Tracking Matter Profitability Over Time, Not Just at Closing

Most firms discover which matters made money at exactly the moment the information becomes useless. The file closes, someone runs a realisation report, and there it is in black and white: two hundred and forty recorded hours, one hundred and eighty billed, one hundred and fifty two collected, and a fee that never had a chance of covering the work once the third round of disclosure landed. Everyone nods. Someone says we should price that type of matter better next time. Then the next one comes in and gets quoted the same way, because nobody remembers the shape of the loss six months later, only the vague feeling that it was a difficult file.

The uncomfortable part is that the loss was visible long before closing. In almost every underwater matter I have looked at, the moment where the economics broke is identifiable and early. It is usually a specific week, tied to a specific event, where the work required stopped matching the fee agreed. A responsive pleading turned into a motion practice. A commercial lease review turned into a landlord negotiation. A family matter with two assets turned into a matter with a disputed business valuation. The information existed. Nobody was looking at it, because the firm's only profitability instrument was a report that fires after the work is finished.

Profitability is not a verdict handed down at the end of a matter. It behaves much more like a bank balance, moving every time someone records time, funds a disbursement, or writes off an entry they were embarrassed to bill. Treating it as a running measurement rather than a closing formality changes what you can do about it, because the only decisions that affect a matter's margin are the ones you still have left to make.

Closing-Time Profitability Is a Post Mortem, Not a Management Tool

A closing profitability report is a historical document. It tells you what was recorded, what was billed, what was written down before the invoice went out, what was written off after the client complained, and what was eventually collected. Every one of those numbers is the residue of a decision that was made weeks or months before you read it. The choice to put a senior associate on document review instead of a paralegal was made in week two. The decision to absorb the cost of a re-drafted agreement rather than have an awkward conversation was made in week seven. By the time the report is generated, all of it is sunk. You are reading an autopsy and calling it management information.

The other problem with post mortems is that they only teach if you get enough repetitions. A firm running four hundred matters a year across three practice areas will eventually see the pattern in closing reports if someone is diligent about reading them. A firm running sixty matters a year will not, because the sample is too thin and the noise too loud. One matter loses money because opposing counsel was obstructive, another because a partner took a call that should have gone to an associate, a third because the client changed instructions twice. Each looks like bad luck in isolation. The systematic problem, the one baked into how you priced the work, stays hidden underneath the anecdotes until it has cost the firm a genuinely serious amount of money.

Running Margin Is Three Numbers You Already Capture

Running margin on a live matter needs surprisingly little. You need the value of work performed so far, the expected total fee for the scope as agreed, and the actual cost of delivering the work performed so far. The first two most firms have. The third is where the exercise usually collapses, because firms measure cost using billing rates rather than cost rates, and a billing rate is a price, not a cost. If you value an associate's six hours at their charge-out rate, the matter looks profitable no matter how inefficiently those six hours were spent, which defeats the entire purpose of the measurement.

The cost rate you want is fully loaded and per person: salary, employment taxes, benefits, plus that person's share of premises, insurance, software, and administrative support, divided by the hours they can realistically be expected to bill in a year rather than the hours they are theoretically available. That last denominator is where most firms flatter themselves. Nobody bills every working hour. Once you build the rate honestly, the picture shifts hard. Work performed by a partner at partner cost on a task a paralegal could have done shows up as a cost problem immediately, which is exactly the signal you want while there is still work left to reassign.

The Inputs Already Exist, They Are Just Never Assembled

Nothing about this requires a new data collection habit. Your time entries already exist because you bill from them. Disbursements are recorded because someone has to be reimbursed. Invoices, payments, and write-offs are captured because your accountant insists. Matter stage is usually tracked informally in someone's head or in a status field nobody trusts. The only genuinely missing ingredient is a structure that puts these next to each other in a way a fee earner can read in fifteen seconds without asking the bookkeeper to build a spreadsheet.

This is where a matter stage tracker stops being a client communication nicety and becomes the denominator of a real financial calculation. In Casely the stage tracker is a clickable stepper, configurable per firm and per practice area, so a conveyancing matter and a personal injury matter carry genuinely different stage sets rather than a generic four-step pipeline neither of them fits. Once stage is a live field rather than a guess, you can ask the only question that matters: how much cost have we consumed against how much of the work we have actually finished. One-click invoicing helps here too, because turning every unbilled hour into a single itemised draft makes hidden work visible rather than letting it sit in a queue where it quietly stops feeling real.

