Attorney Billing: How Law Firm Billing Works, From Time Entry to Payment

Almost every conversation a small firm has about money is really a conversation about the rate. What is far less examined is everything that happens between the moment work is performed and the moment cash lands in the operating account, which is where most of the firm’s revenue is actually won or lost. A rate is one number chosen once a year. A bill is hundreds of small decisions made in a hurry, each of which has to survive a client reading it, and some of which have to survive a fee arbitrator reading it. The rules on those decisions are more specific than most owners realize, and the specificity is good news, because it means the answers are written down.

What attorney billing is, and the four stages it moves through

Attorney billing is the process by which work performed becomes money received. It has four stages, and they are genuinely separate: the work is recorded, the record is reviewed, the reviewed record is rendered to the client as a bill, and the bill is collected. Firms tend to think of this as one activity called “billing” that happens at the end of the month, which is why weaknesses at different stages get diagnosed as the same problem and treated with the same remedy, usually new software.

They are not the same problem. A firm whose lawyers record time three weeks late has an accuracy problem that no amount of invoice polish will fix. A firm whose entries are accurate but whose prebills go out unreviewed has a write-down problem it will only discover when clients start querying. A firm that bills cleanly and promptly but does not follow up has a receivables problem that looks, from the bank balance, exactly like not having enough clients. The term “billing attorney” refers to the lawyer who owns the client relationship and reviews the bill before it goes out, which tells you which of the four stages the profession has historically treated as the one requiring a named owner.

The rest of this piece walks the four stages in order, using the two authorities that speak most directly to each of them. The first is ABA Formal Opinion 93-379, issued in December 1993 and still the reference point for what may and may not be billed on an hourly basis. The four stages below describe hourly work; a matter taken on a percentage of the recovery closes differently, and our guide to contingency fee agreements covers what that writing must say and what has to be produced at the end. The second is the State Bar of California’s Arbitration Advisory 2016-02, written to tell fee arbitrators how to read a lawyer’s bill for padding. The second document is the more useful of the two for a working owner, for an odd reason: it is the only widely available text that describes, item by item, what an experienced neutral looks for when deciding whether your invoice is honest. It is a checklist written for the other side of the table, which makes it the best self-audit a firm can run.

Stage one, recording time: the entry is the product

The first thing to be clear about is how little of a working day is billable even when everything goes right. Yale Law School’s career office publishes an arithmetic exercise for students deciding whether a billable hour target is survivable, and it is worth reading as an owner rather than as a recruit. Assume a lawyer is at the desk from 8:00 to 6:00, a ten hour day. Take an hour for lunch, half an hour for two short breaks, half an hour reading legal updates and general correspondence, and half an hour for internal meetings and CLE. Ten hours at the desk becomes 7.5 hours that can be billed. Across a five day week that is 50 hours present and 37.5 recorded, and across 47 working weeks it is an annual average of 1,762 billable hours.

The exercise is written about large firm targets, and the firms it describes are not small firms. What travels is the ratio. Three quarters of a well-run day is the realistic ceiling for billable capture at the desk, before a single administrative fire, before the bookkeeping, before a client calls about something that is not a matter. Small firm data puts the real figure much lower: measured across a large sample of small and midsize firm billing data, the average utilization rate, meaning the share of an eight hour day actually recorded as billable, was 37 percent in 2024. Our law firm profitability guide works through what that number does to a firm’s economics. The point here is narrower: if capture is the constraint, the fix is at stage one, not stage four.

Ten hours at the desk, and what is left to bill

Given how thin the margin is, the single highest-return habit in a law firm is recording time as the work happens. This is not a preference. It is the difference between a record and a reconstruction, and neutrals treat the two very differently. The California advisory is blunt about it: contemporaneous records are almost universally acknowledged as best practice, and where a lawyer instead goes through the file later and ascribes a time entry to each email, letter and calendar item, the resulting bill may look precise while the appearance of accuracy is deceptive. The longer the gap between the task and the entry, the greater the probability that the entry is wrong, and the more likely it is that a lawyer in a fee dispute will be asked to carry the burden of proving otherwise.

The second habit is specificity, and it is worth seeing the contrast in the advisory’s own examples. One entry reads: “meeting with client to discuss the elements of the separate statement of facts and the source of evidence for each element (1.8); research new opinion on the presumptions and burden of proof under Festo and progeny (2.5).” The other covers the same work: “meeting with client re MSJ and research burden of proof (4.3).” The first tells the client what was done, by whom, and how long each piece took. The second is a single 4.3 hour block that a client cannot interrogate and an arbitrator cannot evaluate. Both bills are for the same work at the same rate. Only one of them will survive being questioned.

