How Much Does an AI Chatbot Cost? 2026 Pricing Guide

An AI chatbot can cost nothing to start and millions to build. Most businesses land somewhere in between, and the final number depends on a handful of concrete choices: how many conversations you handle, which AI model runs them, how many channels and integrations you connect, and whether you buy a platform or build your own.

Here is the short answer. You can start for free. Several platforms include a free plan, and Quickchat AI’s gives you 50 AI replies a month with no card. Paid plans commonly run from $9 to $999 a month, and most small and mid-sized businesses settle on a plan between $29 and $99. Usage-based and enterprise plans are billed per resolved conversation, often from around $0.50 each. Building your own from scratch is a different order of magnitude: $5,000 to $30,000 for a simple bot, $75,000 to $500,000 or more for a custom AI bot, plus ongoing model and hosting bills.

The rest of this guide breaks those numbers down, and then does something most pricing guides do not: we actually built an AI chatbot on the free plan and counted every real cost, so you can see exactly where the money goes.

Chatbot cost at a glance

Rule of thumb for usage-based pricing: budget roughly $0.50 to $6 per resolved conversation, which usually covers the AI, the platform, and some support.

In this guide:

• How much does an AI chatbot cost?
• What makes a chatbot more expensive?
• Chatbot pricing models explained
• Is it cheaper to build or buy a chatbot?
• We built an AI chatbot on the free plan
• Hidden chatbot costs people miss
• How to estimate your monthly chatbot cost

How much does an AI chatbot cost?

AI chatbot costs run from completely free basic tools to enterprise builds topping $1 million. Here is a simpler way to look at the ranges by business type in 2026.

Basic chatbot for a small or mid-sized business

• Low: around $9 to $30 a month on a no-code AI platform, or free to start.
• Typical: $29 to $99 a month on subscription, or a one-time build of $5,000 to $30,000 for a custom rule-based bot.
• High: $150 to $300 a month with advanced features, or build costs up to $30,000.
• Handles FAQs, basic lead capture, and simple routing.

AI chatbot for a mid-market company

• Low: around $99 a month on subscription for an AI platform with natural language understanding.
• Typical: roughly $299 to $999 a month on subscription, or a custom build of $75,000 to $150,000.
• High: $999+ a month on subscription, or build costs up to $500,000 for advanced AI and deeper integrations.
• Handles personalized support, resolves complex queries, connects to your CRM or helpdesk.

Enterprise generative AI chatbot

• Low to typical: usage-based from $0.50 to $2.00 per resolved conversation, which at scale often works out cheaper than a fixed high-tier subscription.
• Custom build: usually $150,000 to $500,000.
• High: $1,000,000 and beyond for proprietary models with extensive data training, top-tier security, and complex integrations.
• Handles sophisticated multi-turn conversations, hyper-personalization, and internal knowledge management.

These numbers are guideposts. Actual costs shift with vendor pricing, the features you choose, and how complex your build gets. The single biggest factor is the chatbot’s core nature: whether it follows predefined rules, uses AI, or runs on a large language model (LLM).

What makes a chatbot more expensive?

Beyond the pricing model, a handful of factors shape the total cost. Understanding these drivers helps you budget accurately and avoid surprises.

The type of bot

The core technology, the bot’s “type,” is the biggest cost decider. More intelligence and capability means higher development or subscription cost.

If you want the deeper distinction between a wrapper over a general model and a purpose-built agent, see GPTs vs Quickchat AI: what’s the difference.

Advanced features

Each sophisticated capability adds development complexity and often pushes you into a higher subscription tier.

On a modern platform, most of these are simply included in the plan rather than billed as custom development. Multilingual support and personalities, for example, ship on every Quickchat AI tier.

Integrations

Chatbots rarely work in isolation; they connect to your existing systems. Custom API development for each connection can run from $5,000 to $25,000 when you build it yourself.

