The Outcome Definition Clause
When a vendor charges per outcome, the definition of the outcome is what decides your bill, so that definition must be agreed in the contract before you sign. A loose definition of resolved turns deflections and reopens into billable events you never wanted to pay for.
Key takeaways
- Under outcome based pricing you pay per result, so the contractual definition of that result controls your invoice.
- Zendesk pioneered outcome pricing per automated resolution, and the definition of resolved must be agreed contractually before signing.
- Exclude reopens, deflections, and false positives, and set who measures the outcome and how disputes are handled.
- Pair the definition with a unit rate lock and a volume cap so neither the price nor the count drifts over the term.
What is an outcome definition clause?
An outcome definition clause states, in the contract, exactly what counts as the billable outcome under an outcome based pricing model. Outcome pricing charges you for a result rather than a seat or a unit of consumption, for example a resolved support ticket handled by an AI agent. Because the result is the thing you pay for, the precise meaning of that result is the single most important commercial term in the deal. The clause is where you pin it down, so that the vendor's meter and your understanding of value describe the same event.
This matters because outcome pricing is spreading. Zendesk pioneered outcome pricing per automated resolution, and the pattern is being adopted across customer service and agent driven products as AI takes on more of the work. Pricing is shifting from seats toward usage, agent, and outcome meters, and outcome is the meter where definition does the most work. Without a tight clause, you are paying against the vendor's internal measurement of success rather than a standard you agreed.
Why does the definition of resolved matter so much?
Because a loose definition of resolved lets events that were not genuinely solved count as billable outcomes. Consider the difference between a ticket the agent actually answered to the customer's satisfaction and a ticket that was merely closed, deflected to a help article, or reopened a day later. If the contract counts all of those as a resolution, you pay for outcomes you did not receive, and the bill rises with the vendor's measurement choices rather than with the value delivered. The looser the word, the more your invoice reflects activity rather than results.
Name the mechanic plainly without accusing anyone of bad faith: when the vendor both performs the outcome and defines it, the incentive runs toward a generous count. That is a structural feature of outcome pricing, not a moral failing, and the remedy is structural too. You agree the definition in writing before signing, so that what counts as resolved is a shared standard you can both measure against, not a number the vendor reports and you accept.
Under outcome pricing the vendor performs the outcome and, without a clause, also defines it. The definition is where you take that back.
How do you draft the clause?
Draft it to define the qualifying outcome, exclude what should not count, and fix who measures it. Start with a positive definition: state the conditions a billable outcome must meet, such as a genuine resolution accepted by the customer or one that goes a defined period without reopening. Then add the exclusions explicitly, because the exclusions are where the money sits. The clause should make clear what does not count, so the meter cannot quietly absorb the grey area.
Specify the measurement and the dispute path as well. State who measures the outcomes, what reporting you receive, how often you can audit the counts, and what happens when you disagree. An outcome you cannot inspect is an outcome you cannot challenge, so the right to see the underlying data and to dispute miscounts belongs in the clause. Read more on bounding the variable side in our note on consumption caps and overage protection, which applies the same discipline to usage meters.
| Clause element | What it controls | Buyer position |
|---|---|---|
| Qualifying outcome | What counts as a billable result | A genuine resolution, defined positively and tightly |
| Exclusions | Reopens, deflections, false positives | Listed explicitly so they cannot be billed |
| Measurement and audit | Who counts and what you can inspect | Your right to the data and to dispute counts |
| Unit rate and volume cap | Price per outcome and total exposure | Locked rate, ceiling on billable volume |
How do you cap the exposure on outcome pricing?
Cap it with a locked unit rate and a ceiling on billable volume, because a per outcome meter is open ended by default. The definition controls what counts; the rate and the cap control how much it costs. Lock the price per qualifying outcome for the term so the unit cost cannot drift upward, and set a volume ceiling or a budget cap so the total exposure is bounded even if outcomes spike. Add notice triggers as you approach the ceiling, so a surge in volume is something you see and manage rather than discover on the invoice.
Treat this as part of the same discipline you apply to any variable pricing. Outcome pricing is a meter, and meters need both a definition and a boundary. The clause that defines resolved and the clause that caps the rate and the count work together: one stops you paying for the wrong events, the other stops you paying without limit for the right ones.
What evidence should you demand before agreeing?
Demand evidence that the outcome is real and the pricing reflects value before you sign, the same way you would challenge any AI premium. Outcome pricing usually arrives attached to an AI agent, and the buyer defense applies: ask for proof of value on your own data, run a scoped pilot, and demand ROI evidence before accepting the premium the outcome model carries. Across the market, AI driven asks run 20 to 37 percent against a historical 3 to 9 percent annual uplift, and disciplined negotiation cuts those asks by roughly 55 percent, landing the average uplift near 12 percent. Outcome pricing is not exempt from that scrutiny.
The pilot also calibrates the definition. Running the agent against your real ticket flow shows you how many outcomes it genuinely resolves versus how many it merely closes, which is exactly the information you need to draft the qualifying conditions and the exclusions with confidence. Negotiate the clause from that evidence, not from the vendor's marketing definition of success.
What are the common mistakes with outcome pricing?
The first mistake is accepting the vendor's definition of the outcome without testing it, which hands the vendor both the performance and the measurement. The second is focusing on the headline price per outcome while leaving the definition loose, so a low unit rate applied to an inflated count costs more than a fair rate on a tight count. The third is signing without a volume cap, which turns a model that sounds aligned to value into an open ended commitment that spikes when usage does. Each feels reasonable at signing and turns expensive once the meter runs.
The disciplined approach inverts all three. Define the outcome yourself from pilot evidence, negotiate the definition and the exclusions before the rate, and cap both the unit price and the total volume so neither can drift. Outcome pricing can genuinely align cost to value, but only when the buyer controls the definition rather than inheriting it. Documented in the order form rather than promised in a deck, the clause is what makes the alignment real instead of rhetorical.
Your next step
Outcome pricing rewards the buyer who defines the outcome before the vendor does. For the full clause set, read the SaaS Contract Terms Guide. To understand the model and how to bound it, see Outcome Based Pricing: The Buyer's View and Consumption Caps and Overage Protection. When you want an outcome clause drafted and won on a live deal, our buyer side team can take the table or coach yours through it.
Common questions
What is an outcome definition clause?
Why does the definition of resolved matter so much?
How do you negotiate an outcome definition clause?
Last reviewed March 2026. Market figures cited are published industry data; figures labelled indicative are directional.