Odoo Live Chat lets you engage website visitors in real time — but you can't have a human sitting behind the screen around the clock. Odoo's built-in chatbot builder solves this by letting you create scripted conversation flows that greet visitors, answer common questions, qualify leads, and route complex issues to the right team member — all without writing a single line of code.
This guide walks through the full setup using only Odoo's native features: enabling Live Chat, building chatbot flows, connecting to CRM and Helpdesk, configuring escalation, and monitoring performance.
Step 1: Install and enable Live Chat
- Go to Apps and install the Live Chat module if it is not already active.
- Navigate to Live Chat → Channels and create a new channel (e.g., "Website Support").
- Add at least one human operator to the channel — this is the fallback agent when the bot cannot resolve a query.
- Under the channel's Widget tab, copy the JavaScript snippet and paste it into your website's
<head>section, or enable Live Chat on your Odoo Website via Website → Configuration → Settings → Live Chat. - Open your website in an incognito window to verify the chat widget appears in the bottom-right corner.
Step 2: Create a chatbot and design your welcome flow
Odoo's chatbot builder uses a step-by-step visual editor where you define what the bot says, what options visitors can choose, and what happens next. No scripting required — everything is configured through the UI.
- Go to Live Chat → Configuration → Chatbots and click New.
- Give the bot a name (e.g., "Support Bot") and upload a friendly avatar image.
- Add your first step as a Text message to greet the visitor — something like "G'day! Welcome to [Company Name]. How can I help you today?"
- Follow the greeting with a Question step that offers category buttons: Sales, Support, Billing, or General Enquiry. Each answer creates a branch you can build out independently.
- Keep the welcome flow to two or three exchanges before either resolving the query or branching into a more specific sub-flow.
Step 3: Build out FAQ and support sub-flows
Each branch from your welcome question should lead into a focused sub-flow. Think of these as mini decision trees that handle the most common queries your team receives.
- FAQ flow — Add a Question step listing your top five most-asked questions (e.g., "What are your delivery times?", "What is your returns policy?"). Each answer triggers a Text step with the relevant information.
- Support triage flow — Ask the visitor to describe their issue category (order problem, product defect, account access). Based on their selection, either provide a self-service answer or escalate to a human.
- Lead qualification flow — For the Sales branch, collect key details: what product or service they are interested in, their company name (via a Free Input step), and their email (via an Email step). This data feeds directly into CRM lead creation.
- Billing flow — Provide links to your online payment portal or invoice enquiry process, then offer to connect with the accounts team if needed.
- End every sub-flow with either a resolution message, a Forward to Operator step, or a Create Lead / Create Ticket action.
Keep scripted flows short — three to four exchanges maximum before either resolving the query or handing off to a human. Visitors abandon chats that feel like phone-tree menus.
Step 4: Configure trigger rules
Trigger rules control when and where the chatbot appears. Thoughtful triggers ensure visitors see the bot at the right moment rather than being pestered on every page.
- In your chatbot's configuration, go to the Trigger Rules section.
- Set page URL conditions to target specific areas of your site — for example, activate the Sales bot only on pricing and product pages, and the Support bot on the help centre and contact page.
- Configure a timer so the bot opens after a visitor has been on the page for a set number of seconds (10–20 seconds works well for most sites). This avoids interrupting visitors the moment they land.
- Use country-based rules to tailor the experience — for example, show Australian-specific shipping information to visitors from Australia.
- If you run multiple chatbots (e.g., one for sales, one for support), assign each to different Live Chat channels and use trigger rules to activate the right bot on the right pages.
Step 5: Connect to CRM for automatic lead creation
- Ensure the CRM module is installed. Odoo Live Chat has built-in integration that creates leads from chat conversations.
- In your chatbot flow, add a Create Lead step at the end of your sales qualification branch. This automatically generates a CRM lead with the visitor's name, email, and any details collected during the conversation.
- The full conversation transcript is attached to the lead as a note, giving your sales team complete context.
- Assign created leads to a specific sales team or salesperson based on the enquiry type or product category discussed in the chat.
- Verify the integration by running through the sales flow as a test visitor, then checking that the lead appears in CRM → Pipeline with all the expected details.
Step 6: Connect to Helpdesk for ticket creation
- Install the Helpdesk module if you use it for support ticket management.
- In your chatbot's support sub-flow, add a Create Ticket step that generates a helpdesk ticket when the bot cannot resolve the issue.
- Map the information collected during the chat (issue category, description, customer email) to the corresponding ticket fields.