Set the Expected Cost at Intake, Not at the First Billing Review

Running margin needs a baseline, and the baseline has to be written down before the work starts. At intake, for any matter type you have run more than a handful of times, you should be able to record a rough hypothesis: which stages this matter will pass through, approximately how many hours each will take, and at what seniority. Not a precise estimate. A hypothesis you intend to test. Firms resist this because it feels like false precision on a matter whose shape is unknown, and that objection is fair for genuinely novel work. It is not fair for the eightieth residential purchase or the fortieth employment tribunal claim.

The value of writing it down is not the accuracy of the guess. It is that a wrong guess becomes informative. If you predicted forty hours to reach the disclosure stage and you are at sixty-five with disclosure still incomplete, you have learned something specific and actionable within the matter, and something even more valuable across the practice area. Without a recorded hypothesis, the same overrun registers only as a vague sense that this file is heavy. Vague senses do not survive contact with a fee negotiation six months later. Numbers written down at intake do.

Read Margin Against Stage, Not Against Calendar Time

Firms that do try to monitor live matters almost always monitor the wrong axis. They pull a list of matters older than ninety days, or matters where unbilled time exceeds some threshold, and they call that a profitability review. Matter age tells you about aging work in progress and receivables risk, which is a genuine cash flow concern but a different one entirely. A matter can be eighteen months old and perfectly profitable because it is a litigation file that spent nine months waiting on a court listing. A matter can be five weeks old and already destroyed because someone spent three days on research that should have been a template.

The correct axis is progress through the work. Cost consumed against stage completed produces a signal you can act on, and it produces it early, because the earliest stages are where scope assumptions get tested. If your intake hypothesis said pleadings would take fifteen percent of the budget and pleadings consumed forty percent, you now know at the earliest possible moment that either the scope was misjudged or the staffing was wrong, and you still have eighty-five percent of the matter left in which to respond. That is the entire argument for stage-based reading. It moves the alarm from the point of no return to the point where response is still cheap.

The Twenty Minute Weekly Review That Makes It Operational

None of this survives as a quarterly initiative. It has to become a short, boring, repeated routine, ideally run by whoever already sits closest to billing rather than by a partner who will deprioritise it in a busy week. The routine is simple: pull every open matter, sort by cost consumed against stage completed, and separate them into three buckets. On track, drifting, breached. Most matters will be on track and require no conversation at all. The point of the review is to shrink the discussion to the handful that are not.

For the drifting ones, the question is diagnostic rather than punitive, and that framing genuinely matters if you want fee earners to cooperate rather than quietly under-record their time to keep files looking healthy. Ask what changed against the intake assumption. Was it scope, staffing, client behaviour, or an external event. Each of those has a different response, and only one of them is a fee earner performance issue. For breached matters, the question is different and more urgent: what is the plan, and does the client need to hear about it this week rather than at the next invoice.

Cost Is More Than Time: Disbursements, Write-Offs, and the Trust Line

Time is the largest cost on most matters but it is never the only one. Firm-funded disbursements distort margin badly when they sit unrecorded or unbilled: counsel fees, expert reports, filing fees, search fees, process serving, medical records, courier and travel. Each one is real cash leaving the firm, and each one is routinely invisible in a margin calculation that only counts hours. A matter can look healthy on time and be losing money outright once you account for an expert report the firm fronted three months ago and has not yet billed.

Client money is a separate discipline and has to stay that way. A trust balance is not revenue and must never be treated as a cushion that makes a thin matter look survivable. Casely enforces this structurally rather than by convention: any disbursement exceeding a matter's actual trust balance is blocked at the database transaction level rather than warned about in a dialog someone can click through, ledgers are isolated per matter, and corrections are voided and remain visible rather than deleted. That last property matters for margin work specifically, because a profitability history built on a ledger where entries can vanish is not a history at all. Rules on when you may bill against funds held, what notice the client is owed, and how disbursements may be funded vary meaningfully between US states, the England and Wales regime, the Canadian provinces, and the Australian states and territories, so confirm the specifics with your own regulator rather than assuming your neighbour's practice applies.

Spotting the Pattern Across Matter Types

One unprofitable matter is an anecdote and should mostly be treated as one. Twenty matters of the same type breaking down at the same stage is not an anecdote. It is a defect, and it lives either in how you price that work or in how you deliver it. The difference between a firm that learns and a firm that repeats is almost entirely whether it can group matters tightly enough to see the repetition, and that requires the grouping fields to exist before you need them rather than being reconstructed by hand from file names.

Practice area alone is too coarse. What you want is the ability to slice by referral source, by matter subtype, by complexity marker, and by whether a matter is connected to others. Casely's contact labels tag roles and referral sources, and connected matters link related files with the reason for the link stated explicitly, which is what lets you ask sharper questions than a practice area report can answer. You may find that one referral channel consistently sends work that overruns at disclosure, or that matters connected to an existing client relationship run leaner because the intake work is already done, or that a subtype you have always treated as routine has quietly become your worst performer since a procedural change last year.

