The setup

The firm had grown to roughly 200 attorneys across four offices and a handful of practice groups, with a respectable timekeeping system, a competent matter-management platform, and a billing engine that had been customized over the better part of a decade. The systems individually worked. Together, they did not produce a number anyone trusted on the day they needed it.

The managing partner could pull a utilization report, but it lagged the close by two to four weeks, came out of a spreadsheet maintained by one person in finance, and didn’t reconcile cleanly to the billing run. Practice-group leaders were running their own variants of the same analysis on extracts of varying age. By the time anyone could see that a given attorney had drifted under target, the month was gone and the conversation was retrospective.

Underneath the visibility problem was a capture problem the firm suspected but couldn’t size. Timekeepers were entering hours into the system, but a meaningful fraction never made it onto a bill — written down at pre-bill, dropped during partner review, or never associated with the right matter in the first place. The firm wanted a defensible answer to “how much are we leaving on the table,” and a way to act on it before the next close.

What we did

We started by treating the three core systems as a single dataset. Timekeeping, matter management, and billing each had a usable export; what they didn’t have was a shared identity layer for attorneys, clients, and matters, or a shared definition of when an hour counted as captured. We built that layer in a governed warehouse, with the join logic version-controlled and tested against the prior twelve months of closed bills so the firm could see exactly which historical numbers it agreed with and which it didn’t.

On top of that layer we built a daily utilization and capture view at the grain partners actually manage to: attorney, practice group, office, and matter type. The view answered the questions the firm was already asking — billable hours per timekeeper, target attainment, realization, write-down rate — but it answered them yesterday rather than next month. Where a number disagreed with the legacy spreadsheet, the lineage was one click away.

The view also exposed the leakage points. Hours that sat in draft for more than a week. Matters with timekeeping but no billing arrangement on file. Pre-bill reviews where the same partner consistently wrote down the same kind of entry. Each of those was a small fix in isolation; together they were the gap between the hours the firm was working and the hours it was capturing.

We didn’t replace any of the existing systems. We made them legible together, and we put the resulting view in front of the people who could act on it.

What changed

Billable capture rose roughly 11% within two full billing cycles of the daily view going live, measured against the firm’s own twelve-month baseline. The lift came from the leakage points the data had made visible — entries cleared out of draft faster, matters opened with a billing arrangement on day one, and pre-bill conversations that happened against a shared number rather than competing extracts. Write-down rate fell 2.3 percentage points over the same window.

The pre-bill review cycle compressed by roughly 40%, because reviewers stopped reconciling between tools and started reviewing the bill itself. Practice-group leaders moved from monthly retrospectives to weekly check-ins, because the data supported the cadence. The managing partner stopped asking finance for a utilization report; the report was already on the dashboard when the day started.

The firm kept its timekeeping system, its matter-management platform, and its billing engine. The change was in how data moved between them and who could see the result.

Why it stuck

The warehouse layer is documented and tested, so the join logic doesn’t quietly drift when a system is upgraded. The view is owned by the firm’s own analytics function, with the definitions written down and agreed across finance and the practice groups, so nobody is relitigating what “captured” means each quarter. The 11% number is the externally visible result; the internal result is that the firm now manages utilization on the day it can still do something about it.