Legal Spend
Do alternative fee arrangements actually save money vs the billable hour?
Updated July 2026
Sometimes, but never automatically. An alternative fee arrangement saves money only when you price it against real historical cost and cycle time for that matter type. Without that data, a flat fee, fixed fee per phase, blended rate, or success fee is a negotiating guess that can overpay the firm or underpay it. The savings come from the data, not the structure itself.
What actually determines whether an AFA saves money?
The quality of the data you price it against. An AFA is a bet on scope: the firm estimates the work, you estimate what it should cost, and the fee locks in. When you know the real historical cost, cycle time, and outcome for that matter type, both sides price from evidence. When you do not, the fee is a guess and someone loses.
This is why the same fee structure can be a win for one legal department and a loss for another. The structure does not create the savings. The accuracy of the price does, and the price is only as good as the historical cost and cycle-time data you set it against.
What are the main alternative fee structures?
Four common structures, each shifting risk differently. A flat fee covers a whole matter for one price. A fixed fee per phase prices discovery, motions, and trial separately. A blended rate charges one hourly rate regardless of who staffs the work. A success fee ties part of the payment to a defined outcome. Each needs different data to price safely.
| Structure | How it shifts risk | Data you need to price it |
|---|---|---|
| Flat fee per matter | Firm absorbs the risk of a matter running long | Full cost distribution across many closed files, not just the average |
| Fixed fee per phase | Risk is fixed only on the phases you can predict | Cost and cycle time by phase for the matter type |
| Blended rate | Removes the incentive to overstaff with senior time | Historical staffing mix and effective hourly rate by matter type |
| Success fee | Ties part of the pay to a defined result | Outcome distribution and a clear, measurable definition of success |
Why does the billable hour still win without data?
Because the billable hour is honest about uncertainty: you pay for the work that happens, no more and no less. An AFA only beats it when you can predict the work well enough to price the fee. Without historical cost and cycle-time data, hourly billing at least keeps you from locking in a bad estimate, even though it gives the firm no reason to be efficient.
- Hourly billing exposes you to inefficiency but never to a locked-in overpayment.
- An underpriced AFA pushes the firm to cut effort on the files it is losing money on.
- An overpriced AFA quietly overpays on every easy file for the life of the agreement.
- Only real cost data tells you which risk you are actually taking on.
How do you build the data to price an AFA well?
Start by categorizing your matters by type, then total the real cost, cycle time, and outcome for each category across closed files. Once you can see what a routine auto file or an employment matter actually costs and how long it runs, you can price a fixed or blended fee that sits close to reality. The structure follows the evidence, not the other way around.
- Fix a small set of matter-type categories and code every matter to one at intake.
- Total real cost, cycle time, and outcome for each category across your closed files.
- Look at the spread within each category, not just the average, so the fee survives a hard file.
- Price the structure that fits the data, and start with the category where your history is deepest.
Common questions
Do AFAs always save money compared to hourly billing?
No. An alternative fee arrangement saves money only when it is priced against real cost and outcome data for the matter type. Priced well, a flat or blended fee removes the incentive to run up hours and gives you cost certainty. Priced badly, it locks in an overpayment or pushes the firm to cut corners on files it is losing money on. The structure itself is neutral. What determines whether it saves money is whether you knew the true cost and cycle time before you agreed to the number. Without that, an AFA is a negotiating guess that can move the cost in either direction.
How to reduce outside counsel spend→What data do we need before proposing an AFA?
At minimum, the historical cost, cycle time, and outcome distribution for the matter type you want to price, drawn from a meaningful number of closed files. You need the spread, not just the average, because a fee set at the median fails the moment a hard file arrives. You also need to know how the firm staffs that work and where its costs concentrate. Without this, you and the firm are both estimating in the dark, and the fee reflects negotiating leverage rather than the real economics of the work. The more closed-file data you can bring, the closer the fee lands to reality and the smaller the risk to either side.
Outside counsel spend by matter type→Which AFA structure is safest to start with?
A fixed fee per phase is usually the most forgiving place to start, because it lets you fix a price only on the stages you can predict and leave the uncertain ones flexible. Discovery and routine motion practice on a high-volume matter type are often predictable enough to price. Trial rarely is. A flat fee for a whole matter demands the most confidence in your data, and a success fee demands a clearly defined, measurable outcome. Start where your data is strongest, usually the early phases of your most routine matters, and extend only as your cost history proves out the pricing.
When to use an AFA on litigation→Do AFAs hurt quality by capping what the firm will spend?
Only if you set the fee too low or ignore how the work is going. A well-priced AFA aligns the firm's incentive with an efficient resolution without starving the file, because the fee was set to cover the real work the matter type requires. The risk appears when a fee is set below true cost: the firm then loses money on hard files and may cut effort where the case needs it most. That is a pricing failure, not a flaw in the model. Pairing an AFA with outcome and cycle-time measurement lets you catch a file where quality is slipping before it becomes a bad result.
How to measure defense attorney performance→CaseGlide is the litigation intelligence platform for Fortune 500 legal departments and insurance claims organizations. It structures live litigation data from defense counsel into executive decisions: reducing defense spend, settling the right cases sooner, and shrinking litigated claim volume.
Next step · See it on your docket
See what your litigation portfolio is telling you
A 30 minute walkthrough on your own docket. No slides, no committee.