Alternative fee arrangements are common on paper and inconsistent in practice.
A 2025 Best Law Firms survey of 4,852 US firms found that 72% offer some form of alternative fee arrangement. Among firms with more than 50 lawyers, the figure rises to 90%. Flat fees, retainers, and contingency arrangements are the most common structures.1
The demand is also clear. Clio reports that 71% of clients prefer a fixed or flat fee.2
The hard part is not convincing firms and clients that alternatives should exist. It is making the arrangement survive contact with real work.
When scope expands, exceptions multiply, or the delivery team cannot see whether the matter is profitable, both sides fall back to the clock. That is not only a pricing problem. It is a data problem.
The firms that make alternative pricing work connect the fee to a measurement layer. They use historical matter data to set the price, live delivery data to manage the work, and post-matter analysis to improve the next proposal.
An AFA is a delivery and data model
Firms often treat alternative fees as a number placed on top of the existing hourly process. The work is scoped loosely, staffed the same way, tracked primarily in hours, and reviewed after the matter closes.
That approach removes the familiar billing mechanism without replacing the management system behind it.
A durable AFA defines:
- the service the client is buying
- the events included in the scope
- the assumptions behind the price
- the team and workflow used to deliver it
- the data needed to monitor cost and progress
- the baseline used to evaluate performance
- the triggers for a change in scope or fee
- the client conversation when a trigger occurs
The price matters. The system around the price is what makes it work. Data analytics gives that system a way to learn.
Fixed pricing without scope control is not certainty. It is unpriced risk.
Build the baseline before quoting
Before proposing a fixed fee, estimate the expected cost to deliver the work. Start with comparable matters, not the firm-wide average.
Group historical matters by the variables that actually change the work: matter type, phase, jurisdiction, document volume, counterparty count, staffing mix, duration, and exception path. This turns a pile of time entries into a useful pricing cohort.
Historical time can be useful, but it is only one input. The model should also include write-offs, rework, third-party costs, collection timing, and the workflow the firm plans to use now.
Start with:
- lawyer and staff time by role
- expected use of approved technology
- review and quality-control time
- third-party costs
- likely exception paths
- an explicit contingency for normal variation
Then compare the expected delivery cost with the proposed fee. Use a range rather than a single optimistic estimate. The spread between similar past matters is often more useful than the average because it shows how much uncertainty the firm is accepting.
For a simple illustration, assume a repeatable service is priced at $6,000. The expected cost of partner, associate, staff, and technology time is $3,600. The expected contribution before shared overhead is $2,400.
If a routine exception adds $1,200 in delivery cost, the matter still has room. If an unmanaged exception adds $3,000, the economics have changed. The firm needs to know which kind of event it is dealing with before it sends the quote.
The useful number is not the discount from standard rates. It is the expected contribution margin, the range around it, and the scope variables most likely to move the result.
Choose the right structure for the uncertainty
Not every matter belongs under a pure fixed fee. The fee structure should match what the firm can know at the start.
Fixed fee
Best for repeatable work with stable inputs, a defined outcome, and enough history to estimate delivery cost.
Phased fee
Useful when the matter has clear stages but later work depends on what happens earlier. Each phase can be scoped and priced when the relevant facts are known.
Fee cap
Useful when the client wants a maximum exposure but the work is still managed largely by time. The firm needs an approval path before the cap is consumed.
Collar
Useful when both sides want predictability but acknowledge a reasonable range. The parties agree how to share the difference if actual effort falls below or rises above the target band.
Retainer or subscription
Useful for a defined portfolio of recurring services, response times, and capacity. It needs usage boundaries and a process for work that exceeds the agreed service level.
Success or contingency component
Useful when the parties can define an outcome and the rules permit the structure. The success measure must be specific enough that both sides calculate it the same way.
Analytics helps make that choice. A narrow cost distribution can support a fixed fee. Clear stage-level variance can point to a phased fee. A broad but measurable range may fit a cap or collar. The model should follow the pattern in the data, not the preference of the pricing committee.
The best structure is not the one that sounds most innovative. It is the one that makes the risk visible and assignable.
Turn scope into measurable fields
The most important part of a fixed-fee proposal is often the definition of done. The most important part of the analytics model is making that definition measurable.
A useful scope has four parts.
Included work
Name the deliverables, number of documents or entities, expected meetings, revision rounds, jurisdictions, and time period.
Client dependencies
State what the client must provide, in what format, and by when. Delayed or incomplete inputs can change the cost of delivery.
Assumptions
Document the conditions behind the price. Examples include expected document volume, availability of approved precedent, absence of contested proceedings, and a defined number of counterparties.
