Help
Every question we get asked, answered at length
If yours is not here, ask us. We would rather answer it once properly than have you guess.
Looking for setup instructions rather than answers about the product? The documentation covers the step by step.
Will it give my customers the right answer?
- Will it make things up or give a customer wrong information?
- It answers from the content you give it and shows which document an answer came from. Anything structured, such as prices, stock, or order details, is a separate case: those render only from the response your own system returns, and the model is prevented from writing those fields itself. When it cannot answer from your content or your systems, it says so and hands over to your team.
- Our documentation is scattered and partly out of date.
- Start by pointing it at your live website, usually the most current thing you have. Add PDFs and Word documents later. It only answers from what you give it, so the question is what to include rather than getting everything perfect first.
- What about customers who write in another language?
- The assistant replies in the language the customer writes in, across common languages. Your source content can stay in one language.
What can it connect to?
- Can it look up real data, not just answer from documents?
- Yes. Connect your CRM, order system, booking tool, or internal database as a tool and the assistant calls it when a question needs live data. The customer sees a short status line while the lookup runs. This is available on Starter and above.
- Will it work with the systems we already use?
- Anything with a REST API or an MCP server, which covers most modern software including the internal tools nobody else supports. We are not waiting to add your stack to a roadmap. Send us your API documentation and we will tell you.
- How does it go on our website?
- One script tag, which works on WordPress, Webflow, and any hand-built site. You set the colours and greeting, and list the domains allowed to load it so nobody can embed your assistant elsewhere.
What does setup involve?
- How long does setup take?
- There are two jobs. Getting it answering from your content is quick, because it can read your existing website rather than waiting for you to gather files. Connecting a live system takes longer and depends on your API, usually an afternoon. We do that part with you.
- Do we need a developer?
- Not for the content side. For connecting a system you need the API details, which your vendor or engineering team can supply. Agent Builder also lets an admin set things up by describing what they want in plain language, and it asks for confirmation before saving anything.
Plans and data
- What happens if we go over the monthly message limit?
- Extra messages are billed per message rather than cut off, so your assistant keeps answering customers. You can watch usage against your plan in the dashboard at any time.
- Is our data used to train AI models?
- No. Your documents and customer conversations are not used to train public models.
Also asked
- Do we have to use your built-in AI, or can we run our own automation behind it?
- Chatevo's own AI is what reads the conversation, decides what to say, and decides when to call a tool, every time, on every plan. What you can change is what happens on the other side of that decision: connect a tool or a webhook to your own n8n, Make, or Zapier workflow, and Chatevo hands that workflow the job once it decides the job needs doing. That is different from replacing our AI with your own agent as the thing driving the conversation. We do not support that today. If that is what you need, tell us on the waitlist form and it will shape what we build next.
Still not answered?
Ask us directly. A person reads every message, and the answer usually ends up on this page for the next person.