Alternative Fee Arrangements, Explained Without the Hype

Most alternative fee conversations start at the wrong end. A partner reads that clients want predictability, announces that the firm will offer fixed pricing on a practice area, and picks numbers by asking three people what feels about right. Six months later the firm is busy, the clients are pleased, and the effective hourly rate on that work has quietly fallen below what the firm pays to produce it. Nobody notices, because the invoices all went out and all got paid. That is the failure mode. Alternative fees do not blow up loudly, they erode.

An alternative fee arrangement is not a discount, a marketing feature, or a modern attitude to billing. It is a transfer of risk from one side of the table to the other, priced. Every structure in the category moves a specific, nameable risk. Once you can say out loud which risk you are taking on and what you are being paid to take it, the choice between a flat fee and a cap and a collar stops being a matter of taste and starts being arithmetic.

What follows is each of the main structures, what it actually shifts, which matter types it fits, where it fails, and the data a firm needs in hand before quoting one. None of this requires a pricing consultant. It requires knowing your own history, which is where most firms are genuinely stuck.

What an Alternative Fee Arrangement Actually Moves

There are only two variables under any legal fee. The first is effort, meaning how much work the matter turns out to require. The second is outcome, meaning whether the work produces the result the client wanted. Straight hourly billing leaves both of those risks entirely with the client. The client pays for whatever effort occurs, whether the matter settles in a week or grinds for two years, and pays the same whether they win or lose. The firm carries almost nothing except the risk of writing off or failing to collect.

Every alternative structure takes one or both of those variables and moves them across the table. A flat fee moves effort risk to the firm and leaves outcome risk with the client. A capped fee moves the top slice of effort risk. A collar splits effort risk in an agreed proportion. A contingency or success fee moves both effort and outcome onto the firm, and prices that with a share of the recovery. A blended rate does something quieter and often misunderstood, which is discussed further down. Before you argue about the number, say which variable you are moving and in whose favour. Firms that skip that step end up defending a price they cannot explain.

Flat Fees Buy Certainty With Scope, Not With Outcome

A flat fee is a fixed price for a defined scope of work. The scope is the product, not the number. Firms get hurt here because they write the fee with great precision and the scope with none, so the engagement says two thousand for a contract review and says nothing about how many rounds of revision, how many counterparties, whether a redline war is included, or what happens if the deal restructures halfway through. When the matter turns ugly the firm has no contractual place to stand, because the only thing it defined clearly was its own obligation to keep working.

Flat pricing suits work where the distribution of effort is tight and repeatable. Company formations, straightforward trademark filings, standard estate planning packages, residential property transactions, uncontested filings, template employment documentation, fixed-scope compliance reviews. The test is not the average number of hours the work takes, it is the shape of the tail. If one difficult opposing party, one uncooperative registry, or one client who cannot make a decision can multiply the effort by four, the average is lying to you. Price against the tail or exclude the tail explicitly, and write the exclusion into the engagement letter in plain language the client will still understand nine months later.

Capped Fees Are a Flat Fee Wearing a Discount

A capped fee bills hourly, with a promise that the client will never pay more than an agreed ceiling. Firms reach for caps because they feel like a smaller commitment than a flat fee. They are not. Above the ceiling the firm is doing free work exactly as it would under a fixed price, and below the ceiling the firm collects less than the flat fee it could have charged. A cap is a flat fee that has given away its upside and kept all of its downside. That can still be the right call, but you should know that is the trade you are making.

Caps make sense when you have genuine hourly history for the matter type and you can set the ceiling comfortably above the eightieth percentile of that history rather than near the average. They also fit situations where the client needs a worst case number for internal approval more than they need a single price, which is common with in-house teams working to a budget. The critical discipline is that everyone keeps recording time after the cap is reached. The moment your people stop entering hours because the hours no longer generate revenue, you have destroyed the only dataset that would have told you the cap was set too low.

Collars Split the Miss in Both Directions

A collar starts with an estimate and puts a band around it. Inside the band the client pays the estimate. Outside the band, in either direction, the parties split the difference at an agreed percentage. If the matter comes in under, the client keeps most of the saving and the firm keeps a slice as a reward for efficiency. If it runs over, the client absorbs part of the overrun and the firm absorbs the rest. It is the only common structure that pays a firm for finishing early, which makes it the honest answer for long matters where both sides know the estimate will be wrong and neither can say in which direction.