Rounding and increments: where a defensible rate becomes an indefensible one

Almost every hourly fee agreement contains a minimum billing increment, and almost no client reads it closely. Formal Opinion 93-379 permits the practice in one clause: a lawyer may not bill more time than she actually spends on a matter, “except to the extent that she rounds up to minimum time periods (such as one-quarter or one-tenth of an hour).” That sentence is the entire authority for rounding, and it does two things at once. It blesses the increment, and it makes the increment the only permitted departure from actual time.

The size of the increment is where the trouble starts. A tenth of an hour is six minutes, and a six minute floor applied to a two minute phone call is a rounding convention. A quarter hour floor applied to the same call is something closer to a pricing decision, and the California advisory works the arithmetic out loud. Take a lawyer billing $300 an hour in quarter hour minimums who completes four separate compensable tasks in a day, each taking under three minutes. The client is billed four times 0.25 hours, which is $300, for under twelve minutes of actual work. The effective hourly rate is above $1,500. Nothing has been fabricated. Every entry is defensible on its own terms. The aggregate is not.

The chart below runs the same comparison across four task lengths. The bars show the effective hourly rate the client ends up paying at a nominal $300 rate, depending on whether the firm rounds to a tenth or to a quarter. It is illustrative arithmetic rather than survey data, and its inputs are stated in full so you can run it at your own rate: the actual minutes, the increment, and a $300 nominal rate.

Effective hourly rate on short tasks, at a $300 nominal rate

Three things follow for a small firm. First, the increment has to be in the written fee agreement, because the advisory treats proper disclosure as what justifies charging the minimum at all. Second, the smaller increment is easier to defend and, for practices that generate a lot of short interactions, often produces a bill the client does not argue with, which is worth more than the difference. Third, whatever increment you use, arbitrators look at the frequency with which minimum entries appear and at whether the lawyer showed judgment in not applying the minimum to trivial activities, by not recording them, by aggregating related ones, or by no-charging the entry. A bill consisting almost entirely of minimum increments tells a story about the firm whether or not the firm intended to tell it.

The related trap is billing every task separately. Reading a routine document takes seconds. Recording a separate minimum entry for each of twenty pieces of correspondence read in a single sitting converts two minutes of work into two hours of billing. This is the same mechanism as the increment problem, operating on volume instead of duration, and it is one of the patterns the advisory lists by name.

What you may not bill, even when the work was real

The most useful sentence in Formal Opinion 93-379 is a reframing. Faced with three hard scenarios, the committee said the question should be considered “not from the perspective of what a client could be forced to pay, but rather from the perspective of what the lawyer actually earned.” Everything else in the opinion follows from that single move, and it resolves cases that otherwise feel genuinely ambiguous.

What could be claimed, against what was earned

Take the three scenarios in turn. A lawyer schedules court appearances for three clients on the same morning and spends four hours at the courthouse, the same four hours each client would have cost alone. A lawyer flies six hours for one client and spends five of those hours drafting a motion for another. A lawyer researches a question for one client and is asked the identical question by a second client months later. In each case the tempting bill is the one that treats the coincidence as the lawyer’s good fortune. The opinion does not: a lawyer who spends four hours on behalf of three clients has not earned twelve billable hours, one who flies six hours while working five has not earned eleven, and one who reuses old work product “has not re-earned the hours previously billed and compensated when the work product was first generated.” The benefit of each economy belongs to the client.

The chart makes the arithmetic visible, and the visible version is harder to argue with than the abstract rule. In the first scenario the gap is eight hours the firm did not work. In the second it is five. In the third the entire charge is a repeat of hours already paid for. None of these bills would look strange on its own; each one would look like a busy week. That is precisely why the opinion supplies a test rather than a list, because the situations that produce the problem are ordinary ones and they arrive without announcing themselves.

Two qualifications keep this from being absolute. The opinion is explicit that a lawyer remains free, with full disclosure, to propose additional compensation for a particularly efficient or outstanding result, or for the benefit the client received from reusable work. What it forbids is achieving the same outcome silently by putting hours on an invoice that were not worked. And the California advisory notes that billing the same time to more than one client is not automatically improper where the arrangement is fair, reasonable, fully disclosed in the written fee agreement of every affected client and explained to each of them. Both exits run through disclosure, which is the theme of the whole subject.

One more pattern belongs here because it is peculiar to small firms. The advisory calls it ghost billing: work performed by one person and billed as another’s time, for example a motion drafted by a paralegal and billed at the partner’s rate, or a contract attorney’s appearance billed as the lawyer of record’s own. It is listed alongside padding as a practice that may be deceptive and may amount to an attempt to collect an unconscionable fee. The safe version is straightforward and is the one most firms would choose anyway: bill the person who did the work, at that person’s disclosed rate. Non-lawyer time is billable when the work is substantive and the client agreed to it, a point we cover in the piece on lawyer hourly rates.

