lateral outcomes depend on a market that moved over the same period, and no instrument can separate the effect of measuring something from the effect of the attention that measuring it attracts. The firm's own reading was that the check-ins mattered less than what the check-ins forced: a named person with a written commitment and a date on which someone would ask whether it had been met.

The secondary effects were easier to attribute. Two hires were exited inside the first year rather than the third, on evidence rather than on a sense that it was not working. Both decisions were uncomfortable and both were cheaper than the alternative, given that a failed lateral costs between 200 and 400 percent of that partner's annual compensation once recruiter fees, guarantees and replacement are counted.

What nearly stopped it

Two objections came close to stopping it.

The first was the argument that laterals would find the check-ins intrusive. They did not. The completion rate ran high from the first round, and the free-text responses were longer at ninety days than at thirty. A partner who has just changed firms generally wants someone to ask how it is going, provided the asking is not done by the person whose decision is under review.

The second was a request, in the second cohort, for a practice group breakdown on a group with three laterals in it. That could not be given without making the three identifiable to a group leader who worked with all of them daily. The answer was no and the finding went out in aggregate. With three respondents any breakdown is attributable, and the laterals had been told at the outset that it would not be.

What actually did the work

The questions were not what made the difference. Integration check-in design is well covered, and most professional development teams could write a workable set.

What the firm did not have was a named person, a written commitment, a date, and an answer sourced from somewhere other than the partner with the most invested in it going well. The first three require nothing but a decision and a calendar entry.

The research behind this, including what the portability data actually shows and why the book is the last thing to move, is set out in our article on lateral partner integration at US law firms.

The client is not identified, at the firm's request. Outcome figures are approved SRA engagement metrics held under client contract, measured against the firm's own pre-engagement baseline and published anonymized. The detection-lag table is SRA instrumentation data across client engagements rather than this firm's own timings, and is labeled as such. Third-party figures are cited inline.

Sources: Decipher Investigative Intelligence, lateral partner portability and retention research. Passle survey of 100 managing partners at top 200 US and top 100 UK firms. Major, Lindsey & Africa Lateral Partner Satisfaction Survey. SRA engagement metrics, 2023 to 2025.