Chatbot Pricing Comparison: Chatbot Prices in 2026

Across the twelve chatbot and AI support vendors compared below, entry plans run roughly $24 to $79 a month, mid tiers land between $100 and $200, and per-seat suites reach $55 to $139 per agent per month. The figure that decides your invoice is not the plan price, it is the meter on top: Intercom charges from $0.99 per Fin outcome, Gorgias $1.00 per AI resolution ($0.90 annual), Help Scout $0.75, Zendesk roughly $1.50 to $2.00, and Front from $0.05 per conversation. Two companies on the same plan can pay wildly different amounts, because success is what the meter bills for.

Last updated July 2026.

Every vendor in this market prices differently on purpose, which makes a straight comparison harder than it should be. One charges per seat, one per resolution, one per active contact, one per conversation, and one per ticket bundle. Below is the price list as each vendor publishes it, followed by what those models actually do to a real bill. Every figure here comes from the vendor's own pricing page unless the row says otherwise, and where a number is not public or where independent trackers disagree, this page says so instead of inventing one.

Chatbot price list: what each vendor charges in 2026

Prices below are monthly and, where a vendor discounts annual billing, the annual rate is shown because that is the one most buyers sign. Read the third column first. It is the part that changes your invoice month to month.

If you want the reasoning behind each of those rows, the individual breakdowns live on the Intercom alternative, Gorgias alternative, and respond.io alternative comparisons, and the full model-by-model explanation sits on the chatbot pricing pillar.

What are the different chatbot pricing models?

There are five, and they behave very differently as you grow. Flat subscription bills one predictable amount per month. Per seat bills for each human agent. Per resolution or per outcome bills each time the AI successfully closes a conversation. Per contact bills for the size of your audience. Per conversation bills for each thread the AI touches, resolved or not.

The distinction that matters is whether the meter is tied to something you want more of. Seats are fine, because you control headcount. Resolutions are the problem: the better the AI works, the more you pay, and you cannot forecast it. A month where a product launch triples your inbound is exactly the month your bill triples too, and it arrives after you have already spent the money. We went through that arithmetic in detail in why per-resolution AI pricing hurts.

What is the average chatbot pricing?

For a small US business, the realistic average is $30 to $150 a month on flat plans, and $200 to $1,000 or more once a vendor meters you. The spread is that wide because usage pricing is not really a price, it is a formula. A 15-person ecommerce brand handling 800 support conversations a month pays roughly $720 in resolution fees alone at $0.90 each, before any seat or platform charge.

That is the number worth calculating before you sign anything. Take your monthly conversation volume, multiply it by the vendor's per-resolution rate, and add the plan cost. If the result is uncomfortable, the plan price was never the real price. There is a fuller worked example in chatbot cost for a small business.

Why do chatbot vendors charge per resolution?

Because it sounds fair and it scales with their revenue rather than your budget. The pitch is that you only pay when the AI works, which compares nicely against a support agent's salary. It is a genuinely reasonable story for a company deflecting 50,000 tickets a month and replacing headcount with software.

It works much worse for everyone else. Most businesses buying a chatbot are not replacing agents, they are covering hours nobody was covering and answering messages that used to go unanswered. Those extra conversations are new volume, not saved salary, so every one of them is a new charge against a budget that did not previously exist. The model also quietly discourages the thing you bought it for, because pointing more traffic at the AI costs more money.

What hidden fees should I expect on a chatbot plan?

Four show up repeatedly, and none of them are on the pricing page headline. Message fee markups are the largest: several vendors resell WhatsApp and SMS traffic at Meta or carrier cost plus roughly 20%, which compounds on high-volume accounts. AI add-ons are the second, where the inbox is on the advertised plan but the AI that does the work is priced separately per seat. Channel gating is the third, where the entry tier only covers one or two channels and going omnichannel forces an upgrade. Overage rates are the fourth, and they are usually higher than the bundled rate, so exceeding your allowance costs more per unit than staying inside it.

Ask each vendor for one number before you commit: the all-in monthly cost at twice your current volume. Any vendor that cannot produce it is telling you something.

Is flat-rate chatbot pricing actually cheaper?

Below a few thousand conversations a month, almost always yes, and the gap widens as you succeed. Here is the same 800-conversation month priced three ways.

Flat pricing stops being the obvious answer at enterprise scale, where a vendor will negotiate a committed rate and you have a finance team modeling it. For everyone under that, predictability is worth more than a lower headline rate, because it lets you point every channel at the AI without watching a meter. That is also when the savings become real for the team: once the routine tier is handled, the people you already employ can move onto the higher-value customer experience operations that were getting squeezed out by inbox triage.

Which chatbot pricing model should I choose?

Match the model to your volume curve, not to your current month. Choose flat if your volume is growing or spiky, if you want to route SMS, WhatsApp, Instagram, and web chat at the AI without rationing, or if you need to forecast a number. Choose per resolution only if your volume is flat and predictable and you are genuinely retiring headcount. Choose per contact only if your list size is stable, which for most growing businesses it is not.

One more filter: check what the AI is allowed to do. Several of these vendors meter resolutions but the AI only closes support tickets, so qualifying a lead or booking an appointment either costs extra or is not offered. If the AI is supposed to sell as well as support, price the whole job rather than the deflection rate. Our own pricing is flat by tier with qualifying, booking, and upselling included, and carrier and Meta message fees passed through at cost.

How often do these prices change?

Frequently enough that any comparison older than a quarter should be treated as a starting point rather than a quote. Zendesk changed how automated-resolution allowances work in January 2026, ManyChat restructured its tiers into a new ladder in 2026, and Drift stopped being sold as a standalone product entirely. This page is re-verified against vendor pricing pages on a rolling basis, and rows where a vendor no longer publishes numbers say so rather than repeating a figure from an old blog post.

If you want the shorter version aimed at a first-time buyer, start with how much an AI chatbot costs, or the model-by-model explainer in chatbot pricing models. Agencies reselling this to clients need a different set of numbers, which are in chatbot as a service pricing.