Expenses and disbursements: your overhead is already in the rate

The second half of Formal Opinion 93-379 deals with everything on the invoice that is not professional time, and it sorts those charges into three buckets that are worth memorizing because they resolve most of the questions a small firm has.

General overhead is the first. The opinion’s position is that a client who has engaged a lawyer for a fee would be justifiably disturbed to find additional charges for general office overhead on the bill, and that absent advance disclosure to the contrary, the client should reasonably expect the cost of maintaining a library, carrying malpractice insurance, renting office space and buying utilities to be subsumed within the charge for professional services. The California advisory reaches the same place from the other direction and adds the practical explanation: overhead is why a firm in an expensive downtown market generally charges a higher hourly rate than a firm renting cheaper space further out. The rent is already in the rate. Billing it again is billing it twice.

Disbursements are the second. When a firm tells a client it will charge for disbursements, the client should expect to be billed the actual payments the firm made on the client’s behalf: what the court reporting service charged, what the airfare and the hotel cost. The opinion’s view is that a surcharge on those amounts is improper absent disclosure to the contrary, unless the firm genuinely incurred additional expense beyond the direct cost. And if the firm receives a discounted rate from a third party provider, the benefit of the discount belongs to the client rather than being retained as undisclosed margin. Clients, the committee noted, could quite properly view the alternative as creating hidden profit centers.

In-house services are the third and the fuzziest. Photocopying, long distance calls, computer research, special deliveries and secretarial overtime may be recouped, but the charge has to reasonably reflect the firm’s actual cost for the service rendered rather than being priced as a product. Adopted rule text tends to say the same thing with a second permitted route: a lawyer may seek reimbursement for in-house services either by charging an amount the client agreed to in advance or by charging an amount that reasonably reflects the cost incurred. Either way the client has to be able to see the charge and understand what it is for.

The table is worth keeping next to your fee agreement rather than next to your invoices, because three of the four rows turn on what was disclosed before the engagement began. Almost nothing in this area is forbidden outright. What is forbidden is arriving at it silently, on an invoice, after the client has already committed.

Administrative and ministerial work sits just outside this category and is worth naming separately, because it is where small firm bills most often go wrong by accident. Preparing a proof of service, stuffing and stamping an envelope, updating a calendar and reconciling an account are not legal services, and a lawyer’s rate is already understood to include consideration for support staff wages. The California advisory observes that it is not uncommon to see a client billed up to an hour for preparing and serving form interrogatories when the only properly chargeable component was the five to ten minutes the lawyer spent deciding what to check. Running a practice takes dozens of administrative and managerial tasks a day, and in the vast majority of cases it is inappropriate to charge a client for them. That is not a hardship. It is the reason the rate is what it is.

Stage two, the prebill: the one review that decides the invoice

A prebill is a draft bill, read and edited before it is sent. Larger firms have always run this as a formal step owned by the billing partner, and the edits made at that step have a name: billing judgment. Small firms very often skip it, on the reasonable-sounding ground that the person who recorded the time is the person who would review it, so what would the review add.

What it adds is the only opportunity to remove the entries that should not be on the bill. The 4.3 hour block that should have been three separate lines. The hour spent on a task that turned out to be a dead end for reasons the client should not pay for. The minimum increment applied twenty times in one afternoon. The disbursement that was passed on at list price when the firm was invoiced at a discount. Every one of these is cheap to fix in a draft and expensive to fix after a client has read it, because after that point the fix is a credit, and a credit is a conversation about whether the firm’s bills can be trusted.

Edits should be able to run in both directions, but the two directions are not symmetrical. The California advisory says that any evidence a prebill was adjusted upwards should be examined carefully, because it may indicate the client is being charged more time than the lawyer actually expended. Increasing an entry to a contractually agreed minimum is fine. Restoring time the lawyer inadvertently left out is fine. Increasing a charge to improve the look of the month is not, and absent clearly defined and explained terms in the written fee agreement permitting upward adjustment, the increase may itself be evidence of padding. The advisory is equally alert to why the review gets skipped in practice: billing partners are busy, the lawyer who did the work is not available, the stack is high and the firm wants the bills out. Those are the reasons in a large firm. In a two lawyer firm they are the same reasons with fewer people to absorb them.

Read your own bill the way a fee arbitrator would

This is the part of the subject nobody publishes for firm owners, so it is worth setting out properly. Arbitration Advisory 2016-02 exists to tell a neutral how to assess a bill when a client alleges it is padded. It lists the patterns that draw scrutiny. Read as a self-audit rather than as an accusation, it is the most efficient billing review a small firm can run, and it takes about twenty minutes on a single month’s invoices.

