How does Intercom charge $0.99 per Fin outcome, and how do we estimate outcomes from our current ticket volume?

Fin’s $0.99 “per outcome” pricing is designed to track what actually matters in support: resolved conversations, not just messages sent or time spent. To model costs from your current ticket volume, you’ll map today’s conversations to likely “Fin outcomes,” then apply Fin’s typical resolution rates and your own constraints (channels, topics, hours, identity checks).

Quick Answer: Intercom charges $0.99 for each successful Fin outcome—when Fin fully resolves a customer query with no human intervention. To estimate outcomes from your current ticket volume, segment your conversations by type/complexity, apply a realistic Fin resolution rate (often 50–70% for eligible volume), then multiply projected outcomes by $0.99.

The Quick Overview

What It Is: A usage-based pricing model where you pay $0.99 only when Fin successfully resolves a customer query (an “outcome”), not for every interaction or attempt.
Who It Is For: Support teams that want to scale AI resolution across web, mobile, email, and messaging channels, with clear, predictable economics tied to real impact.
Core Problem Solved: Traditional AI/chatbot pricing rewards activity, not results. Fin’s outcome-based pricing aligns cost with resolved conversations—so you can scale AI confidently without runaway usage bills.

How It Works

At a high level, an “outcome” is counted when Fin has taken a conversation from initial customer question to complete resolution—no human agent needed. You can think of it as the AI equivalent of a fully solved ticket.

Fin’s pricing flow looks like this:

Customer asks a question.
The customer reaches you via Messenger, your Help Center, or another Fin-enabled channel.
Fin engages and attempts resolution.
Fin reads the question, pulls from your Help Center, procedures, and connected systems, and either:
- Resolves the query end-to-end, or
- Hands off to your team in the Helpdesk/Inbox.
Outcome is recorded if fully resolved by Fin.
If Fin completes the conversation with no human takeover, that counts as one charged outcome at $0.99. If a human steps in (or you’ve deliberately limited Fin to partial assistance), that conversation does not bill an outcome.

In practice, you configure where Fin is allowed to run (channels, hours, topics), how it escalates, and what data it can use. Fin’s average resolution rate is around 66% across customers and continues to improve over time, so your outcomes—and ROI—grow as the system learns.

How Intercom Charges $0.99 Per Fin Outcome

Let’s break down what is (and isn’t) included in that $0.99.

What counts as a Fin outcome

A Fin outcome is typically counted when all of the following are true:

Fin handled the query end-to-end.
The customer’s question is answered, the intent is satisfied, and the conversation is concluded without a human taking over.
The conversation is “finished.”
The customer has either:
- Stopped responding after Fin provided a complete answer, or
- Explicitly indicated resolution (e.g., “Thanks, that’s all”).
No agent participation in the resolution.
If an agent replies in the conversation thread as part of solving the customer’s issue, that conversation won’t be billed as a Fin outcome.

Important: Intercom only charges when Fin truly resolves the query. If Fin contributes context but a human has to finish the job, that’s not an outcome and not billed at $0.99.

What does not count as an outcome

Scenarios that typically don’t generate a billed outcome include:

Fin responds, but the customer requests a human and the conversation is handed off.
Fin attempts an answer, but the agent steps in to correct or expand it.
Fin provides an article or a partial answer, but your workflow always routes this topic to humans by design (e.g., high-risk billing disputes where humans must approve).
The conversation is abandoned before Fin gets to a complete, relevant answer.

You’ll still see Fin’s impact in reduced handle time (through Copilot) and better self-serve rates, but you won’t pay outcome fees unless Fin takes the query across the finish line.

Channels where Fin outcomes can occur

Fin outcomes can be generated anywhere Fin is active, for example:

In-product or website Messenger
Your Help Center (article suggestions + conversational answers)
Email-style flows where Fin answers inbound questions and you control when it replies using predicates like “Email To” vs. “Email Cc” in Workflows
Messaging channels you integrate via Intercom, once Fin is enabled on those workflows

Because Fin runs in the same Helpdesk and Inbox as your humans, the handoff logic and outcome tracking is consistent across channels.

Estimating Fin Outcomes from Your Current Ticket Volume

To estimate your monthly Fin outcome volume—and therefore cost—you’re essentially answering:

“Of the conversations we handle today, how many can Fin safely and completely resolve, given our content, channels, and policies?”

Here’s the practical, operator-style way to do that.

