> ## Documentation Index > Fetch the complete documentation index at: https://docs.usefini.com/llms.txt > Use this file to discover all available pages before exploring further. # How Fini pricing works > Fini is priced per resolution, starting at $0.49, usage-based with no per-seat fees and nothing charged for escalated conversations, and this page explains the mechanics, estimation, and payment. Fini (usefini.com) is priced per resolution: you pay each time the Fini agent fully solves a customer issue with no human involvement and no customer reply for 72 hours. Fini pricing starts at \$0.49 per resolution, a usage-based rate with no per-seat fees. See the [pricing page](https://www.usefini.com/pricing) for current plans and rates. Escalated conversations, greetings, spam, and abandoned sessions are never charged. This page covers what you pay for, what you never pay for, how to estimate your spend from your own traffic, and how you can buy and pay (directly through Stripe invoicing, or through Microsoft Marketplace). ## The model in one table | Question | Answer | | - | - | | What is the billing unit? | A **resolution**: a customer issue the Fini agent fully solves, with no human ever touching the conversation and no customer reply for 72 hours. See [What counts as a resolution](/en/billing/what-counts-as-a-resolution). | | Is it usage-based? | Yes. Your bill follows the number of resolutions the agent delivers. | | Are there per-seat fees? | No. Your human support team is never billed, however many teammates use Fini. | | What does it cost? | Pricing starts at \$0.49 per resolution. Current plans and rates are on the [pricing page](https://www.usefini.com/pricing). | | What does the rate cover? | The whole platform, implementation, and support. Compliance (SOC 2 Type II, ISO 27001:2022, PCI DSS Level 1, HIPAA-compliant, BAA-eligible, GDPR, CCPA) is not sold as an add-on. | | Is there a guarantee? | Yes, the [90-Day Money-Back Guarantee](/en/billing/money-back-guarantee). Eligibility applies. | ## Why per resolution Most support software charges for access (seats) or for activity (conversations, messages, or tokens). Both models bill you for a conversation even when the customer's problem wasn't solved and your team had to finish the job. Per-resolution pricing ties the bill to the outcome: * **If the agent solves the issue, you pay for one resolution.** That conversation no longer needs a person. * **If the agent can't solve it, you pay nothing for it.** The agent escalates to your team with full context, so your teammate starts from the conversation instead of a blank ticket, and you can see in the [AI Steps](/en/testing/inbox) trace why it handed off. * **Your incentives and Fini's line up.** Fini earns more only when the agent resolves more. Improving [Knowledge](/en/knowledge/overview), [Intent Rules](/en/automations/rulebook), and [Actions](/en/api-reference/actions) raises your resolution rate, and every extra resolution is a conversation your team didn't have to handle. ## What you are never charged for | Not charged | Why | | - | - | | **Escalated conversations** | Any conversation tagged **Escalated to Human Team** in [Analytics](/en/analytics) is a human handover, so it is not a resolution. This includes escalations triggered by a [Guardrail](/en/configuration/guardrails), by an [escalation reason](/en/analytics#escalation-reasons) such as missing knowledge or a failed API call, and by customers who ask for a human. | | **Conversations waiting on the customer** | A conversation in **Waiting for Customer** has not been solved. If the customer never returns, it is an abandoned session, which is never billable. | | **Greetings and spam** | Never billable. | | **Conversations the customer reopens within 72 hours** | A resolution requires 72 hours with no customer reply, so a reopen inside that window is not a resolution. | | **Conversations a teammate touched** | If anyone on your team replies to or works on the conversation, it is not a resolution. | | **Seats** | No per-seat fees. Your human team is never billed. | The same rules as a decision path, from an incoming conversation to the bill: ```mermaid theme={null} --- title: When a conversation becomes billable --- flowchart TD CONV(("Conversation")) REAL{"Real support request?"} STATUS{"Conversation status"} RES["Resolved by AI"] WAIT["Waiting for Customer"] ESC["Escalated to Human Team"] REOPEN{"Customer reply or human touch