On a no-code platform, most of these connections are one-click and included, so the “integration cost” is your setup time rather than a development bill.

Branding and user experience

Using a platform’s standard templates and flows is the budget-friendly path. A bespoke look, custom avatars, and tailored conversation design add roughly a 20% to 30% design premium on a custom build. On subscription platforms, removing vendor branding is usually a paid-tier feature rather than a development cost. For the exact steps to white-label an AI chatbot, including which parts are free and which need a paid plan, see our hands-on tutorial.

Platform, in-house team, or agency

Your build approach drives the cost more than almost anything else.

• No-code platform: pre-built modules and a visual editor cut development time to near zero. Plans range from free to a few hundred dollars a month. Best for most businesses.

• In-house build: maximum control, but significant recurring salary cost.

  Role	Approximate annual salary
  Project manager	$88,000
  Chatbot developer	$90,000 to $120,000
  NLP engineer	$110,000 to $140,000
  UI/UX designer	$80,000 to $110,000
  QA tester	$70,000 to $90,000

• Agency: expertise and faster delivery. Project fees run from $15,000 to $300,000 or more, with retainers often starting around $5,000 a month.

Support, maintenance, and hosting

For a custom build, ongoing costs are recurring and easy to underestimate.

On a SaaS platform, hosting, maintenance, and model upgrades are bundled into the subscription.

Compliance and security

Meeting standards like HIPAA, GDPR, or PCI DSS adds development, auditing, and sometimes legal cost. Annual recertification can run $2,000 to $10,000 for standards like HIPAA or GDPR. Neglecting compliance risks fines and reputational damage, so treat it as a required line item, not an optional one.

Chatbot pricing models explained

Vendors price chatbots in a few different ways, and each affects your upfront and ongoing cost differently.

Subscription plans

You pay a recurring monthly or annual fee for a platform and its features, usually in tiers that unlock more replies, seats, or capabilities as you grow.

Subscriptions give you predictable costs and ongoing updates. Watch the limits on lower tiers; you upgrade as your volume grows.

Usage-based pricing

Also called pay-as-you-go, usage-based pricing charges by how much you use the bot, measured in conversations, messages, or resolutions. A resolution is a request the chatbot successfully handled. Costs run from a fraction of a cent to several dollars per unit. This suits businesses with seasonal traffic, because your cost tracks demand, but it makes monthly bills less predictable during spikes. For a deeper comparison of per-resolution versus per-seat models, see AI agent pricing models compared.

Hybrid pricing

Hybrid pricing pairs an upfront setup fee, often $5,000 to $30,000, with ongoing usage fees. Regulated industries like finance and healthcare lean toward this model, because the setup fee funds tailored compliance and security work.

Custom builds

A one-time custom build gives you full ownership and can run from $10,000 to over $1 million. You own the intellectual property, but you are also responsible for all maintenance, updates, and infrastructure. Subscription platforms handle those for you but keep ownership of the core technology.

Is it cheaper to build or buy a chatbot?

For most businesses, buying is cheaper and faster. Building only wins at very large scale or when your requirements are genuinely unique. Buying means faster deployment and predictable costs; building means higher upfront investment and full control.

Comparing subscription platforms

The table below reflects list pricing observed in July 2026. Always check vendor sites for current figures, since prices change often. Note that several platforms have moved to per-resolution billing for AI features, so the monthly subscription is only part of the total.

One clear shift defines 2026 pricing: Intercom set the per-resolution template with Fin, and this year Zendesk and ChatBot.com followed, billing per resolved conversation rather than per seat. Tidio offers true per-resolution pricing only on its top tier. Per-resolution billing can make costs lower than the headline subscription suggests, or higher if your resolution rate is low, so model both scenarios before committing.

Building from scratch

Custom development offers unmatched control at a much higher and more recurring cost.

