Why Per-Matter P&L is Broken at Most Law Firms
Most firms cannot tell you which matter made money last quarter. The four hidden costs that distort matter-level profit, and the one-page P&L view that actually works.
The 19-point gap
We rebuilt 92 matter P&Ls by hand across the firms in our 2026 benchmark sample. The median matter margin dropped 19 points. About a fifth flipped from reportedly profitable to actually loss-making, almost none of them the matters the originating partner would have named.
The question is not whether Indian law firms are profitable, on the right view, most are. The question is whether the partnership knows which matters and which partners are doing the carrying. Mostly, they do not.
The maths is not hard. The conventional view leaks profit in four predictable places, and the matters that look most profitable are almost always the ones leaking the most.
Ninety-two matters, two ways
The standard view is wrong in the same direction every time
When firms build a per-matter view at all, it usually has four lines. Revenue is the invoiced amount. Cost is fee-earner hours at a blended rate, often the associate rate. Overhead is a flat percentage of revenue. The leftover is called profit. The originating partner sees the number, nods, and the partnership moves on.
Every line is wrong in the same direction. Revenue is overstated, pre-invoice write-downs pool at the firm level. Cost is understated, partner time is half-recorded, paralegal and ops time is missing. Overhead is misallocated, flat percentages ignore what the matter actually consumed. And the cost of carrying the matter for the 142 days between work done and cash collected is nowhere on the page.
The result is that the matters that flatter the originating partner most, high-revenue, partner-led, generous discount cycles, long-tail collection, sit at the top of the league table. The matters that quietly subsidise the firm, repeat-client opinion work, well-scoped fixed fees, monthly-billed disputes, sit in the middle, where nobody looks.
Leak 1: Partner time that never gets recorded
On every matter of consequence, the most expensive person in the room is a partner. On most timesheets we audit, the partner records less than half of attributable time. The median across our sample is 47%. Senior partners are worse than juniors. Disputes partners are worse than corporate partners.
The downstream effect on per-matter P&L is severe. A matter that recorded 18 partner hours actually consumed 38. At a loaded partner cost of INR 14,000 per hour, the weighted median for senior equity partners across our 19-firm financial sample, that gap is INR 2.8 lakhs of cost the matter never saw. On a matter that reported a 41% margin on a INR 30 lakh fee, the true margin is closer to 32%. Roll that effect across a portfolio of 200 matters and firm-level margin is overstated by 7 to 11 points.
The fix is not better timers, it is a definitional change. Partner time is firm cost whether or not it is billed. The timesheet is for the firm’s management accounting, not just the client’s invoice. Most firms conflate the two, which is why most matter P&Ls flatter the partner more than they help the firm.
Leak 2: Write-downs that pool at the firm level
This is the leak that surprises managing partners most, and it is the largest in rupee terms.
The pattern is universal. A matter runs over budget. The billing partner pulls hours off the draft invoice before it goes to the client. The matter records the invoiced amount as revenue. The written-down hours sit somewhere, but, at most Indian firms, that somewhere is a firm-wide realisation pool, not the matter that generated them.
On the matter’s own page, the partner appears to have collected 100% of what they billed. At the firm level, realisation runs at 71%. Both numbers are technically correct. They are also useless, because nobody can answer the question that follows: which matters caused the 29 paise gap?
The correct treatment is unromantic. Every write-down is charged to the matter that generated it, before margin is calculated. The ranking reorders within a quarter. Matters with discounts agreed upfront, flat fees, capped retainers, usually look fine. Matters with quiet pre-invoice write-downs, the ones the originating partner did not want to surface, drop sharply. Those are the matters that should have been re-scoped or re-priced months earlier.
- Recorded value · INR 42.0 lakhs
- Invoiced · INR 31.5 lakhs
- Reported revenue on matter · INR 31.5 lakhs
- Cost · INR 19.0 lakhs
- Matter margin · 40%
- Recorded value · INR 42.0 lakhs
- Write-down assigned to matter · INR 10.5 lakhs
- Reported revenue on matter · INR 31.5 lakhs
- True cost incl. unrecorded partner time · INR 26.0 lakhs
- Matter margin · 17%
Same matter. Twenty-three points of margin, gone, once two leaks are accounted for. Across the 92 matters we rebuilt by hand, the median downward revision was 19 points. The largest single revision was 47 points, a matter the originating partner had defended in two consecutive partnership meetings.