Collars demand two things most firms are not ready for. The first is a defensible estimate, which means real history rather than a partner's instinct. The second is visibility, because a collar only survives if the client can watch effort accumulate against the estimate as the matter runs instead of receiving one shocking reconciliation at the end. This is where a live client portal stops being a convenience and becomes structural. Casely's portal updates in real time and filters privileged material automatically per document, so a client can follow the matter without anyone on your side assembling a status pack, and a matter stage tracker gives them a clickable view of where the work actually sits.

Success Fees Are a Portfolio Decision, Not a Matter Decision

A contingency or success fee takes both variables at once. The firm funds the effort and only gets paid if the outcome arrives. Any single matter priced this way is a bet. Forty matters priced this way, selected with discipline against a known win rate, is a business. The distinction matters enormously for small firms, because a firm without enough volume to average out its outcomes is not running a contingency practice, it is gambling with its own payroll. If you cannot describe your intake screening criteria in a sentence, you are not ready to take outcome risk.

This is also the structure with the widest jurisdictional variation, so treat any general statement about it with suspicion, including this one. Contingency arrangements are standard in United States plaintiff-side work but are commonly prohibited or restricted in criminal and domestic relations matters, and the permitted percentages and disclosure requirements differ state by state. England and Wales use conditional fee agreements and damages-based agreements, which carry their own caps, formalities and costs consequences and do not map cleanly onto the American model. Canadian provinces and Australian jurisdictions each regulate uplifts and percentage fees under their own professional rules. Confirm what is permitted where you practise, in writing, before you offer anything of this shape.

Blended Rates Are a Staffing Commitment in Disguise

A blended rate charges one hourly figure for everyone who touches the matter, regardless of seniority. Clients like it because it removes the suspicion that partners are padding matters with expensive juniors or that juniors are being learned on at partner prices. It is easy to sell and easy to explain, which is exactly why firms agree to it without doing the underlying calculation.

The calculation is this. A blended rate is only profitable if the real staffing mix on the matter is more junior-weighted than the mix implied by the blend. Quote a blend that assumes a partner does one hour in five, then staff the matter with a partner doing one hour in two, and you have given away the difference on every single hour. So a blended rate is not a pricing decision at all, it is a staffing commitment you have made in advance and must then honour operationally. Before quoting one, look at the actual timekeeper mix on your last twenty comparable matters rather than the mix your org chart implies, and be honest about the matters where partner attention is the thing the client is genuinely buying. Those matters should never be blended.

Variance, Not Value, Decides Which Structure Fits

The instinct is to match structure to matter size, with small routine work going flat and large complex work staying hourly. That instinct is wrong often enough to be expensive. The deciding factor is variance. Work with a tight, well understood effort distribution can be priced flat at almost any size, and a large corporate filing programme can be more predictable than a small landlord dispute where the other side is a private individual with time on their hands and a grievance.

So sort your work by how well you can predict the hours, not by how much it bills. Low variance and repeatable goes flat. Moderate variance with a distribution you can actually see goes to a cap or a collar. High variance attached to a monetary outcome you can influence is contingency territory, if your jurisdiction permits it and your volume supports it. High variance with no monetary outcome, which describes a great deal of contentious and regulatory work, should stay hourly, and you should say so plainly rather than inventing a structure to look modern. Refusing to price something is a legitimate answer and clients respect it more than a number you visibly guessed.

Price the Phase, Not the Matter

Large matters are not one pricing decision, they are several. A litigated dispute has a pleadings phase with fairly predictable effort, a discovery or disclosure phase with wide and largely uncontrollable variance, an interlocutory phase that depends entirely on the other side, and a trial phase that is expensive and rare. Applying one structure across all of that guarantees that the structure is wrong for at least three of the four. Firms that price by phase get most of the client's predictability benefit while keeping the firm off the hook for the part of the matter it genuinely cannot forecast.

Phase pricing needs the matter record to hold the phases explicitly rather than in someone's head. A configurable stage tracker, set up per practice area so litigation stages and transactional stages look different, gives you a place to hang the fee terms for each phase and a trigger point where the next fee conversation is supposed to happen. Where a dispute spawns related proceedings, linking them as connected matters with the reason recorded keeps the pricing history intact instead of leaving three separate files that nobody can compare later.

The Four Numbers You Need Before You Quote

The first number is the distribution of hours by matter type, meaning not just the mean but the eightieth and ninetieth percentiles and the worst case you have actually lived through. The second is your real cost per hour by timekeeper, built from salary, benefits, overhead and realistic capacity rather than from the rate card, because the rate card tells you what you hoped to charge and not what the work costs to produce. Without those two numbers, every fixed price you quote is a guess wearing a suit.