within 72 hours?"} BILL["One billable resolution"] FREE["Not charged"] CONV --> REAL REAL -->|"greeting or spam"| FREE REAL -->|"yes"| STATUS STATUS --> RES STATUS --> WAIT STATUS --> ESC ESC --> FREE WAIT -->|"customer never returns"| FREE WAIT -.->|"customer replies"| STATUS RES --> REOPEN REOPEN -->|"no"| BILL REOPEN -->|"yes"| FREE classDef source fill:#F7F7F7,color:#131415,stroke:#E8E8E8 classDef agent fill:#131415,color:#FFFFFF,stroke:#131415,stroke-width:3px classDef surface fill:#FFFFFF,color:#131415,stroke:#131415 classDef human fill:#C3EE5E,color:#131415,stroke:#131415,stroke-width:2px class CONV agent class REAL,STATUS,REOPEN,RES,WAIT,ESC surface class FREE source class BILL human ``` Analytics also shows **Deflection rate**, which is `100% - Human escalation rate` and counts waiting conversations as deflected. Billing does not follow deflection. Use **AI resolution rate**, which counts only **Resolved by AI** conversations, when you think about spend. See [Resolution vs deflection](/en/performance/resolution-vs-deflection). ## How to estimate your spend You can estimate billable resolutions before you launch, then replace the estimate with real numbers once the agent is live. Use your helpdesk's monthly inbound volume across the channels you plan to put Fini on (chat and email; voice rates are listed on the [pricing page](https://www.usefini.com/pricing)). Once Fini is live, use **Conversation volume** from the [Analytics](/en/analytics) KPI cards instead. Subtract greetings, spam, and conversations customers abandon. If you don't have a baseline, review a sample of a few hundred recent tickets and count them. Multiply what's left by the share you expect the agent to resolve end to end. Plan with the [rollout timeline](/en/rollout-timeline) in mind: the knowledge agent is live on day 1, agentic workflows come by day 14, and full autonomy by day 30, so resolution rate climbs over the first month as you add Intent Rules and Actions. After launch, use **AI resolution rate** from Analytics. Billable resolutions times your per-resolution rate gives your estimate. Pricing starts at \$0.49 per resolution; your plan's rate is on the [pricing page](https://www.usefini.com/pricing). Conversations the customer reopens within 72 hours are not resolutions, so leave some margin for them. **Worked example (volumes only).** A team receives 40,000 conversations a month. About 2,000 are greetings, spam, or abandoned, leaving 38,000. In the first month the agent resolves 50% of those end to end, so roughly 19,000 billable resolutions. By month three, with refunds, cancellations, and order changes running as Intent Rules, the agent resolves 75%, so roughly 28,500. The other 9,500 escalate to the team with full context and are not billed. The numbers in this example are illustrative, not a benchmark. The fastest way to lower your cost per solved issue is to raise resolution rate, not to cut volume. Use the [Knowledge performance](/en/analytics#knowledge-performance) and [Intent rule breakdown](/en/analytics#intent-rule-breakdown) tables to find high-volume slices with a low **AI Resolve Rate**, then close the gap with [Articles or Sources](/en/knowledge/sources) or a new Intent Rule. ## Voice The Fini agent also answers inbound phone calls (see [Voice](/en/deploy/voice)). Voice rates are listed on the [pricing page](https://www.usefini.com/pricing). ## How you buy and pay ### Direct contract, invoiced through Stripe Under the [Terms of Service](https://www.usefini.com/security/terms-of-service), fees are invoiced through Stripe with 30-day payment terms. ### Microsoft Marketplace (MACC-eligible) Fini is listed on the Microsoft Commercial Marketplace. Buying through the Marketplace means: * **One contract and one invoice through Microsoft.** No separate vendor onboarding. * **MACC-eligible spend.** Purchases count toward your Microsoft Azure Consumption Commitment. * **Azure-hosted deployment.** Fini can be deployed in your Azure tenant, with Entra ID and Conditional Access, Private Link and VNet support, and customer-managed encryption keys. See [Fini on Microsoft Azure](https://www.usefini.com/better-together) for the listing and the Azure deployment details. ### Changing how you're billed Custom volume and contract terms are available for enterprise deployments. Contact [founders@usefini.com](mailto:founders@usefini.com) or use the [contact page](https://www.usefini.com/company/contact). ## Related The exact definition, examples, and how it maps to Analytics statuses. Eligibility, the three performance conditions, and how claims work. Payment terms, Marketplace purchases, and what shows up on your invoice. Track AI resolution rate, the number that drives your bill. ## Related topics - [RFP fact sheet](/en/evaluate/rfp-fact-sheet.md) - [Questions to ask an AI support vendor, with Fini's answers](/en/evaluate/questions-to-ask.md) - [Billing FAQ](/en/billing/billing-faq.md) This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.