Leak 3: Overhead that pretends every matter is the same
Most firms apply overhead as a flat percentage of revenue, or as a flat uplift on fee-earner cost. It is fast, defensible at audit, and structurally wrong. Two matters with identical revenue can consume wildly different amounts of the firm’s shared cost.
A short opinion for a repeat client consumes almost no firm overhead, a quick conflicts check, no KYC refresh, no library time, no paralegal, no COO time, no dedicated workspace. A two-year arbitration for a new client consumes all of it, continuously, for two years. Allocating the same flat percentage to both is a polite way of saying we would rather not look.
The fix does not require academic activity-based costing. Three buckets, applied per matter, get you 80% of the way:
- Intake load. A one-time charge per matter for conflicts, KYC, engagement letters, file opening, and matter setup. In our sample this runs INR 18,000 to INR 35,000 per matter, depending on practice. It is roughly the same for a INR 2 lakh opinion and a INR 2 Cr deal; pretending otherwise distorts small matters disastrously.
- Support load. Paralegal, ops, knowledge management and secretarial time, attributed by hours actually consumed, not by revenue share. Most firms can recover this from existing timesheets, if their support staff records to matter codes. Most do not.
- Seat cost. Premises, technology, leadership, finance, allocated per fee-earner-month on the matter, not per rupee of revenue. A matter that ran a senior associate for six months consumed six months of seat cost, irrespective of what was billed.
Move from flat-rate overhead to these three buckets and a familiar pattern appears: short opinions look worse, and long, low-rate engagements that appeared profitable often turn out to be loss-making. The uncomfortable part is the point.
Leak 4: The carrying cost the matter never sees
A matter is not finished when the invoice goes out. It is finished when the cash lands. The median Indian firm sits at 142 days of lock-up between work done and cash collected. Across a firm, that translates into crores of working capital tied up in matters the partnership treats as closed.
Per matter, this never appears. The 90 days the client took to pay, the 60 days the partner sat on the draft, the 30 days the matter ran in WIP, none of it costs the matter anything on the page. It costs the firm interest, opportunity, or, in the worst case, the line of credit it had to draw on to make partner distributions.
The correction is small but decisive. Charge each matter for the days it sat. At a working-capital cost of 11% per year, a matter holding INR 30 lakhs in WIP and AR for 160 days costs roughly INR 1.45 lakhs. On a matter with a reported margin of INR 5 lakhs, that is 29% of the margin, paid in carrying cost the matter was never asked to account for.
The point is not to bill clients for it. Two matters with identical reported margins are not equally profitable when one collects in 60 days and the other in 220. The monthly-billed, disciplined-collection matter is structurally more valuable. Most firm P&Ls treat them as equal.
The corrected one-page P&L
Once the four leaks are accounted for, a per-matter P&L fits on a single page and reads honestly. Six lines, in order.
| Line | What it captures | Example, mid-size M&A matter |
|---|---|---|
| Standard-rate revenue | All recorded hours at standard rate, including unbilled partner time | INR 42.0 lakhs |
| Less: assigned write-downs | Pre-invoice discounts charged to this matter, not pooled at firm level | (INR 10.5 lakhs) |
| Less: loaded fee-earner cost | Partner + associate cost at fully-loaded rate, on actual time spent | (INR 17.8 lakhs) |
| Less: matter overhead | Intake load + support hours + fee-earner-month seat cost | (INR 8.2 lakhs) |
| Less: lock-up carrying cost | Days in WIP + AR × daily working-capital rate | (INR 1.5 lakhs) |
| True matter margin | What the matter actually contributed to the firm | INR 4.0 lakhs · 13% |
From INR 42 lakhs to INR 4 lakhs
Conventional view: 40% margin. Corrected view: 13%. The 27-point gap is not pessimism, it is cost the firm absorbed, recorded somewhere other than on this matter.
Across a portfolio of 200 to 400 matters, three things become visible that the conventional view systematically hides.
- The most profitable matters are usually not the largest or the most visible. They are repeat-client, well-scoped, monthly-billed engagements with disciplined partner time.
- Between 15% and 25% of matters are loss-making once fully loaded. Most are not the ones partners would have guessed.
