How Much Does a Chatbot Cost? (2026 Pricing Guide)

How much does a chatbot cost? In 2026 the honest answer is a range, because a simple menu bot that answers five canned questions and a custom AI assistant that reads your knowledge base and books appointments are not the same product. A basic off-the-shelf bot can run as little as $0 to $50 a month, while a custom AI chatbot trained on your business typically costs $2,000 to $15,000 to build plus $50 to $500 a month to run. In this guide I will break that down by chatbot type, separate the one-time build from the ongoing monthly costs, explain what an LLM actually adds to the bill, and show you how to pick the cheapest path that still does the job. I work with clients across the US, Europe, and Israel, and these numbers reflect what an experienced freelancer charges rather than agency pricing.

How much does a chatbot cost by type

The biggest factor is whether your bot follows fixed rules or actually understands language. Here is the spread I see, with the build cost and the realistic monthly running cost side by side, because quoting one without the other is misleading.

To put real money on it: a rule-based bot on a platform like Intercom or Tidio can be live for the price of a subscription, around $30 to $100 a month (about 110 to 370 ILS), with little or no build cost. A standard custom AI chatbot that answers from your own content and hands off to a person when stuck usually lands around $4,000 to build (about 15,000 ILS) and $100 to $200 a month to run. An advanced assistant that takes actions like booking a meeting or updating your CRM starts around $10,000 and climbs with volume and complexity.

Rule-based vs AI/LLM chatbots

This is the fork that decides most of the price, so it is worth understanding clearly.

Rule-based bots

A rule-based bot follows a decision tree you design: if the visitor clicks this, show that. It is cheap, predictable, and never says anything you did not script. The trade-off is that it breaks the moment someone phrases a question in a way you did not anticipate, and building a tree that covers real conversations gets tedious fast. For a small site that just needs to route visitors and capture leads, a rule-based bot is often all you need and the cheapest option by far.

AI/LLM bots

An AI bot uses a large language model to actually understand what the visitor typed and answer in natural language, usually grounded in your own documents so it does not make things up. It handles the infinite ways people phrase the same question, sounds human, and works in multiple languages. The cost is twofold: a higher build, because grounding it in your content and keeping it accurate is real engineering, and a per-message running cost, because every conversation calls the LLM. If you want the deeper picture of what these bots can do, I cover it in my guide to adding an AI chatbot to your website.

The monthly running cost, explained

The build price is only half the picture. An AI chatbot has real per-month costs that a rule-based bot mostly does not, and ignoring them is the most common budgeting mistake I see.

• LLM usage fees: every message the AI answers costs a small amount in tokens. At low traffic this is a few dollars a month; a busy support bot handling thousands of conversations can run $100 to $500+ a month. This is the cost most people forget.
• Platform or hosting: an off-the-shelf tool charges a flat $30 to $300 a month. A custom bot needs hosting, usually $10 to $50 a month at small scale.
• Knowledge base updates: when your products, prices, or policies change, the bot's source content needs updating so it stays accurate. Budget a little time or a small retainer for this.
• Maintenance: models get deprecated, integrations change, and edge cases surface. A realistic figure is 10 to 20 percent of the build cost per year.

I always quote the build and the expected monthly cost together. A cheap off-the-shelf bot with a flat fee can be the better deal at low volume, while a custom bot that runs on usage-based pricing wins once you control the model and the prompt. The right answer depends on your traffic.

What drives the price up or down

Two chatbots that sound identical on a call can differ in price by 5x. Here is what actually moves the number, roughly in order of impact.

• Rule-based vs AI. The single biggest lever. Understanding language costs more to build and to run than following a script.
• Actions vs answers. A bot that only answers questions is one price. One that books a meeting, creates a ticket, or updates a record is doing real work and crosses into AI agent territory, which adds scope.
• Knowledge base size and accuracy needs. Grounding a bot in a handful of FAQs is quick. Keeping it accurate across hundreds of documents, with no hallucinations on pricing or policy, takes real engineering.
• Integrations. Connecting to your CRM, helpdesk, calendar, or order system each adds build and ongoing maintenance.
• Channels. A website widget is one build. Adding WhatsApp, Messenger, and SMS multiplies the work.
• Languages. A single-language bot is simpler than one that switches between Hebrew, English, and more cleanly.
• Human handoff. A bot that knows when it is stuck and routes to a live agent is more robust, and more work, than one that just guesses.

Off-the-shelf vs custom

Where you get your bot matters as much as what it does. Each path has an honest place.

Off-the-shelf

Platforms like Intercom, Tidio, and many others give you a working bot for a monthly fee, no build required. For a simple FAQ or lead-capture bot, this is a legitimate and cheap start. The trade-offs show up as you grow: per-seat or per-conversation pricing that climbs with success, limited control over how the AI behaves, and your bot living on someone else's rules. If your needs are basic, the simplicity is worth it.

Custom

A custom bot is the right call once the off-the-shelf options cannot answer accurately from your specific content, cannot take the actions you need, or get too expensive at your volume. You own it, you control exactly how it behaves and what it costs to run, and there is no per-conversation tax on your growth. The shift that changed my 2026 pricing is that AI-assisted development has collapsed the timelines. A custom bot that took weeks a couple of years ago now ships in days, so custom is no longer automatically the slow, expensive path. The honest limit: the tools make a good engineer much faster, but knowing how to keep a bot accurate and what to leave out still comes from experience.

So, how much does a chatbot cost for you?

For most businesses in 2026, the realistic answer is somewhere between free for a simple rule-based widget and $25,000 for an advanced AI assistant that takes real actions. Most of the custom projects I take on land in the $2,000 to $8,000 build range with $50 to $300 a month to run, because that is where a bot that genuinely understands your customers and answers from your own content delivers the best return. The right number is the one that matches what the bot actually needs to do, with the monthly cost factored in from the start. If you want to compare the build numbers against a broader automation project, my guide to how much business automation costs puts it in context, and you can get a rough figure in minutes with my project cost estimator.

If you want a straight, no-pressure estimate for your specific case, book a call and tell me what you want the bot to handle. I will give you an honest range, the expected monthly cost, and the leanest path to get there. You can also reach me through the contact form.