Step 1: Quantify your current volume

Start with the last 1–3 months:

Total monthly conversations (or tickets) handled by your support team.
Breakdown by channel:
- Web/Messenger
- Help Center contact forms or “contact us” from articles
- Email
- Messaging apps (WhatsApp, SMS, etc.)
Breakdown by intent or category if you have it (billing, account access, how-to, bugs, refunds, etc.).

If you’re migrating from another helpdesk (e.g., Jira Service Management, Zendesk), export your ticket stats and map categories to something like:

Password / login / account access
Plan / billing / invoices
How-to questions about existing features
Setup/onboarding questions
Bugs/incidents
Policy/sensitive actions (cancellations, data requests, etc.)

This categorization is the backbone of your Fin outcome estimate.

Step 2: Define Fin-eligible volume

Not every conversation should be automated, especially early on. Filter down to the slice you’re comfortable letting Fin handle.

Typically Fin-eligible volume includes:

How-to and feature questions
“Where do I change my password?”, “How do I export a report?”, “Can I invite another team member?”
Procedural queries that have clear playbooks
“What’s your refund policy?”, “How long does shipping take?”, “How do I update my billing address?”
Account and billing questions that can be answered safely from your knowledge base or via Data connectors and Fin Tasks with proper identity verification.

Usually excluded, at least initially:

Complex multi-party escalations or high-touch commercial conversations.
Edge cases around compliance, legal, or bespoke enterprise deals.
High-risk actions (account closure, legal requests) where policy mandates human review.

As a rule of thumb, many teams find 40–70% of their volume is “Fin-eligible” once knowledge and procedures are in good shape.

Step 3: Apply a realistic Fin resolution rate

Fin’s reported average resolution rate across customers is about 66% and increases by roughly 1% each month as the system improves. That’s across all customers and use cases; your initial rate will depend on:

How complete and accurate your Help Center and internal procedures are.
Whether you connect key systems (via Data connectors and Fin Tasks/Procedures) so Fin can actually resolve, not just explain.
The breadth of topics you initially allow Fin to handle.

For estimation, I usually model three scenarios on Fin-eligible volume:

Conservative: 40–50% resolution rate
Expected: 60–70% resolution rate
Optimistic: 75–80%+ once the system is mature

This gives you a band rather than a single number, so you’re not surprised either way.

Step 4: Estimate monthly outcomes and cost

Here’s a concrete example.

Assume:

20,000 total monthly conversations.
60% are Fin-eligible → 12,000 eligible conversations.
You forecast a 65% Fin resolution rate on eligible conversations.

Then:

Estimated Fin outcomes = 12,000 × 65% = 7,800 outcomes/month
At $0.99 per outcome = 7,800 × $0.99 ≈ $7,722/month

Model a conservative and optimistic case on the same base:

Conservative (50%): 12,000 × 50% = 6,000 outcomes → $5,940/month
Optimistic (75%): 12,000 × 75% = 9,000 outcomes → $8,910/month

Now you can compare that against:

Current support headcount + hiring plan you’d otherwise need.
SLA/response time targets and backlog reduction.
Value of 24/7 coverage without adding shifts.

Step 5: Account for channel and workflow choices

Your configuration will influence how many conversations ever get the chance to become a Fin outcome:

Where Fin is enabled:
If Fin only runs in your Messenger and Help Center, but 70% of your volume is still email, the immediate outcome pool is smaller. As you deploy Fin across more channels via Workflows, your eligible base grows.
When Fin is allowed to answer:
In email flows, for example, you can use Workflows predicates like:
- “Email To” is support@yourcompany.com → Fin can answer.
- “Email Cc” contains support@yourcompany.com → route to humans only.
  That choice directly limits or expands Fin’s opportunities to generate outcomes.
Operating hours and languages:
Some teams start by letting Fin handle after-hours queries only. That still generates outcomes, but from a smaller subset of volume. Once you see performance, you can expand to business hours and additional languages.

In other words: your initial Fin outcome estimate is not fixed—it’s a function of where and how you deploy Fin.