- The ranking of partners by true profit contribution is rarely the same as the ranking by revenue origination.
The last of those is the finding that, more than any other, changes how the firm runs.
What actually changes, and who fights it
Two things change once a firm runs this view consistently, one operational, one political. The political change is why this is not already standard practice.
The operational change is straightforward. Loss-making matters get renegotiated, repriced or, occasionally, exited. Pricing on new mandates is set against the loaded view, not the blended associate rate the proposal team has been quietly using. Billing moves from milestone to monthly on long engagements, because the carrying cost now sits next to the margin where the partnership can see it.
The political change is harder. Once partner contribution is computed on true matter margin, the partnership table looks different. The biggest originators sometimes contribute less than the quiet ones, because the matters they bring in consume more partner time, eat more support load, and lock up cash for longer. The corrected view does not just measure differently; it threatens compensation, which is why most firms quietly resist rolling it out.
Whoever proposes the corrected view first is, by default, accusing the firm’s top-billing partners of being its biggest cost.On Partner Politics
Partner contribution, two views
One firm in our sample ran the corrected view on the prior 18 months of matters. Their conventional partnership ranking on revenue had been A, B, C, D, E. The corrected ranking, by true profit contribution, was D, A, E, B, C. Partner D, a quiet IP-practice head who had never made the top three on revenue, was the largest contributor. Partner C, a high-origination corporate partner, was the smallest.
The first partnership meeting after that reading was the most consequential the firm had held in five years. Two of the three structural changes that followed, a fee committee on new mandates above INR 50 lakhs, and a monthly mandatory WIP review at practice-head level, were proposed by Partner C, who knew which way the wind was now blowing.
Why this is harder than it sounds
Running this view monthly across three or four hundred matters, with assigned write-downs, activity-weighted overhead, fully-loaded partner cost, and live lock-up days, is hard to do without a system designed for it. That is why most firms don’t.
Annual rebuilds in Excel are possible, we have done them with senior associates on secondment, and the output is real. The moment you try to run the view monthly, the spreadsheet collapses. And the entire value is in the cadence: catch a loss-making matter in month two, not at year-end when the damage is locked in.
That is where most firms stop. It is also where the top-quartile firms in our benchmark, 84% realisation, 89-day lock-up, INR 4.1 Cr PEP, pull away. None of them are using a spreadsheet by the time they cross 50 fee-earners.
What to do this quarter
Three things, in order. The first two cost nothing.
One. Take your ten largest matters of the last twelve months. Rebuild each one on the six-line view, by hand. The exercise will take a senior associate two days. The output will reorder your sense of which matters and which partners are actually profitable. Do not skip this. The leverage of every later decision depends on first seeing the gap honestly.
Two. Stop pooling write-downs. Whatever the system, from this quarter on, write-downs get assigned to the matter that generated them. The firm-level realisation number will rise in honesty and fall in headline. The matter-level conversations that follow are the ones the firm should have been having for years.
Three. Pick one practice group and run the corrected per-matter P&L monthly for two quarters. Not annually. Monthly. By the end of the second quarter the partners in that group will be using the view without prompting. That is the signal to extend it to the rest of the firm.
A standing offer
The reason per-matter P&L is broken at most firms is not that the maths is hard. It is that the conventional view flatters the partners who originated the matters, and the partners build the views. Fixing it is a definitional move, partner time as firm cost, write-downs as matter-level events, overhead as activity-driven, lock-up as a real charge, and then the discipline to look at the resulting numbers without flinching.
The firms that do this do not become more profitable because they have new revenue. They become more profitable because they stop subsidising the matters that were always losing money, and start defending the matters that were always paying for the firm.
A note on the data. Median values referenced in this piece are drawn from the same survey set as the 2026 Indian Law Firm Benchmark: nineteen Indian law firms across eighteen months, nine of which shared the line-item time, billing and write-down ledgers behind their financials. The 92 matter-level rebuilds come from those nine firms. All firm and partner examples are anonymised.
Firmtalk and this view. Of the nine firms in the matter-rebuild subset, six now run a version of the corrected per-matter P&L monthly through Firmtalk. The remaining three run it quarterly on Excel exports. The firms running monthly added an average of 6.3 points of firm margin in their first year of the cadence. The firms running quarterly added 1.9.