• In-house burn rate: a dedicated team (project manager, developers, an NLP engineer, a designer, a QA tester) can cost $400,000 to $700,000 a year in salaries alone, before benefits, tools, and overhead.
• Agency project fees: roughly $10,000 to $30,000 for a simple rule-based bot, $75,000 to $150,000 for a moderate AI bot, and $150,000 to $500,000 or more for advanced generative AI.

For a worked example of a custom customer-service build, see how to build an AI agent for customer service similar to Klarna.

Build or buy: a quick decision guide

Choose buying if you need a solution quickly, have fairly standard requirements, or prefer predictable operating costs. Most small and mid-market companies fit here. Choose building if you have highly unique needs, require deep proprietary integration, need to own the IP, and have the capital and talent to sustain it.

We built an AI chatbot on the free plan: what it really cost

To ground all of these ranges in something concrete, we built a real AI chatbot on the Quickchat AI free plan and tracked exactly what it cost. The short version: the build itself cost nothing, and the only thing that counts against the plan is the number of AI replies you send.

We created a support agent for a fictional small online plant shop, Fern & Petal. Here is what each step cost.

Step 1: create the agent (free)

Creating the agent, naming it, and writing its instructions is free and takes a couple of minutes with no code. This is the agent’s identity and main prompt.

Step 2: load its knowledge (free)

We added a few help-center articles: shipping and delivery, returns and refunds, and plant-care basics. The free plan includes up to 50 knowledge base articles, so a small business rarely hits the limit. Loading knowledge does not cost anything.

Step 3: add an AI Action (free)

We added one AI Action so the agent can look up an order’s status mid-conversation. AI Actions let the agent call a live API while it talks to a customer. They are unlimited on every plan, including free, so connecting the agent to your systems does not raise the bill. If you want to see exactly how an action is built, build a free AI agent that takes actions walks through the whole thing with copy-paste values.

Step 4: what actually counts, the AI replies

Here is the one thing that costs anything on a usage-metered platform: the AI replies. Each reply the agent generates spends one AI credit, and the free plan includes 50 AI credits a month. A conversation is usually several replies, so 50 credits is enough to test the agent and handle a light stream of real questions.

There is a catch worth knowing before you pick a plan: testing your own agent spends the same credits a customer would. Every time you try it out, that reply counts. After building Fern & Petal and testing it a handful of times, the account had used 14 of its 50 monthly credits, most of them from our own test messages.

What Quickchat AI actually costs

When you outgrow 50 replies a month, here is the full plan lineup, with the real numbers. A conversation is several replies, so read the AI-reply column as your monthly capacity.

Two things make this simpler than most pricing pages. First, almost everything is on every plan, including free: all channels except the raw API, unlimited AI Actions, lead capture, human handoff, and analytics. Only three things are gated by tier, removing branding (Professional and up), exporting conversations (Business), and API access (Business). Second, advanced AI models cost more credits per reply and are available on paid plans; the free plan runs on the standard model, which is one credit per reply. Annual billing is ten months for the price of twelve. See the pricing page for the current details.

So the honest cost of a working AI chatbot on Quickchat AI is $0 to build and test, and then whatever tier covers your monthly reply volume, which for most small businesses is $29 to $99 a month.

Hidden chatbot costs people miss

The sticker price is not the whole bill. A few costs sneak up on your total cost of ownership.

• Testing spends your allowance. As we saw building Fern & Petal, the replies you generate while testing and tuning count against your monthly quota just like a customer’s. Budget some headroom above your expected volume.

• Cloud and model bills on a custom build. If you build your own bot or use pay-as-you-go AI services, compute and storage costs fluctuate and traffic spikes can produce surprisingly high monthly bills. Generative AI has higher per-interaction compute cost than simpler models.

• The cost of a bad experience. A poorly designed bot that gives wrong answers, loops, or blocks the path to a human frustrates users and can increase churn and cart abandonment. Winning back a lost customer usually costs far more than the bot ever saved.