Block billing deserves the extra line because the number attached to it travels. In its earlier 2003 advisory the same committee opined that block billing “hides accountability and may increase time by 10% to 30%,” and the 2016 version reprints that language while making clear that an arbitrator’s discretion is not confined to that range. Courts have said the same in harder words. California appellate authority holds that trial courts retain discretion to penalize block billing when the practice prevents them from discerning which tasks are compensable and which are not, and that where counsel cannot further define the entries the court may assign a reasonable percentage to them or cast them aside. None of that makes block billing a rules violation in itself. What it makes it is a bill that cannot be defended line by line, which amounts to the same thing at the moment it matters.

Stages three and four, rendering and collecting: the invoice is not the end of the work

What has to appear on the bill itself is more prescribed than most owners expect, and in some states it is set by statute rather than by the conduct rules alone. The conduct rule sets the floor everywhere: the scope of the representation and the basis or rate of the fee and expenses must be communicated to the client, preferably in writing, before or within a reasonable time after the representation begins, any change to that basis or rate must also be communicated, and on the client’s request in any matter the lawyer must communicate the basis or rate in writing.

California goes further and legislates the bill. Business and Professions Code section 6148 requires that all bills rendered by an attorney to a client clearly state their basis; that the fee portion include the amount, rate, basis for calculation or other method of determination of the fees and costs; and that the cost and expense portion clearly identify the costs and expenses incurred and their amounts. It also puts a clock on the firm: on a client’s request, a bill must be provided within 10 days, unless one was provided in the previous 31 days, in which case it is due within 31 days of the last one. Failure to comply with the section renders the fee agreement voidable at the client’s option, leaving the attorney entitled to collect a reasonable fee. Whether or not you practice in California, that statute is a serviceable specification for what a bill should contain, and a firm whose invoices would satisfy it will rarely be asked to explain them.

Two details in that list are easy to miss and both are about timing rather than content. The first is that a bill has to be producible on demand, not merely at month end, which is a systems requirement as much as a compliance one: a firm that can only generate an invoice by running a full billing cycle cannot meet a ten day request without disrupting everything else. The second is that the consequence of getting the paperwork wrong is not a fine but a voidable agreement, which converts a clerical failure into a fee dispute the firm starts from behind. Both are arguments for keeping the billing system simple enough to run at any point in the month.

Then the bill has to turn into money, and this is the stage most firms treat as somebody else’s department. It is not a small problem. In a survey of more than 2,800 legal professionals published in March 2025, primarily solo and small firm practitioners, 68 percent described fee collection as a significant or moderate challenge, 57 percent reported difficulty collecting reimbursable expenses and 55 percent named expense tracking as a challenge. The same survey found 82 percent of firms now accept credit or debit card payments, up from 78 percent in 2023, and 59 percent of firms using payment technology reported improved collection rates alongside better client satisfaction.

Getting paid, as reported by firms in 2025

Two things in that picture are worth separating. Accepting cards is a payment-channel decision, and the profession has largely made it. Collecting is a process decision, and the same survey suggests the profession has largely not. A firm that bills on the last day of the month, sends nothing else until the next bill, and starts asking questions at ninety days has no collection process; it has a hope. A firm that bills on a fixed date, applies any retainer or advance fee deposit against the invoice on a stated schedule, sends a short reminder at fifteen days, and makes one call at thirty, has one. Neither firm is working harder. One of them has decided in advance what happens, and the other is deciding each time.

Two mechanical cautions on the way through. Applying client funds held in trust to an invoice is a trust accounting event with its own rules about notice and timing, covered in our piece on lawyer trust accounting, and it is not made compliant by the fact that the money was always going to be yours. And card surcharges, where they are used at all, have to be handled so that no processing fee is ever taken out of trust principal. Your practice management and billing software should be configured for this at setup rather than reconciled into shape later.

What to fix first, in the order that pays

If the four stages are all imperfect, which most are, the sequence that returns the most for the least effort is the order they happen in, because each stage inherits the errors of the one before it. Nothing downstream can repair an entry that was never recorded.

Start by closing the gap between doing the work and recording it, and measure the gap rather than estimating it: look at when last month’s entries were actually created. Then institute the prebill review as a scheduled event with a named owner, even in a firm of one, and give yourself permission to strike entries rather than only to move them. Then run the arbitrator’s checklist above across one month of finished invoices and fix what it finds, which will usually be two or three recurring patterns rather than anything alarming. Only then change the increment, the rate or the software, because those are the changes everyone reaches for first and the ones that do the least on their own.

The through line in every authority quoted here is the same and it is not really about money. A bill is a communication, and the conduct rules treat it that way: the duty is to explain the basis of the charge at the outset and then to render statements that let the client see how that basis was applied. A firm that bills so that a stranger could follow the reasoning has already satisfied most of what the rules require, and has also, not coincidentally, built the bill that gets paid without a phone call.