Intercom Chatbot Pricing: The Fin Outcome Fee Explained
Intercom bills two ways at once: a seat for every human, and $0.99 every time Fin reaches an outcome. The second one is what surprises people, because it scales with how well the AI works.
Prices verified 20 July 2026. Vendors change pricing often, so every figure links to its source.
One thing to know before the numbers, because it affects whether you want a multi-year commitment here. On 15 June 2026 Salesforce signed a definitive agreement to acquire Fin, the company formerly known as Intercom, for approximately $3.6 billion. The deal is expected to close in the fourth quarter of Salesforce's fiscal 2027. The helpdesk is still sold as Intercom and Fin is still its AI agent, so nothing about today's pricing has changed, but the roadmap now belongs to Salesforce.
The line that decides your Intercom bill is on their pricing page: Fin costs "$0.99 per outcome", and you are "only charged once per conversation, even if multiple questions are answered". Fin's own pricing page states a 50-outcome monthly minimum and a separate $9.99 charge per qualification.
That is a usage fee sitting on top of a per-seat subscription. It is worth understanding before you scale, because unlike a seat licence, the AI line goes UP as the bot gets better at its job.
How Intercom's pricing is structured
Three seat tiers, plus a per-outcome AI charge. The seat figures below come from third-party pricing trackers, because Intercom's own page did not render them when we checked.
| Component | Reported price | Source |
|---|---|---|
| Essential (per seat / mo) | About $39 paid yearly | Third-party trackers |
| Advanced (per seat / mo) | About $85 paid yearly | Third-party trackers |
| Expert (per seat / mo) | About $132 paid yearly | Third-party trackers |
| Fin AI | $0.99 per outcome | Intercom pricing page (verified) |
| Fin billing rule | Charged once per conversation | Intercom pricing page (verified) |
| Standalone Fin | Minimum monthly commitment applies | Intercom pricing page (verified) |
We are being deliberate about that Source column. Intercom's plan names and the Fin mechanics are quoted from their page. The seat dollar amounts are widely reported but we could not read them off the vendor page, so treat them as indicative and confirm on Intercom's pricing page before you budget.
Why the per-outcome fee is the part to model
A per-seat fee is predictable. A per-outcome fee rises with volume and with the AI's success rate.
Fin charges when it reaches an outcome, once per conversation. So the bill is roughly: conversations that reach an outcome, times $0.99. Push more traffic into chat, or improve the bot so it resolves more, and that line grows in direct proportion.
That is not a criticism of the model, it is just a different shape of cost. Usage pricing is fair in the sense that you pay for work done. It is simply harder to forecast, and it penalises exactly the outcome you are optimising for. Budget it as a variable cost, not a subscription.
Worked examples at three team sizes
Assumption stated up front so you can change it: half of chat conversations reach a billable Fin outcome. Seat rates are the third-party figures above, paid yearly.
3 seats, 2,000 conversations
- 3 Advanced seats at ~$85~$255
- 1,000 outcomes at $0.99$990
About $1,245 / month
The AI line is already ~4x the seat line.
10 seats, 8,000 conversations
- 10 Advanced seats at ~$85~$850
- 4,000 outcomes at $0.99$3,960
About $4,810 / month
25 seats, 20,000 conversations
- 25 Expert seats at ~$132~$3,300
- 10,000 outcomes at $0.99$9,900
About $13,200 / month
Roughly $158,000 a year at this volume.
Intercom vs matram.ai on cost
matram.ai is flat: $29, $69 or $199 a month, unlimited seats, no per-outcome fee.
At the 10-seat, 8,000-conversation example above, the difference is roughly $4,810 a month against $69. The reason the gap is that wide is structural: matram.ai does not charge per seat or per resolution, so volume and team growth do not move the price.
The honest counterweight is scope. Intercom is a customer engagement platform: product tours, outbound campaigns, onboarding flows, a mature inbox and a large app ecosystem. matram.ai is an AI chatbot that trains on your content, cites its sources, captures leads and hands off to a human. If you actively use Intercom's wider platform, you are not paying purely for chat and the comparison is not like for like.
If the chatbot is the piece you care about, the Intercom alternative page covers the switch, including what you lose.
When Intercom's pricing makes sense
- You use the engagement platform, not just chat: tours, outbound, onboarding, surveys.
- Your conversation volume is low enough that the per-outcome fee stays small.
- You want one vendor across support and product messaging, and will pay for that consolidation.
- You are early enough that a seat or two plus modest Fin usage is genuinely cheap.
It stops making sense when chat volume grows, because the outcome fee grows with it, and when you are paying platform prices for a product you only use as a chatbot.