The real math of Intercom + Fin pricing (worked examples)

Intercom charges on two meters at once: a monthly fee for every human seat, and a separate 99-cent charge every time its Fin AI agent resolves a conversation. Most people budget for the first meter and get blindsided by the second — because the resolution meter grows with your traffic, and at real volume it becomes the larger number by far. This post works the full bill for three common team shapes, using Intercom's published pricing as of mid-2026, so the number stops being a surprise.

We build a competing product, so read this with that in mind — but every figure here comes straight from Intercom's public pricing page, and the arithmetic is yours to redo. The goal is not to dunk on anyone. It is to show you where the money actually goes, because the pricing model is genuinely easy to under-count. Intercom's bill is not one number. It is two meters running side by side, and the one you forget is the one that scales.

How the bill is actually built

Intercom's published pricing as of mid-2026 has two independent parts.

The first is seats — a fee per human agent, per month, on annual billing: roughly $29 on the Essential plan, $85 on Advanced, and $132 on Expert. Monthly billing runs noticeably higher than annual, commonly 25 to 35 percent more. The plan you land on is not a free choice: the features most support teams actually want — the Workflows builder, multiple team inboxes, round-robin assignment, custom reports — sit on Advanced, and enterprise needs like SSO or HIPAA sit on Expert.

The second is Fin, Intercom's AI agent, billed at $0.99 per outcome. And this is the line everyone reads too fast. At roughly ₹95 to the US dollar, that is about ₹94 every time the AI closes a conversation. It is not a subscription. It is a per-event charge that fires thousands of times a month if your volume is high. You do not pay Intercom for the AI. You pay Intercom every time the AI works.

Throughout, we will assume Fin resolves half of incoming conversations — a middle-of-the-road number; real deployments land anywhere from 40 to 60 percent depending on how well the help content is seeded. Plug in your own rate; the shape holds.

Team one: the solo founder (~300 conversations/month)

For a one-person business doing about 300 conversations a month, the realistic bill is around $150 a month — roughly ₹14,000.

Here is the make-up. One Essential seat at $29. Fin resolving, say, 40 percent of those 300 conversations (a fresh setup, help content still thin) is 120 outcomes at $0.99, or about $119. Total: $148 a month, about ₹14,000 — call it ₹1.7 lakh a year for a single-founder support stack.

Notice the split: the human seat is $29, and the AI is $119. Even at a solo founder's tiny volume, four-fifths of the bill is the resolution meter, not the seat. The founder's instinct — "it's twenty-nine bucks a seat, that's fine" — is off by roughly five times, because the seat was never the expensive part.

Team two: the five-person team (~2,000/month)

A five-person team handling about 2,000 conversations a month lands near $1,400 a month — roughly ₹1.35 lakh.

Five people who want Workflows and multiple inboxes are on Advanced: five seats at $85 is $425. Fin resolving half of 2,000 conversations is 1,000 outcomes at $0.99, or $990. Total: about $1,415 a month, roughly ₹1.35 lakh — near ₹16 lakh a year.

The seat bill and the AI bill have now crossed over. At two thousand conversations, the AI meter ($990) is already bigger than your entire human seat bill ($425). And it is the meter you cannot cap, because it rises every time your business grows. A good month — more signups, more traffic, more questions — is also a bigger invoice. Success is billable.

Team three: the fifteen-person team (~8,000/month)

A fifteen-person team at about 8,000 conversations a month is looking at roughly $5,200 a month — around ₹5 lakh.

Fifteen Advanced seats at $85 is $1,275. Fin resolving half of 8,000 is 4,000 outcomes at $0.99, or $3,960. Total: about $5,235 a month, roughly ₹5 lakh — close to ₹60 lakh a year. Move to Expert for SSO and the seat portion alone jumps to nearly $2,000, pushing the whole thing past $5,900 a month.

Look at the ratio at this scale: the AI meter ($3,960) is more than three times the seat bill ($1,275). By the time you are a real support team, you are barely paying for people — you are paying, over and over, for the machine to answer. That is a defensible model if the resolutions are worth $0.99 each to you. The point is simply that you should know it is the number, because it is the one that runs your annual bill toward the price of a car.

The mechanics people miss

Four details reliably surprise people mid-invoice.

What counts as a "resolution." Fin's $0.99 does not only fire on a clean, happy answer. Intercom bills an "outcome," which per their published terms includes a resolution, a procedure handoff, or a disqualification — broadly, the AI engaging and the customer not coming back for more help counts. Lead-qualification outcomes are billed higher still, at $9.99 each. You are not billed for correct answers. You are billed for outcomes — a wider net than "the AI solved it."

Seats stack on top, always. There is no plan where Fin replaces your human seats. You pay per seat and per resolution, together. The AI does not subtract from the seat bill; it adds a second bill beside it.

Tier gates move you up the seat ladder. The moment you need Workflows, multiple inboxes, or custom reporting, you are on Advanced ($85) rather than Essential ($29) — for every seat, not just the ones using the feature. One needed feature can nearly triple your entire per-seat line.

The resolution meter is uncapped by design. Seats are predictable; you know how many people you employ. Resolutions are not — they float with your traffic, so the scary version of the bill is a viral week, not a steady month. Standalone Fin (bolted onto another helpdesk) even carries a 50-outcome monthly minimum of $49.50, so the meter has a floor but no ceiling.

None of this is hidden — it is all on the public pricing page. It is just easy to add up wrong, because the human brain anchors on the seat price it saw first and quietly forgets the meter that does the real damage.

The takeaway, and a different shape

Run your own version before you commit: your monthly conversations, times your expected resolution rate, times $0.99, plus your seats times the plan price. For most teams the answer is two to five times what the seat sticker suggested. The seat price is the ticket. The resolution meter is the ride.

We built Iris with the opposite instinct, and it is worth being plain about why: a flat per-seat price (published in rupees, GST included) with AI answers metered at a low, flat, published per-answer rate — not a 99-cent charge that climbs with your success. A busy month should make you money, not cost you more to support. That is the whole reason the pricing works the way it does; the contrast with a two-meter bill that scales against you speaks for itself once you have done the arithmetic above.