Isn't it riskier for the provider?

It is, which is why it demands closed scope and written acceptance criteria before starting. Without those, outcome pricing is just an expensive way to argue.

Does it work for any engagement?

No. It works when the outcome can be defined and measured. For pure exploration — where nobody yet knows what needs doing — a short bounded phase first, then fixed price.

← Full glossary

Outcome-based pricing

service-model

outcome-driven pricing · fixed-scope pricing · impact-linked fees

A model where the provider charges for work delivered or effect achieved, not for hours spent or per-seat licences. It is becoming the standard in AI services because the cost of repeatable work is collapsing.

Outcome-based pricing breaks the link between what a service costs and how long someone spends on it. It comes in three forms worth keeping apart: fixed price per deliverable (I know what I pay and what I get), price per unit of work processed (per ticket resolved, per invoice extracted, per document reviewed), and a fee tied to a business indicator (the rarest and hardest to attribute). The first works for AI implementation because scope can be closed; the third usually ends in an argument about how much of the result the provider caused. When AI makes repeatable work cost a fraction, charging by the hour forces the provider to be slow to earn the same: the incentive is inverted. That is why the model is spreading from the large consultancies to AI-native law firms.

How it differs from

Hourly billing: By the hour, the provider earns more the longer it takes. By outcome, they earn more the better they solve it.
Per-seat licence: A licence charges for access to software whether used or not. Outcome pricing charges for work done.

Examples

A fixed price to leave a process automated and running, with scope defined up front.
Charging per document reviewed instead of per reviewer hour.
A monthly fee to run the agents in production, independent of hours consumed.

FAQ

Isn't it riskier for the provider?: It is, which is why it demands closed scope and written acceptance criteria before starting. Without those, outcome pricing is just an expensive way to argue.
Does it work for any engagement?: No. It works when the outcome can be defined and measured. For pure exploration — where nobody yet knows what needs doing — a short bounded phase first, then fixed price.

TechCrunch — AI law startup Norm raises $120M · 2026-07-07 · E3

Outcome-based pricing

How it differs from

Examples

FAQ

Related terms