$50 free credit for new accounts - ends in

Claim $50

Comparison

Botpress pricing compared: what it costs for support teams

Priya NairHead of Customer Experience
9 min readAug 4, 2026

You are pricing out Botpress for your support team. The landing page shows a free tier and a friendly per-message cost. You jot down a few numbers and think, “This looks reasonable.”

But then you start adding up what it really takes to get a support bot live. A designer builds conversation flows. A developer hooks it up to your website and help center. Someone tests answers and updates them when your product changes. You realize a per-message price is only the ticket to enter – the total ride is longer.

If you pick a tool without accounting for those hidden costs, you risk slower answers, more agent workload, and a bot that frustrates customers instead of helping them. A fair comparison needs to start with how the pricing actually stacks up.

How Botpress pricing works

Botpress offers a free plan with a set number of included messages. After you pass that limit, you pay for each extra message you send and receive. Higher-volume teams move to a paid plan or an enterprise tier. Pricing is usage-based, so your bill grows with every chat.

The exact figures change over time, so we will not quote numbers here. The key concept: you pay for messages consumed, whether the bot answers correctly or not. You also need to build and host the bot yourself. That means you carry the engineering cost of creating the conversational logic and keeping it accurate.

For a lean support team, this model can feel like buying a car by the mile but being expected to assemble the engine, too.

Where the real costs show up

A per-message price looks clean on paper. The less visible cost lives in the work behind it.

Building the bot. Botpress is a conversational AI platform. Its strength is letting you design complex dialogue trees. That flexibility comes with a time investment. You or a teammate must map out every path a customer might take, write the replies, and test them.

Keeping answers accurate. Products change, policies shift, and prices update. Each change means someone revisits the conversation flows. With manual upkeep, bots drift. A drifting bot gives outdated answers, and outdated answers push customers to contact your team anyway.

Human handoff gaps. When a customer asks something outside the bot’s flow, you want a person to step in instantly. Botpress lets you build a handoff, but you have to wire it up to your live-chat system and monitor it. That work adds to your support stack, and it adds to your cost.

Hosting and maintenance. Botpress runs on your own infrastructure or a self-hosted server unless you use their cloud. The free tier is self-hosted, so you pay for compute, uptime, and security patches. Some teams are comfortable with that; many support leaders simply want the tool to work.

How Chatref pricing works differently

Chatref follows a different philosophy. You buy prepaid credits and pay only for what you use. There are no seats to license, no hosting bills, and no charges for teammates who log in to watch chats or take over.

A few credits cover a conversation that resolves a question on its own. Chatref charges per resolved interaction – not per message ping-pong. If a chat needs a human, the human jump-in is included. You do not pay extra for the handoff.

Because Chatref is built for support teams, the heavy lifting is done for you. You add a snippet to your site, upload your knowledge (docs, site pages, FAQs), and the agent learns your business. It answers in your brand’s voice from that content. You do not design dialogue trees. You do not manage cloud servers. You do not pay for seats.

Pricing model: per-message versus credits

Botpress ties your cost directly to message volume. Every “hello,” every follow-up, and every clarification burns a message. Support chats are messy – customers rephrase, add details, and ask something else while waiting. Message-based pricing can surprise you at the end of the month.

Chatref uses a credit system. You know the cost of a resolved conversation upfront. Credits are deducted when the AI gives a complete, accurate answer or when a human takes over smoothly. You can forecast your monthly spend based on how many support queries you typically handle, not on how chatty your customers are.

For a team that pays their support agents per hour, a pricing model that mirrors resolved work feels more intuitive. It aligns cost with the outcome: a question answered.

Where Chatref fits: support that learns from your content

Chatref is not a general chatbot builder. It is an AI support agent that learns your business. You add a chat to your website or connect it to Slack, email, and WhatsApp. It answers from your own knowledge base, not from generic training data. So answers stay factual.

When a question goes beyond the agent’s scope, a real person joins the live chat with full context. There is no need to build that bridge yourself; it is part of the platform. The agent also works across all channels so you only train it once.

This focus on support means a different kind of pricing clarity. You pay for the value of a resolved interaction, not the engineering hours to set it up or the infrastructure to keep it running.

