Originally published on LinkedIn on 4 June 2026.
On paper, the corporate team is the success story, instructions up, utilisation high, and the new partner is winning work the firm wouldn’t have seen two years ago. Then the finance report lands, and two numbers are moving the wrong way.
Lock-up is climbing. Realisation is slipping. The team is busier than it has ever been, and the firm is converting that work into cash more slowly and keeping less of it.
Nothing in the headline figures explains why. The work is there. The demand is there. The talent is there. The problem sits underneath it all, in the part of the operation the dashboard does not show.
What the numbers are telling you
Lock-up is climbing, and the pressure is in the front half of the cycle, not invoices sitting unpaid, but work sitting unbilled between completion and billing. Meanwhile realisation is falling, and a flat-out team with falling realisation should not sit together. When they do, it usually means some of the work being done never reaches a bill.
Both numbers are symptoms and neither tells you where the loss is happening. For that, you have to look at how the work moves through the team.
Where the money is actually going
There are two leaks.
The first is the over-engineering bottleneck. Files reach the point where they are ready to go out, and then they wait for final review and every file is held to the same exhaustive standard. The straightforward matter reviewed as closely as the complex one. Some of that queue is capacity. But often it is a review standard that has not adjusted to the volume or the stakes.
The instinct is sound because quality matters but applied without distinction, it leaves matters sitting behind a single reviewer who cannot clear them fast enough.
This costs the firm twice. 1) It ages WIP, because every file held in review is work done but not yet billed, and 2) it erodes margin, because the most expensive person in the team is re-checking work already done competently below them with partner rate applied to associate output.
A mid-size firm in 2026 will likely have AI review tools pointed straight at this, running a first-pass check and taking the mechanical load off the reviewer. On paper that clears the bottleneck. In practice it relocates it. The file still does not move until the partner signs it off, and the partner inclined to review everything to the same depth will go over the AI output as closely as the original draft, distrusting the flag and re-checking what was already checked.
The tool compresses the work and the partner re-expands it.
The second leak is communication variance. In a fast-moving corporate team, a lot of valuable advisory happens informally, such as a call to reassure a client, an email clarifying a point, a view offered in an unscheduled meeting. Each is real work, often high-value, and in an exhausted team much of it is never recorded. On hourly work, that is billable time leaking straight out of the system.
On fixed-fee or capped work it is worse: the time was never going to be itemised, so it vanishes into margin with no trace and no one ever decides to write it off. It’s work the firm did for free without choosing to.
What it costs if it runs unchecked
Neither leak corrects itself. They compound.
Industry benchmarks put median lock-up at around 75 days therefore the work a team completes now does not convert to cash for roughly two and a half months. A team whose files are stuck in review sits the wrong side of that, and the gap widens every month. The cash that should be funding the growth is tied up in unbilled work that ages a little more each week.
The leakage annualises into a number that would stop a managing partner cold if it ever appeared as a single line. A handful of unrecorded interactions per fee earner per week, across a growing team, across a year, is not a rounding error. It’s a meaningful share of the team’s output, earned and discarded.
There is also a slower, more dangerous cost. The team carrying this is already stretched. A bottleneck that never clears and a standard never quite reachable wear people down. The strongest performers, the ones with options are the first to conclude the growth is being delivered on their backs. They leave, taking client relationships and institutional knowledge with them, and the firm spends more replacing them than it would have spent retaining them.
Why the usual fixes don’t hold
Most firms reach for the obvious levers. They chase the WIP, and the numbers improve for a month before drifting back. They tighten time recording, and compliance lifts briefly and fades. They add an associate on the assumption the problem is capacity.
None of these hold, because none touch the cause. The WIP is not ageing because nobody is chasing it; it’s ageing because work is stuck in review. Time is not leaking because people forgot the policy; it’s leaking because the expectation to capture it was assumed rather than set. Adding an associate to a team with a single reviewing bottleneck does not relieve the bottleneck. It feeds it.
The cause underneath the numbers
Sometimes rising lock-up is genuinely structural. A billing system that slows invoicing down, or a major client that simply pays late, will push the number up on their own. Those causes are worth ruling out, because the dashboard alone will not tell you which one you have. The numbers show the symptom, not whether the cause is structural or sitting with the people leading the work. That’s exactly why it pays to look properly before spending money fixing the wrong thing.
When the structural explanations are ruled out and the numbers still move, what is left is usually a leadership pattern, and it traces to one place: a new partner in the role, still operating the way that got them there. The review standard has not adjusted to the new role.
The instinct to own every detail personally was an asset when the work was theirs to do but when applied to a team’s output it becomes a constraint. Secondly, the expectation to capture ad-hoc advisory was never set because to the partner it is obvious, so it goes unsaid, and what goes unsaid leaks.
Neither is a flaw in the partner. Both are the predictable result of hiring a new partner for individual excellence and asking them to lead without recalibrating how they operate. These are operational blind spots, not failures of effort or talent, but costs that form in the gap between what the dashboard shows and how the work is actually being led.
How to fix it
Address the leadership pattern, and the numbers follow.
Start with the review standard: an explicit, agreed definition of what “good enough to release” means for each tier of work, so the routine matter is not held to the standard of the most complex one. That judgment clears the bottleneck and returns the partner to the work only they can do.
Then set the expectation on “capture” explicitly. Not a reminder or a policy email, but a stated standard that ad hoc advisory is recorded as a matter of course, with the partner modelling it. Expectations that are demonstrated land. Expectations that are assumed leak.
Put a light structure around both: a short, regular review of where files are sitting and why, but structure only works once the pattern beneath it has shifted. Systems do not fix a bottleneck made of judgment, and policies do not fix a standard that was never set.
The team has not changed. The market has not changed. What needs to change is small, specific, and sits with one person. When it does, the files move, the time gets captured, and the cash the firm has already earned finally reaches the bank.
Operational blind spots don’t show up on the P&L until they’re expensive. The free leadership diagnostic is where you find them first. This is the link to schedule a call. https://calendly.com/andre-28/30min?month=2026-06