Intercom's Per-Resolution Pricing: What You'll Actually Pay in 2026

When Intercom launched Fin, its AI agent, it repositioned itself from a per-seat platform to a per-resolution model. The pitch was compelling: only pay when the bot actually solves something. No resolution, no charge.

In practice, the model creates a billing incentive that works against customers. This post breaks down exactly how Intercom Fin pricing works, where the assumptions fall apart, and what the real monthly cost looks like at different conversation volumes - compared to Velaro's conversation-based model.

How Intercom Fin Pricing Actually Works

Intercom's Fin AI agent is priced at $0.99 per resolved conversation. That's on top of the base platform fee, which starts at $29/month for the cheapest plan and scales quickly. Enterprise teams are typically negotiating annual contracts in the $20,000-$100,000+ range before Fin resolution fees are added.

The platform fee buys you access to the inbox, routing, and basic automation. Fin - the AI agent that actually deflects tickets - is billed separately, per resolution.

The resolution fee model sounds clean on paper. The problem is in the definition of "resolved."

The Assumed-Resolution Problem

Here's where the billing model has a fundamental flaw that Intercom doesn't prominently advertise.

When Fin handles a conversation, it marks the conversation as resolved based on its own assessment - not based on explicit customer confirmation. If the customer doesn't respond after Fin's last message, the system assumes the issue was resolved and the $0.99 charge fires.

Think about how many customer service interactions end without a reply. A customer asks about return policy, gets an answer, and closes the tab without saying "thanks, that helped." Under Intercom's model, that's a billable resolution. A customer asks where their order is, the bot replies with tracking info that's wrong, and the customer gives up in frustration and calls your 1-800 number instead - also billed as a resolution.

"We were charged for over 400 resolutions in one week where customers contacted us again within 24 hours on a different channel. Those were not resolved. We paid for 400 failures."

The assumed-resolution flaw means you're paying for:

• Conversations where customers gave up and found another channel
• Conversations where customers didn't reply but weren't satisfied
• Conversations that were "resolved" but reopened within 48 hours
• Mis-deflections where the bot answered a slightly different question than was asked

Running the Numbers at Scale

Let's look at what Intercom's pricing actually costs across a range of conversation volumes, using conservative assumptions: 60% bot resolution rate, 15% of "resolutions" are assumed (not confirmed).

These are just the Fin fees. Add your platform subscription (typically $400-$1,200/month for a mid-market team) and your per-seat charges for human agents on top.

The Billing Surprise Pattern

Across customer reviews on G2 and Trustpilot, a consistent pattern emerges with Intercom's per-resolution model. Teams sign up at low conversation volumes, the per-resolution fee seems manageable, then their support volume grows. The Fin charges grow with it. By the time the annual invoice arrives, the total cost is 2-4x what was budgeted.

Several teams have reported attempting to cancel or downgrade after experiencing this shock, only to find that Intercom requires 30-60 days notice before annual contract renewal. Missing that window means another year at full price.

What "Per Resolution" Means for Your Incentives

Beyond the billing math, the per-resolution model creates operational problems that are less obvious.

When your AI vendor profits more as resolution rates increase, the vendor has an incentive to count more things as resolved - not to make your customers happier. Assumed resolutions after silence are the direct result of this incentive. The model is optimized for billing events, not customer outcomes.

A conversation-based model like Velaro's charges for conversations regardless of outcome. That puts the incentive in the right place: help you reduce conversation volume through better self-service and knowledge bases, not inflate resolution counts to generate billing events.

Where Intercom Still Makes Sense

To be fair: if your support team is handling under 500 bot-resolved conversations per month and you're paying $495 or less in Fin fees, the model can work out. Intercom's interface is polished, the inbox UX is well-regarded, and for very small teams the all-in-one pitch is real.

The model breaks down at mid-market scale - typically once you're above 3,000-5,000 bot resolutions per month. At that point, Fin fees alone often exceed what a full alternative platform costs.

Velaro's Pricing Model: The Alternative

Velaro prices by conversation volume on a tiered monthly basis. The tiers are: 500 contacts, 5,000 contacts, 25,000 contacts, and 100,000 contacts. Every tier includes both bot and human agent conversations. There are no per-resolution fees.

For a team handling 10,000 monthly conversations with a 60% bot resolution rate, the cost difference compared to Intercom's model is substantial - often $3,000-$5,000 per month in Fin fees alone, before accounting for platform subscription differences.

The comparison also needs to account for what's included. Velaro includes omnichannel routing, conversation analytics, and co-browse in standard plans - features that Intercom charges separately for or doesn't offer at all.

Questions to Ask Before You Sign with Any AI Chat Vendor

Regardless of which platform you're evaluating, these questions will expose pricing model risks before you commit:

1. How exactly do you define a "resolved" conversation? Does the customer need to confirm, or is silence treated as confirmation?
2. Is there a monthly cap on resolution fees? If your volume spikes, what's the ceiling?
3. Can I see a bill from a similar-sized customer? Marketing math and actual invoices often diverge significantly.
4. What's the cancellation window? Annual contracts with 60-day notice windows can trap you for 14 months if you miss the renewal date.
5. Are there volume discounts, and at what thresholds? Per-resolution pricing sometimes has tiered rates that aren't advertised upfront.

The per-resolution model isn't inherently dishonest. But the assumed-resolution definition, combined with unpredictable scaling costs, makes it a difficult model to budget accurately at any volume above a few thousand conversations per month. Understand exactly what you're signing before you commit.

The Bottom Line

Intercom Fin charges $0.99 per "resolved" conversation - including conversations where the customer simply stopped replying. At 10,000 monthly conversations with a 60% bot resolution rate, that's nearly $6,000/month in resolution fees alone, before your base subscription. Velaro charges by conversation volume, not outcome - your bill doesn't go up because your AI got better at its job. For teams above 3,000 bot resolutions per month, the math strongly favors a conversation-based model.