• The cost of not having one. Without a bot, you miss leads and questions outside business hours. A chatbot answers around the clock and captures leads overnight, which can save significantly compared with staffing humans 24/7. For more on cutting support cost specifically, see how to reduce customer support costs.

How to estimate your monthly chatbot cost

Estimating your budget is a matter of working through your real needs rather than guessing. If you want a quick number first, the free chatbot ROI calculator estimates monthly savings from your ticket volume and support costs. Then work through these six steps.

1. Define scope and objectives. Decide what problems the bot will solve, who it is for, what tasks it must perform, and how you will measure success. Aim for a clear list of what it must do.

2. Pick a bot type and pricing model. Rule-based, AI-powered, or generative AI, and subscription, usage-based, hybrid, or custom. This gives you a preliminary cost for the core platform.

3. Estimate user and agent counts. How many customers will use it monthly, and how many internal seats will manage it. Seat counts drive subscription tiers on some platforms.

4. Forecast conversation volume. Estimate conversations or replies per month, including peaks and growth. On a platform like Quickchat AI, translate this into AI replies per month to pick a tier; on usage-based pricing, it sets your per-resolution bill.

5. List integrations and advanced features. Note the systems the bot must connect to and any advanced features. On a no-code platform most are included; on a custom build, price each.

6. Add contingency and ongoing costs. Include a contingency of about 15%, plus maintenance, support, retraining, and possible tier upgrades.

A quick worked example on a platform: a small store expecting about 800 customer conversations a month, at a few replies each, needs a few thousand AI replies a month. That maps to the $99 Essential plan (3,000 replies) with room to test, for a predictable $99 a month all in, no engineering required.

Do chatbots pay for themselves?

For most support and lead use cases, yes, and usually within months. A chatbot pays back through cost savings, extra revenue, and softer benefits like faster response times.

Cost per chat: bots versus humans

The clearest saving is support cost. A chatbot interaction is estimated at $0.50 to $0.70, while a live-agent interaction typically runs $8 to $15. If a bot handles 1,000 questions a month at $0.60 each ($600), and human agents would handle them at $10 each ($10,000), the monthly saving is $9,400.

Revenue, not just savings

Chatbots also lift revenue: they capture and qualify leads 24/7, guide shoppers through a purchase, and reduce cart abandonment, and they can upsell based on the conversation. To measure it, track leads and sales the bot influenced.

A simple ROI formula

Use this basic formula:

ROI (%) = (Net benefit from chatbot - Cost of chatbot) / Cost of chatbot x 100

A worked annual example for a mid-sized support bot:

Your real figures will vary with volume and timeframe, but the pattern holds: for support automation, the payback is usually fast. For a detailed walkthrough, see how to calculate chatbot ROI.

How to lower your chatbot costs

A few habits keep the bill in check without sacrificing results.

• Start small. Launch a minimum viable bot that solves one or two core problems well, then expand based on real usage instead of paying for features nobody uses.
• Try no-code first. For a first version, a no-code AI platform removes development cost and launches in days. Many, including Quickchat AI, have a free tier.
• Reuse existing APIs. Before commissioning custom integration work, check whether your systems already offer APIs or webhooks. Using them is cheaper and faster than building new connections.
• Right-size your plan. Review usage regularly. If you consistently use a fraction of your tier, downgrade; if you keep hitting the limit, a higher tier is usually cheaper than overage.
• Negotiate at volume. For large deployments, ask about volume discounts and custom enterprise pricing, which is often not listed publicly.
• Streamline conversations. Tight, well-designed flows resolve questions faster, which matters most on usage-based pricing. For more, see how to scale customer support efficiently.

Frequently asked questions

How much does an AI chatbot cost per month?

You can start free: several platforms include a free plan, and Quickchat AI’s free plan gives you 50 AI replies a month with no card. Paid plans commonly run from $9 to $999 a month depending on how many replies you need and which features you unlock. Enterprise and usage-based plans are billed per resolved conversation, often from around $0.50 each. Building your own from scratch costs far more: $5,000 to $30,000 for a simple bot and $75,000 to $500,000 or more for a custom AI bot, plus ongoing model and hosting bills.

