Every price, on one page.
for the audit. After that, each stage has a pricing rule written down before it starts, and digital staff are paid only on results you can verify.
Performance-based pricing ties what you pay to the outcome an AI delivers: a proven sale, a completed task, a resolved ticket. Not a per-seat licence, not a per-token bill. It is the model that suits agents hired to do work rather than software operated by people, and it is the model Quantrim is built on.
The pricing rules.
One rule per stage. The outcome metric for digital staff is fixed in writing before the trial starts, so measurement is never a debate.
- 01
The free audit
Two weeks, three hours of your time, a report you keep. No cost, no obligation, whether or not you hire us.
$0
- 02
Fractional AI leadership
A day rate on a monthly retainer sized to the roadmap, typically two to eight days a month. Scale up, scale down or stop with a month's notice.
Day rate. Monthly retainer. No lock-in.
- 03
The Digital Brain and workflows
Scoped from the audit, priced before it starts, and owned outright by you when it is delivered. Models plug in underneath at their own cost, which you see.
Fixed scope, fixed price.
- 04
Digital staff
A free trial on one scoped workflow. Then payment per outcome: Mary on sales she is proven to generate, Adam per task completed and verified, Eliza monthly against the briefs she delivers.
Paid on results you can verify.
- 05
Training and adoption
Included in every engagement, because tools that people work around return nothing.
Included.
Worked examples.
Illustrations of how each rule plays out, with round numbers chosen for clarity. Your numbers come from the audit.
Mary on cart recovery
Attribution rule agreed up front: a sale counts when it completes from a cart Mary recovered or a lead she nurtured. Free trial for the first scoped workflow. Then a percentage of attributed sales, invoiced monthly against your store's own records. No sale, no fee.
Adam on order entry
A task counts when an order is entered and verified against the ERP, or an exception is resolved within the agreed rules. Free trial on the mapped workflow. Then a fee per verified task, reported weekly. Tasks Adam queues for a person are not charged.
Two days a month
A 150-person firm with a board that wants an answer on AI. Two days a month at the day rate covers the roadmap, the policy, build-versus-buy calls and a quarterly board report. A fraction of an executive salary, and it can stop with notice when a full-time hire makes sense.
Three ways to pay for software.
Software has been sold per seat or per usage. Outcome pricing is the third model, and it moves the risk to where it belongs.
| Seat-based | Usage-based | Outcome-based | |
|---|---|---|---|
| You pay for | Each user licence | Volume consumed | Results delivered |
| Cost predictability | High | Variable | Tied to value |
| Vendor incentive | Sell more seats | Drive more usage | Make you succeed |
| Risk sits with | Buyer | Buyer | Shared, mostly vendor |
| Best for | Team tools | Infrastructure | Autonomous AI agents |
The honest trade-offs
Both sides must agree how the outcome is measured and attributed: what counts as a proven sale, and how much credit the agent gets. Vendor revenue is less predictable, and fair measurement needs clean data. A clear definition of the outcome, fixed before the trial, and transparent tracking against your own records resolve most of it.
Questions.
What counts as a “proven result”?
Whatever outcome metric we agree before the engagement starts: a completed sale attributed to the agent, a ticket resolved without human help, a task completed and verified. The definition is fixed up front so measurement is never a debate.
Is there any minimum commitment?
The engagement starts with a free trial on a scoped workflow. There is no payment until the agent performs against the agreed metric.
Does Quantrim charge per seat?
No. Quantrim uses a free trial followed by performance-based payment on proven results, so cost is tied to value delivered.
Who pays for the models underneath?
Model costs are passed through at cost and shown on the invoice. Routine work runs on cheaper or local models; frontier models are used only where the task needs them. The Digital Brain is decoupled so you can swap providers when prices move.
What is the drawback of outcome pricing for us?
You need clean enough data to measure the outcome fairly, and you need to agree an attribution rule. The audit checks the first; the second is a one-page agreement before the trial.
Can we get a quote before the audit?
For leadership, yes: the day rate is fixed. For the Digital Brain and digital staff, the honest answer is no, because the scope comes from the audit. That is why the audit is free.
What is performance-based AI pricing?
A model where you pay for the outcomes an AI delivers (such as a proven sale or a resolved ticket) instead of a fixed per-seat or per-usage fee.
How is outcome-based pricing different from usage-based pricing?
Usage-based pricing charges for volume consumed, such as API calls or tokens, regardless of result. Outcome-based pricing charges only when an agreed result is achieved.
What are the benefits for buyers?
Lower adoption risk (you pay when it works), cost that tracks value, and a vendor whose incentives are aligned with your success.
What are the drawbacks of performance-based pricing?
Both sides must agree how the outcome is measured and attributed, vendor revenue is less predictable, and fair measurement needs clean data.