Chatbase review

Chatbase is a no-code platform for building AI agents for customer support, sales, and product guidance. You set an agent’s role, connect information sources, and define guardrails, then test it against scenarios before deployment. The platform is designed to publish an agent across chat, email, voice, WhatsApp, and Slack from one dashboard.

Agents can use built-in Actions to call functions such as looking up an order or retrieving an invoice. Chatbase also includes a shared helpdesk inbox where AI and human agents can work from the same ticket queue. Its homepage lists analytics for conversation topics, sentiment, and trends; a model playground lets teams test agents against multiple underlying models.

The fit is broader than a website chatbot, but the pricing is credit-based. That makes plan limits and the cost of extra credits important to check before switching.

Chatbase’s main appeal is putting agent setup, deployment, actions, and human handoff in one product. Teams can build without code, test before publishing, and use the same agent across supported channels. The shared inbox is useful if support staff need to take over conversations rather than manage a separate queue. The homepage also lists sentiment and topic analytics, though it does not provide performance figures in the published material.

There are several entry points by team size. According to the pricing page, Hobby costs $40 monthly or $32 per month billed yearly and includes 700 message credits, 200 test credits, one agent, and two seats. Basic is $150 monthly or $120 on annual billing, with 4,000 message credits, three seats, helpdesk inbox, voice, API access, and 10 concurrent calls. Business costs $500 monthly or $400 annually and includes 15,000 message credits, five seats, advanced analytics, and 20 concurrent calls. A seven-day trial is listed.

That gives buyers a clear starting price and a path to more usage, but not unlimited capacity.

The plan caps matter. Hobby’s 700 monthly message credits and single agent may be restrictive for a live support team; Basic includes 4,000 credits and 10 concurrent calls, while Business includes 15,000 credits and 20 concurrent calls. The pricing page lists extra credits at $40 per 1,000, extra agents at $25 per month each, and branding removal at $99 per month. Enterprise pricing is custom, so larger deployments have no published price to compare.

A Reddit thread about site training reports that some sitemap links failed to train and that support did not resolve the issue. That is one user’s account, not proof of a general failure, but teams relying on broad site crawling should verify their own source coverage. Another Reddit post alleges high credit consumption and slow support; a commenter identifying as a Chatbase employee disputed the post’s credit claim while acknowledging support speed needed improvement. The published pricing details do not explain how credits map to each model or response, so confirm expected usage and support terms before committing.

Chatbase starts at $40 a month for Hobby when billed monthly, or $32 a month on annual billing. The annual price is a monthly equivalent and reflects a 20% discount; paying yearly means committing to the annual billing term. Hobby includes 700 message credits and one agent, but only two workspace seats.

Basic jumps to $150 monthly or $120 on annual billing. That buys 4,000 message credits, three seats, a helpdesk inbox, voice and API access, and 10 concurrent calls. Business is $500 monthly or $400 annually, with 15,000 message credits, five seats and 20 concurrent calls. Enterprise pricing is custom. The tier jump matters: Basic costs $110 more per month than Hobby at monthly rates, while Business costs $350 more than Basic.

Budget for add-ons too: extra agents cost $25 per agent monthly, extra storage is $25 per 10 MB monthly, and removing Chatbase branding costs $99 monthly. Auto-recharge is $40 per 1,000 message credits. There’s a seven-day trial, but no ongoing free plan.

Plan to configure your agent’s role, data sources and guardrails, then check its actions, integrations and target channels before launch. Chatbase says agents can be deployed across chat, email, voice, WhatsApp and Slack, with a shared inbox for AI and human agents. The published information doesn’t specify an import tool for agents from another platform, so budget time to recreate and validate those pieces rather than assuming a one-click transfer.

Before leaving, check what you can export: the published materials don’t publish details about conversation, agent or analytics exports, or how portable configured actions are. Confirm those points with Chatbase before moving customer data or rebuilding elsewhere. Reddit evidence about why users leave, migration problems or export limits wasn’t included in the published material, so we can’t responsibly attribute common complaints to users. For either direction, inventory your data sources, integrations, channels, helpdesk workflows and credit usage first; those are the items most likely to shape the work and replacement cost.