Different Fee Models Need Different Margin Clocks

Hourly work has the gentlest failure mode, because overrun mostly converts into a bigger invoice, and the real margin risk sits at the write-down conversation rather than in the work itself. That means the running measure for hourly matters should track recorded value against expected billable value and flag entries at risk of being written down, not merely hours consumed. Flat fee work is the opposite. Every hour past the assumption is a direct hit with no recovery mechanism, which is why flat fee matters need the tightest stage-based monitoring of anything in your book and why the intake hypothesis is not optional for them.

Contingency and conditional arrangements need a different frame entirely. Cost accrues continuously while revenue stays at zero until resolution, so a running margin figure is meaningless and a running exposure figure is the useful number: how much firm cost is currently at risk in this matter, and how much across the whole contingency book. That is a portfolio question, not a matter question. Blended arrangements need their blend assumption tested against the seniority mix actually delivering the work, which frequently drifts senior over time. Casely supports hourly, flat-fee, contingency and blended billing natively and exports LEDES 1998B where institutional clients require it. Be careful with the underlying legal position: contingency fees, conditional fee agreements and damages-based agreements are permitted, capped, or prohibited very differently across US states, England and Wales, Canada and Australia, and are commonly restricted in criminal and many family matters, so confirm what is available in your jurisdiction before you build pricing around it.

Feeding the Pattern Back Into the Next Fee Quote

This is the whole point, and it is where most profitability work quietly fails to arrive. The output of running margin across a body of similar matters is a distribution, not an average. You now know that eight out of ten commercial lease reviews land near the quoted fee and two of them double because the landlord's solicitor reopens the schedule of condition. Quoting from the average of that distribution guarantees you lose money on the tail. Quoting from the best case, which is what most firms do because the best case is the one they remember, guarantees it more thoroughly.

The better response is usually structural rather than a blanket price increase. Price the stage that actually overruns, or scope it out of the fixed portion and price it separately when it triggers. Write the engagement scope so the boundary sits exactly where your own data says the work reliably breaks, rather than where a precedent letter happened to draw it. That is a far easier conversation to have with a client than a mid-matter fee variation, because it is specific, it is evidenced, and it is offered before they have committed. Firms that price this way tend to win more of the work they want and lose more of the work that was going to hurt them, which is the correct outcome in both directions.

What To Do When A Matter Is Already Underwater

Finding out early is only valuable if you do something with the finding, and the options narrow fast as the matter progresses. Early on you can restaff, moving work down to the appropriate level and keeping the partner on the parts that genuinely need a partner. You can tighten process, replacing bespoke drafting with a firm template where the risk profile allows it. You can raise scope with the client while the extra work is still ahead of them rather than behind them, which is a materially different conversation from asking someone to pay more for work already delivered. And you can decide, deliberately, to absorb the loss because the relationship justifies it, which is a legitimate business decision as long as it is a decision rather than an accident.

What you cannot do is treat profitability as a reason to disengage carelessly. Duties to a current client do not soften because the matter turned out to be unprofitable, and the rules governing withdrawal, including whether court permission is required once proceedings are underway and what notice the client is owed, vary between jurisdictions and often between courts within one jurisdiction. Confirm the position locally before acting on it. The practical takeaway is that early visibility widens your set of legitimate options, and late visibility narrows it to absorbing the loss or having a conversation that damages the relationship. That gap is the entire return on running margin.

Making It Real Inside the Firm

The obstacle to any of this is almost never analytical sophistication. It is that the underlying record is scattered across a time-recording tool, an accounting package, a shared drive, and someone's memory of where the file has got to, and reconciling those four things takes long enough that nobody does it weekly. When time entries, matter stage, disbursements, invoices and the trust ledger all live against the same matter record, the weekly review takes twenty minutes instead of two days, and a twenty minute routine is one that actually survives a busy month.

Start narrow rather than firm-wide. Pick your highest volume matter type, write down the expected shape of the work at intake for the next ten of them, and read cost against stage every week. Within a quarter you will have something no closing report will ever give you: a live picture of where that work reliably breaks, and enough evidence to price the next one properly. If you want to see how the pieces fit together, our legal reporting and analytics and legal time tracking pages walk through the underlying record in more detail.

Casely is cloud-native with no local install, documents are encrypted with AES-256 under a per-firm key, and there is a free plan so you can test this on real matters before committing anything. The measurement is not the hard part. Choosing to look while you can still act is.