Change triggers
Define the events that require a new estimate, a new phase, or a change order. Do not wait until the matter is over to decide that it was out of scope.
The trigger should appear in the matter workflow as structured data. When a new jurisdiction, extra revision round, unexpected filing, or material volume increase occurs, the system should record the event and prompt the responsible lawyer to pause and have the client conversation.
Free-form notes are useful for context, but they are difficult to aggregate. A small set of consistent matter fields lets the firm compare estimates with outcomes and see which assumptions repeatedly break.
Use leading indicators during the matter
Many firms review AFA performance at the end, when there is nothing left to manage.
The matter team needs a simple live view:
- fee agreed
- delivery cost to date
- work completed
- forecast cost to finish
- scope changes pending
- margin range
- client approvals outstanding
The dashboard should emphasize signals the team can act on. Cost to date is useful, but cost to complete is more useful. Total scope changes matter, but unresolved scope changes are what put the fee at risk. A simple variance threshold can flag a matter before the margin disappears.
Time capture can still be useful under an AFA. It provides cost and capacity data, even when the client is not billed by the hour. The difference is that hours become an internal management input rather than the product sold to the client.
That shift also makes AI measurement more honest. If an AI-assisted workflow reduces drafting time but increases review time, the firm can see the net delivery cost. If the workflow improves speed and margin without increasing corrections, the firm can price the service with better information next time.
The point is not to build the largest dashboard. It is to give the matter owner a short list of decisions: continue, reallocate, clarify scope, or return to the client.
Connect the data before adding more reports
A pricing model cannot improve if the estimate, matter, time, billing, and outcome data cannot be joined.
Use a common matter identifier across the systems involved. Define a small metric dictionary so finance, pricing, and matter teams calculate the same concepts in the same way. At minimum, agree on:
- expected delivery cost
- actual delivery cost
- forecast cost to complete
- scope variance
- write-off and adjustment treatment
- contribution margin
- cycle time
- rework or quality exceptions
Start with a reliable dataset that answers a few commercial questions. A trusted weekly view is more valuable than a real-time dashboard built on inconsistent definitions.
Access also matters. Pricing data can expose client terms, lawyer performance, and matter economics. Give each team the information needed for its role, and keep the underlying definitions governed.
Give the team a playbook
Pricing committees can design a strong AFA that fails at the matter level because the team does not know how to operate it.
The playbook should answer:
- Who owns scope?
- Who can approve an exception?
- What must be recorded in the matter system?
- When does the client need to be notified?
- Which work is standardized or automated?
- Which work requires partner review?
- How often is the forecast updated?
- Who reviews the result after closing?
The client-facing language matters too. A fixed fee should be positioned as cost certainty and shared discipline, not an automatic discount. The firm is taking delivery risk. In return, the client commits to a defined scope and decision process. Outside counsel guidelines should make the fee structure, staffing expectations, reporting, and exception process explicit.3
Close the feedback loop
Every completed AFA should improve the next one.
At matter close, compare:
- estimated and actual delivery cost
- estimated and actual duration
- expected and actual exception paths
- review and rework time
- client response and collection time
- the contribution margin
Analyze the result by cohort, not only matter by matter. Look for recurring drivers of variance. One practice may consistently underestimate document volume. Another may lose margin during a specific phase. A third may have a pricing problem that is actually a collection problem.
Then update the template, workflow, and price. A pricing model becomes reliable when the data behind it compounds.
This is one reason firms struggle when pricing data lives in spreadsheets disconnected from matter management and billing. The estimate, the work, and the result need to share a common matter identifier so the firm can learn from what happened.
A 60-day analytics-led pilot
Weeks 1 and 2: select the service
Choose one repeatable matter type. Build a small historical cohort and identify normal variation, missing data, and the variables that explain cost.
Weeks 3 and 4: design the model
Define scope, assumptions, delivery workflow, cost model, fee structure, change triggers, and the metrics used to judge the pilot.
Weeks 5 and 6: configure the workflow
Add only the matter fields, alerts, dashboard views, approvals, and pre-bill checks needed to run the model. Train the team on the client conversation and on consistent data capture.
Weeks 7 and 8: run and review
Pilot the arrangement, monitor the forecast, capture exceptions, and compare the result with the baseline. Adjust the model with real delivery data.
Alternative pricing sticks when the firm can see the work clearly enough to manage it. The commercial promise, operating workflow, and analytics layer have to be designed together.
Ready to make fixed fees operational?
Jinka helps law firms connect pricing models to matter workflows, governed delivery data, change control, dashboards, and AI-assisted operations.