AI Support Tool Pricing 2026: 7 Vendors Ranked by Cost Model

The honest map of AI support pricing in 2026 splits on one axis: who publishes a rate.^[1][5] Three vendors (eesel, Gorgias, Intercom Fin) put a number on the page you can plan against. Four (Decagon, Sierra, Ada, Forethought) route everyone through sales and publish no dollar figure. That divide matters more to a solo founder than any cents-per-resolution gap, because you can only budget against a published rate. This guide ranks all seven by transparency and works through the cost model that fits each stage.

1. The four pricing models

AI support tools price along four shapes:

• Per ticket (per task): a flat fee for every ticket the AI works, resolved or not. eesel at $0.40 per task is the clearest example.^[2]
• Per resolution: a fee only when the AI fully closes a conversation on its own. Gorgias ($0.90 annual) and Intercom Fin ($0.99) use this.^[3][1]
• Outcome-based: a fee tied to a defined business outcome (resolution, retention, upsell). Sierra uses this, with no public dollar rate.^[4]
• Platform / custom enterprise: usage-based contracts with a platform fee, priced through sales. Decagon, Ada, and Forethought sit here.^[5]

2. All 7 vendors, ranked by transparency

Vendor	Model	Published rate	Self-serve
eesel	Per ticket[2]	$0.40 per ticket[2]	Yes
Gorgias AI	Per resolution[3]	$0.90/res annual, $1.00 monthly[3]	Yes (with helpdesk plan)
Intercom Fin	Per resolution[1]	$0.99 per resolution[1]	Yes (14-day trial)
Sierra	Outcome-based[4]	Not publicly documented (as of May 2026)[4]	No, sales-led
Decagon	Custom enterprise[5]	Not publicly documented (as of May 2026)[5]	No, sales-led
Ada	Custom enterprise	Not publicly documented (as of May 2026)	No, sales-led
Forethought	Outcome / custom	Not publicly documented (as of May 2026)	No, sales-led

Four rows reading "not publicly documented" is not a gap in the research; it is the finding. Decagon, Sierra, Ada, and Forethought publish no per-resolution dollar rate, and inventing one would be worse than naming the absence. If you need to know the bill before signing, the field is really three vendors deep.

3. The vendors that publish a rate

eesel charges $0.40 per regular task, one ticket equals one task regardless of message count, with no base fee, no seat fee, and no monthly minimum.^[2] It is the lowest published unit rate and the lightest to layer onto an existing helpdesk.

Gorgias AI Agent charges $0.90 per automated resolution on annual billing ($1.00 monthly), billed only when the AI fully resolves a conversation on its own, and sits on top of a Gorgias helpdesk plan starting at $10/month.^[3] Best if Gorgias is already your helpdesk.

Intercom Fin charges $0.99 per resolution (outcome) with a 50-outcome monthly minimum standalone and no platform fee against an external helpdesk; bundled with Intercom, seats add $29 to $132 per seat per month.^[1] The most mature ecosystem of the three.

The unit caveat runs through all three: eesel charges per ticket worked, Gorgias and Fin charge only on full resolutions. At a 60% resolution rate, eesel bills 100 tickets while the other two bill 60, which narrows the apparent rate gap. The eesel vs Gorgias comparison works that arithmetic in detail.

4. The quote-only enterprise vendors

Sierra publishes its outcome-based model (pay on achieved outcomes, no charge in most cases on escalation) but no dollar rate, set per contract.^[4] Decagon publishes nothing and routes to a demo.^[5] Ada is quote-based and states it fits companies with at least 300,000 annual conversations. Forethought is outcome-based but contact-sales for a quote.

The common thread: these four target enterprise volume and price through sales, so a solo founder cannot model the cost in advance. Third-party coverage circulates contract ranges, but those are estimates, not published rates, and do not meet the bar for a cited price here. For each, the honest line is: not publicly documented (as of May 2026). The head-to-heads on the published rivals are in Intercom Fin vs Decagon and Sierra vs Intercom Fin.

5. The human baseline every vendor replaces

Whatever the pricing model, every vendor is sold against human-agent cost. The engine below prices a 4,000-ticket-per-month operation at a 60% resolution rate, inside the 55-70% tier-1 AI resolution band industry benchmarks report for 2026. The engine's AI-cost figure is a model token-and-escalation cost, not a vendor invoice; treat the published vendor rates above as the real per-resolution line, and the engine as the human-vs-AI economics it sits inside.

At this volume the human-only cost is $28,000 a month, and an AI-first setup resolving 60% of tickets cuts it roughly in half. Against that gap, the vendor per-resolution fee is the smaller line: 2,400 resolutions at $0.99 (Fin) is about $2,376, far below the human cost it displaces. The vendor decision is which pricing model is predictable and which fits your stack, not whether automation pays. Run your own inputs in the AI vs Human Support Cost calculator.

6. How to choose by stage

1. Solo founder or SMB, want a budgetable rate: eesel, Gorgias AI, or Intercom Fin. All three publish self-serve rates. Pick eesel for the lowest unit price and lightest stack fit, Gorgias if it is already your helpdesk, Fin for ecosystem maturity.^[2][3][1]
2. High resolution rate expected: per-ticket (eesel) tends to win because you pay either way and the unit rate is lower.
3. Noisy or low-resolution volume: per-resolution (Gorgias, Fin) avoids paying for tickets the AI never closes.
4. Enterprise volume, custom workflows: Decagon, Sierra, Ada, or Forethought, but budget for a sales process and a contract-defined rate rather than a published number.^[5]

The support cost methodology is at the AI vs Human Support Cost methodology page.^[6]

Frequently asked questions

What is the cheapest AI support tool per resolution in 2026?

Among the vendors that publish a rate, eesel is the lowest unit price at $0.40 per ticket, followed by Gorgias AI Agent at $0.90 per resolution on annual billing and Intercom Fin at $0.99 per resolution. The catch is that the unit differs: eesel charges per ticket worked, while Gorgias and Fin charge only on resolutions. Decagon, Sierra, Ada, and Forethought do not publish rates and are quote-only enterprise platforms (verified May 2026).

Which AI support vendors publish their pricing?

As of May 2026, eesel ($0.40 per task), Gorgias AI Agent ($0.90 per resolution annual), and Intercom Fin ($0.99 per resolution) publish self-serve rates. Sierra publishes its outcome-based model but no dollar figure. Decagon, Ada, and Forethought publish no pricing at all and route buyers through sales, so their per-resolution cost is not publicly documented.

Is per-resolution or per-ticket pricing better for AI support?

Per-resolution pricing (Gorgias, Intercom Fin) only charges when the AI fully closes a conversation, so you never pay for tickets it escalates, which suits noisy ticket volume. Per-ticket pricing (eesel) charges for every ticket worked at a lower unit rate, which wins when resolution rates are high. The right model depends on your resolution rate and how much of your volume the AI can actually close.