Features & Benefits Breakdown

Core Feature	What It Does	Primary Benefit
Outcome-based pricing	Charges $0.99 only when Fin fully resolves a customer query.	Aligns cost with real value—resolved conversations—so you don’t pay for partial or failed attempts.
Shared Helpdesk + Fin inbox	Fin and human agents operate from the same Helpdesk and Inbox with a shared customer view.	Seamless handoffs and complete context mean no duplicated effort and consistent customer experiences.
AI Insights & continuous tuning	Surfaces topics Fin can’t yet resolve and shows performance by channel and intent.	You can systematically improve content and workflows—so resolution rates and ROI increase over time.
Data connectors & Fin Tasks	Connect Fin to external systems and orchestrate multi-step procedures with identity checks and business logic.	Lets Fin handle complex workflows (e.g., disputes, refunds) end‑to‑end, increasing outcome volume safely.
Copilot for agents	Assists humans during live conversations with suggestions, summaries, and translations.	Even when Fin doesn’t own the outcome, agents resolve more queries per day (31% more in testing).

Ideal Use Cases

Best for high-volume, repetitive support:
Because a large share of queries are how‑to or policy questions that Fin can resolve reliably, quickly raising your AI resolution rate and reducing pressure on your team.
Best for scaling teams in the “breaking point” zone:
Because you can offload a big chunk of volume to Fin instead of adding headcount, while outcome-based pricing keeps costs proportional to real resolution.

Limitations & Considerations

Fin needs good knowledge to resolve:
If your Help Center and internal procedures are sparse or outdated, Fin’s resolution rate—and therefore outcome volume—will underperform. Plan for a content pass and use AI Insights to prioritize gaps.
Not every topic should be automated on day one:
Sensitive or high-risk workflows (e.g., financial changes, data deletions) should be gated behind identity checks, Fin Tasks, and clear escalation paths. That may reduce short-term outcome volume but is essential for safety and trust.

Pricing & Plans

Outcome-based pricing for Fin sits alongside your Intercom Customer Service Suite or Helpdesk plan. Think of it in two layers:

Base platform (Helpdesk + Messenger + Help Center + Inbox + Workflows + Copilot):
This gives you the connected system where AI and humans work together.
Fin AI Agent usage at $0.99 per outcome:
Billed based on the number of successful Fin outcomes per month.

Exact plan names (e.g., Starter, Pro, Premium) and platform pricing will depend on your region, scale, and configuration. But you can think about fit broadly as:

Core Suite + Fin for growing teams:
Best for teams that want to replace or modernize an existing helpdesk and introduce Fin as a primary resolution layer.
Fin-first on top of existing tools:
Best for teams who want to keep their current helpdesk for now but use Fin to resolve a large share of front-line queries and hand off cleanly to humans.

To get precise pricing for your workspace—base plan plus expected Fin outcome volume—Intercom will model your actual data with you.

Frequently Asked Questions

How exactly is a Fin outcome defined for billing?

Short Answer: A Fin outcome is counted when Fin fully resolves a customer query without human intervention.

Details:
An outcome is billed when Fin handles the conversation end‑to‑end on a Fin-enabled channel, provides a complete and relevant answer, and no human agent participates in resolving the issue. If an agent steps in, the customer is routed directly to an agent based on your workflows, or the conversation ends without a clear resolution, that interaction is not billed as a Fin outcome. You can see Fin’s performance and handoffs in reporting, so you can validate how many conversations are being resolved by Fin versus humans.

How do we build a reliable forecast of Fin cost from our current ticket data?

Short Answer: Segment your current volume by topic and channel, estimate the Fin-eligible slice, apply a conservative and expected resolution rate, then multiply projected outcomes by $0.99.

Details:
Pull 1–3 months of historical data from your existing helpdesk (or Intercom if you’re already using it):

Count total monthly conversations and group by channel and category (e.g., login, billing, feature how‑to, bugs).
Mark each category as Fin-eligible or not for your first phase, excluding high‑risk or bespoke issues.
Multiply total volume in the Fin-eligible categories by a conservative resolution rate (e.g., 50%) and an expected rate (e.g., 65–70%) based on how mature your knowledge and procedures are.
Multiply those estimated outcomes by $0.99 to get a monthly cost range.

Once Fin is live, use AI Insights to refine your model by actual intent and resolution rate per topic and channel. Most teams see resolution improve over the first few months as they add articles, connect systems via Data connectors, and automate more complex workflows with Fin Tasks/Procedures.

Summary

Fin’s $0.99 per outcome model is built for leaders who care about resolved conversations, not just AI activity. You pay only when Fin fully handles a query, and you can tune where it runs, which topics it covers, and how it hands off to your team—all from the same Intercom Helpdesk and Inbox your agents use every day. To estimate cost from your current ticket volume, treat it like any production system rollout: quantify your volume, define what’s safely automatable, apply realistic resolution-rate scenarios, and then iterate as AI Insights sharpen your view of where Fin can do more.

Next Step

Get Started