When Botpress makes sense

Botpress shines if you need deep conversational logic – guided workflows, multi-step forms, or custom routing that changes based on user input. A developer can build conversations that feel like a native app inside the chat window. If that is your primary use case and you have a team comfortable with coding, Botpress gives you extraordinary control.

It also fits organizations that already run their own servers and want a self-hosted product with total data sovereignty. The pricing model may work if you can predict message volume accurately and your bot handles simple, high-volume requests where per-message costs stay low.

When Chatref fits better

If your goal is to deflect support tickets without building a bot from the ground up, Chatref is the more direct path. You train it on the content you already have – help articles, onboarding guides, product data – and it begins answering customers in minutes.

No per-seat fees mean your entire team can use the shared inbox, watch chats live, and step in when needed. A sudden spike in support queries does not spike your infrastructure cost because the pay-as-you-go credit system is predictable. And because answers come from your own material, the quality does not depend on a team member remembering to update a flow chart.

Chatref also fits teams that want one agent working across web, email, Slack, and WhatsApp from day one. That omnichannel scope is part of the product, not a paid add-on.

A quick comparison of cost drivers

Cost driverBotpressChatref
Core pricingFree tier, then per-message paid plansPrepaid credits, pay per resolved interaction
Team accessUsually per-seat (varies by plan)No per-seat fees
HostingSelf-hosted or cloud (self-hosted is free)Fully hosted, included
Bot-building effortMust design conversational flowsLearns from your content; no flows to build
Human handoffMust be custom-built or integratedBuilt-in live takeover with context
OmnichannelRequires separate connectorsIncluded across web, email, Slack, WhatsApp
Multilingual supportCan be configured, depends on flowsAutomatic in 11 languages

Choosing without overpaying

The right decision comes down to what problem you are solving. If you need a no-code conversational designer and have the time to craft every interaction, Botpress is a strong tool. Its free self-hosted tier can be an affordable starting point for technically minded teams.

If you need a support agent that hits the ground running and keeps costs aligned to resolved chats, Chatref offers a simpler path. You are not paying for messages that go in circles. You are not buying seats for every supervisor who wants to see the inbox. And you are not maintaining a server when you could be improving your help content.

Key takeaways

  • Botpress charges by message, which can become unpredictable when chats involve many back-and-forth turns.
  • The hidden cost of building and maintaining conversation flows often matters more than the per-message price.
  • Chatref uses prepaid credits per resolved interaction, with no per-seat fees and no hosting costs.
  • A support-first tool like Chatref learns from your existing content, so accuracy stays high without manual flow updates.
  • Human handoff is included in both tools, but Chatref builds it into the platform so you do not need extra integration work.

Frequently asked questions

Does Botpress charge per seat or per message? Botpress primarily bills based on the number of messages your bot sends and receives. Some paid plans may include team seats, but the main cost driver is message volume, not how many people manage the bot.

Can I try Chatref without paying? Yes. Chatref offers a free start so you can add it to your site, upload your knowledge, and test how it answers customers. You only add prepaid credits when you are ready to go live with higher volume.

Which tool handles human handoff more easily? Chatref includes a shared inbox and built-in live takeover. A person can jump into any active chat with full context at no extra cost. Botpress allows handoffs but requires you to set up the integration and monitor it yourself, which adds setup time.

Is Chatref suitable for high-volume e-commerce support? Yes. The credit system stays predictable even during seasonal spikes, and the agent can answer across web, email, and WhatsApp from a single knowledge base. Multilingual support lets you help customers in 11 languages without extra translation work.

What if I already use Botpress and just want to compare? This comparison is meant to help you weigh the real total cost. If Botpress fits your workflow and you can manage the upkeep, it remains a solid choice. If you would rather shift spend toward resolved interactions and less maintenance, Chatref is worth a look.

Moving your support forward should not feel like a guessing game. Start with the tool that gives you a clear picture of what you will pay and how fast your team can get value. Start free and see how Chatref answers your customers from your own content.

Priya Nair · Head of Customer Experience

Priya has spent over a decade helping support teams answer faster and stress less. She writes about the day-to-day of great customer support and how AI can carry the load.

Try this in your own workspace.

The best way to learn is to build as you read. Start free and follow along.