Publishing the Split of Non-Billable Hours
The remainder is where the diagnosis lives. Seven categories, in hours, with no target on any of them.
One team's non-chargeable quarter, split by what the hours were
312 hours. The cost of winning the work that will be billed next year.
243 hours. The firm's quality mechanism, and non-chargeable.
206 hours. Systems, compliance, approvals, internal reporting.
184 hours. The first category to fall when anything presses.
148 hours. Templates, methodology, the knowledge base.
105 hours. Available, willing, not staffed.
97 hours. Work sold and blocked on the other side.
1,295 hours, and the diagnosis lives in all seven lines rather than in the ratio above them. Five of the seven are things somebody other than the team decides. One firm's own quarter, not a benchmark.
Utilisation reports the share of capacity spent on client work. The rest of the week is reported as a single residual, if at all.
The commercial question in “Publishing the Split of Non-Billable Hours” needs a clean distinction between time recorded, time approved, time billed and money received. Teams evaluating consult the current Monitask guide for billable and non-billable hours can strengthen the work-record part of that chain while leaving invoice totals, tax and settlement to the appropriate finance system.
The residual is where the diagnosis lives. Splitting it into seven named categories, in hours, turns an evaluative number into a usable one.
As a separate benchmark for “Publishing the Split of Non-Billable Hours”, see ILO working-time resources. External guidance is most useful when it prompts a specific check rather than replacing case-by-case judgement.
The seven categories
Training and development. Supervision and review. Business development and pitches. Internal projects. Administration. Waiting on client input. Bench.
Seven lines, monthly, by team, in hours.
Why hours and not percentages
Because hours can be acted on. "Supervision fell by 58 hours" names a quantity somebody can restore.
A percentage of a residual that is itself shrinking is hard to reason about and easy to misread.
What each line tells you
Training: whether development is happening. Supervision: whether the quality mechanism is running. Development: the cost of winning work. Internal: whether improvement work survives. Administration: the overhead load. Waiting: client-side process problems. Bench: the gap between capacity and work.
Seven questions, each answered by a line.
The categories that must not have targets
All of them. A target on any line produces the same reclassification the rest of this collection describes, one category down.
The split exists to be read. Its value is entirely in being an honest record.
Why waiting deserves its own line
Hours lost to blocked work are the cleanest evidence a firm has about client-side process problems, and they are routinely absorbed into administration or bench.
Separated out, they support a conversation with the client that is otherwise impossible to have with evidence.
Why administration deserves one
Because it is the category most likely to be growing unnoticed. Systems, compliance, internal reporting, approvals.
A firm that cannot see its administrative load cannot tell when it has doubled.
The movement line
Each month, the largest change among the seven, named in a sentence.
That sentence is what makes the page worth reading. Without it the seven lines are a table nobody interrogates.
What the split makes visible
Deferral, immediately. Any category that falls by more than half in a quarter was reclassified or postponed, not eliminated.
That check is the main defence against everything in the behaviour section, and it only works if the categories exist.
Setting the categories up
Timesheet codes mapped to seven groups, once. Most firms already have twenty or thirty non-billable codes and no grouping.
The mapping is an afternoon's work and it is the whole implementation.
Resisting the urge to add more
Seven is enough. Twenty categories produce a table nobody reads and rows too small to be meaningful.
Where a firm needs more detail, it drills into one category rather than splitting the page further.
What the numbers will show in the first month
Usually that supervision is lower than anybody believed, that administration is higher, and that bench time is concentrated in one or two teams.
Three findings, none available from the ratio, all actionable.
The reporting pair
Utilisation at team level, with this split beside it. Two tables on one page.
The first says how much capacity went to clients. The second says what the rest was for, which is the question the first one raises and cannot answer.
What to do with a category that looks wrong
Ask, rather than act. A supervision line at two per cent might mean review is not happening or might mean it is being coded to the engagement.
The distinction matters and is settled by asking two people how they record it. Acting on the number without asking produces an instruction to record differently, which is the only thing an instruction can produce.
Most surprising numbers in the first few months of this report are coding questions rather than findings.
Keeping the seven stable
The categories are only useful across time, so they should not be revised. A category redefined in year two breaks the comparison with year one.
Where a new category is genuinely needed, it is added and the old ones left alone, with the addition dated.
This is the same restatement discipline the definitions section describes, applied to the composition rather than to the ratio. The temptation to tidy the categories arrives annually and should be resisted: a slightly awkward set of seven with four years of history is worth more than an elegant set of nine with none.
The mapping exercise
Most firms have twenty or thirty non-billable codes accumulated over years, several of them near-duplicates.
Mapping them to seven groups is an afternoon with a spreadsheet, and the mapping is the whole implementation: no new codes, no change to how anybody records, no retraining.
The old codes keep working and the report reads the groups, which also means the historical data maps too and the first version arrives with years of history.
Why this is the highest-value addition
Because it costs no new data collection, it has no target to game, and it answers the questions firms otherwise try to answer with a ratio that cannot.