The third number is what you have historically collected against what you originally quoted on similar work, which tells you whether your estimates are systematically optimistic. Most are. The fourth is a written definition of what finished means for that matter type, because you cannot price a deliverable you have not defined and half of all fee disputes are really definition disputes arriving late. If your firm cannot produce these four things this week, that is the project, and it comes before any pricing announcement to clients.

Keep Tracking Time After the Fee Stops Depending On It

The most common operational failure in alternative fee work is that time recording collapses the moment the fee stops being calculated from hours. It is an understandable reaction and it is fatal. Under hourly billing, time entries generate the invoice. Under alternative fees, time entries generate the pricing model for next year. If your people stop capturing hours on fixed-fee matters, you lose the ability to tell a profitable flat fee from an unprofitable one, and you will keep repricing on the same bad guess indefinitely.

This is worth stating explicitly to the team, because the usual instruction is that time recording exists to bill clients, and that instruction becomes false the day you go fixed. The framing that works is that hours now measure cost rather than revenue. Practically, it means keeping capture friction low enough that people do it when there is no invoice riding on it, and it means your system needs to hold hourly, flat-fee, contingency and blended matters natively rather than forcing fixed-fee work through an hourly-shaped workaround. Casely handles all four billing types natively and still turns unbilled time into an itemised draft in one click, which is how you keep the internal cost picture even on matters where the client sees only one number.

Trust and Ledger Mechanics Change Under a Fixed Fee

Alternative fees change what the money in your client account represents. Under hourly billing, funds come out of trust as work is performed and invoiced, in small regular movements you can reason about. Under milestone-based fixed pricing, larger amounts move at defined points, and the temptation to draw against a milestone slightly before it is genuinely complete is real and is exactly where regulators find problems. The safeguard that works is structural rather than behavioural. Casely enforces the trust rule at the database transaction level, so a disbursement exceeding a matter's actual trust balance simply cannot be written, and corrections are voided and remain visible rather than being deleted, which means the audit trail survives the mistake.

Institutional clients who ask for alternative fees are usually the same clients running formal e-billing, so the structure and the file format arrive together. Fixed-fee and phase-based work still has to land in a LEDES 1998B export in the shape their system expects, and per-matter isolated ledgers are what make phase-level reconciliation possible without a spreadsheet rebuild at year end. If your billing system was designed only for hourly work, alternative fees will show up first as a formatting problem and only later as a profitability one.

Measure the Effective Rate, Then Reprice Every Year

The only scoreboard that matters is effective rate per matter type per structure, calculated as total fees received divided by total hours actually recorded. Run it annually against every fixed price, cap, collar and blend you offer, and compare it to what the same work would have earned hourly. Some of your alternative fees will be beating your rate card, which is worth knowing and worth expanding. Some will be far below it, and those are usually the ones the partners are proudest of, because the client loves them.

Then run the second number, which is how many of those matters you won specifically because you offered the structure. An arrangement that lowers the effective rate but materially raises conversion may still be the right commercial decision, particularly for a growing firm buying its way into a client relationship. The failure is not offering a fee that earns less per hour, it is offering one without knowing that it does. Reprice once a year, in a scheduled review, using your own closed-matter data rather than what the market is said to be doing.

Pricing Is a Discipline, Not a Pitch

Alternative fee arrangements are worth doing, and the firms that do them badly are not doing them wrong so much as doing them blind. Every structure in this article works somewhere. Flat fees work on tight, repeatable scopes. Caps work when you have real history and set the ceiling high enough. Collars work on long matters where visibility is genuine. Success fees work at portfolio scale in jurisdictions that permit them. Blends work when the staffing plan is honest. What none of them survive is being announced before the underlying numbers exist.

Start with one matter type. Pull its history, look at the distribution rather than the average, cost it properly, write the scope boundary before you write the price, and run it for a year while continuing to record every hour. That single exercise will teach you more about your firm's economics than any pricing seminar, and it will tell you honestly whether the next matter type is ready for the same treatment. If the data is not there, the answer is to build the data, not to quote anyway and hope.

If you are moving toward fixed and phase-based pricing, the systems underneath have to keep up first, which usually means clean time capture that continues after the fee stops depending on it and billing that handles more than one model without workarounds. Our legal billing software supports hourly, flat-fee, contingency and blended arrangements natively, with LEDES 1998B export for institutional clients and per-matter trust ledgers that block an overdraft rather than warning about one. The Free plan costs nothing to start, which is enough to run the exercise on a single practice area and see what your own numbers actually say.