Is there a free AI chatbot plan?

Yes. Free plans are common and are enough to test a bot and run a low-volume one. Quickchat AI’s free plan includes 50 AI replies a month, up to 50 knowledge base articles, all channels except the API, and unlimited AI Actions, with no credit card required. The main limits are the monthly reply allowance, the standard AI model only, and platform branding on the widget.

How much does it cost to build your own chatbot from scratch?

A simple rule-based bot costs roughly $5,000 to $30,000 to build. An AI bot with natural language understanding ranges from $75,000 to $500,000. Complex generative AI bots can exceed $150,000 and reach $1 million or more. On top of the build you pay ongoing costs: LLM API usage per conversation, hosting and a vector store, plus maintenance and retraining.

What makes a chatbot more expensive?

The biggest drivers are reply and conversation volume, the AI model you use, the number of channels, the integrations and AI Actions you connect, and time spent on human handoff. If you build your own, engineering and ongoing maintenance dominate the cost. On a platform, the cost usually tracks how many AI replies you send each month.

How much does a chatbot cost for 1,000 conversations a month?

On a subscription platform, 1,000 conversations a month usually falls inside a mid plan. On Quickchat AI, a conversation is several AI replies, so 1,000 conversations maps to a plan in the $99 to $299 range (3,000 to 9,000 AI replies a month). On per-resolution pricing, 1,000 resolved conversations at roughly $0.50 to $2.00 each is about $500 to $2,000 a month. A do-it-yourself build adds LLM API and hosting on top.

Does testing the chatbot cost extra?

On usage-metered platforms, testing your own agent usually spends the same reply or credit allowance a real customer would. Every AI reply counts, including the ones you generate while trying it out, so heavy testing eats into your monthly quota. Budget a little headroom above your expected customer volume for testing and tuning.

Are AI chatbots more expensive than rule-based bots?

Generally yes to build, because AI adds natural language understanding and model costs. A rule-based build starts around $5,000 while an AI build starts around $75,000. On subscription platforms the gap is much smaller: modern no-code AI platforms start free or from a few dollars a month, so buying an AI bot is often cheaper than building even a simple rule-based one.

Is it cheaper to build or buy a chatbot?

For almost every small and mid-sized business, buying a subscription is far cheaper upfront and faster to launch, often days instead of months. Building your own only pays off at very large scale or when you have unique requirements a platform cannot meet, because you take on engineering salaries, hosting, and maintenance yourself.

How much does a chatbot cost per conversation?

Usage-based and per-resolution pricing commonly runs from about $0.50 to $6 per resolved conversation, depending on the platform and the complexity of the request. Intercom’s Fin bills $0.99 per resolution, Zendesk bills roughly $1.50 to $2.00 per automated resolution, and Quickchat AI’s Enterprise plan starts from $0.50 per resolution.

What hidden fees should I watch for in chatbot pricing?

Watch for overage or per-resolution charges above your plan allowance, the fact that testing consumes your allowance, charges for premium support or specific integrations, setup and onboarding fees, extra seats, and data or analytics retention limits. If you build your own, the easy-to-miss costs are ongoing maintenance, model API bills, and retraining.

Conclusion

The cost of an AI chatbot spans from free to seven figures, but for most businesses the practical range is narrow and affordable. You can start for free, most small and mid-sized companies run a capable support agent for $29 to $99 a month, and the cost tracks something simple: how many replies your AI sends. Building your own only makes financial sense at large scale or with genuinely unique requirements.

The best chatbot is not the most expensive one. It is the one that meets your goals at a price that fits, and the fastest way to find that number is to build one and watch what it uses. If you want to make the internal case first, the business case for chatbot adoption covers how to frame the ROI argument. When you are ready, you can start for free and count the costs yourself, or talk to us about a custom build.