- Assign tickets to the appropriate helpdesk team based on the issue type selected during the conversation.
- This ensures no support conversation falls through the cracks — every unresolved interaction becomes a trackable ticket with the full chat history attached.
Step 7: Set up escalation to human operators
- Add Forward to Operator steps at key points in your chatbot flows — after the bot has collected initial details but cannot resolve the query, or whenever the visitor explicitly asks to speak with a person.
- Ensure the full conversation history is visible to the human agent when they take over. Odoo preserves the chat transcript automatically, so the visitor never has to repeat themselves.
- Configure operator availability — if no human agents are online, the bot should collect the visitor's email address and either create a helpdesk ticket or CRM lead, then let the visitor know someone will follow up.
- Set your operators' working hours in their user preferences so Odoo correctly shows availability. For Australian businesses, make sure these reflect your AEST/AEDT time zone.
- Consider adding a "Talk to a human" option at the top level of your chatbot menu so visitors can bypass the scripted flow entirely if they prefer.
Step 8: Use canned responses and templates
Canned responses speed up your human operators when they take over from the bot. They also ensure consistent, accurate answers across your team.
- Go to Live Chat → Configuration → Canned Responses and create responses for your most common follow-up messages.
- Use a short keyword shortcut for each response — operators type the shortcut (e.g.,
:shipping) in the chat and the full response auto-fills. - Create responses for frequent scenarios: shipping timeframes, returns policy, business hours, pricing enquiries, and escalation to specialist teams.
- Include placeholder text that operators can quickly personalise (e.g., "Hi [Name], thanks for your patience. Let me look into that for you.").
- Review and update canned responses monthly to keep them aligned with current policies and promotions.
Step 9: Test the full conversation flow
- Open your website in an incognito window and start a Live Chat session as a visitor.
- Verify the chatbot welcome flow fires correctly — greeting, initial question, branching into sub-flows.
- Walk through each branch: Sales (lead qualification), Support (triage and ticket creation), Billing, and General Enquiry. Check that the right steps appear in the right order.
- Test escalation — select "Talk to a human" and verify the handoff works smoothly with the conversation history intact.
- Test the offline scenario — log all operators out and verify the bot collects contact details and creates a lead or ticket.
- Check that CRM leads and helpdesk tickets are created correctly with all the expected data and conversation transcripts.
- Test on mobile devices — the Odoo chat widget is responsive by default, but verify it works well on smaller screens.
Step 10: Monitor performance with Live Chat reports
- Navigate to Live Chat → Reports to access built-in analytics.
- Track key metrics: total conversations, bot-resolved vs. human-escalated, average response time, and customer satisfaction ratings.
- Review conversation transcripts regularly. Look for patterns: common questions the bot handles well, points where visitors drop off, and unnecessary escalations.
- Use the Sessions report to identify peak chat hours — this helps you roster human operators for the times they are needed most.
- Check operator performance metrics to ensure human agents are responding promptly when the bot hands off.
- Consider adding visitor ratings at the end of conversations. Enable this in Live Chat → Channels → [Your Channel] → Options.
Start with a limited scope: have the chatbot handle only your top ten FAQs and basic lead qualification initially. Expand its flows as you review real conversations and identify new patterns. A bot that confidently sends visitors down the wrong path is worse than no bot at all.
Best practices for Australian businesses
- Use Australian English — Set your chatbot messages to use local spelling (enquiry, colour, organisation) and terminology. Visitors notice when the language feels foreign.
- Display business hours in AEST/AEDT — When the bot mentions availability or follow-up times, always reference Australian time zones. Avoid ambiguous references like "business hours".
- Respect privacy obligations — If your chatbot collects personal information (names, emails, phone numbers), ensure your privacy policy covers live chat data collection. Under the Australian Privacy Principles, visitors should know what data you collect and how you use it.
- Include GST in any pricing mentioned — If your bot provides pricing information, always quote GST-inclusive prices as required by Australian Consumer Law for B2C transactions.
- Account for public holidays — Update your operator availability and bot messaging around Australian public holidays, including state-specific holidays if you serve particular regions.
- Don't over-automate — Australian consumers consistently rank the ability to reach a real person as a top priority in customer service. Make the path to a human operator obvious and easy.
A well-configured chatbot in Odoo Live Chat transforms your customer support from reactive to proactive — handling routine questions instantly while your team focuses on high-value interactions. Everything covered in this guide uses Odoo's built-in features, so there's no custom development to maintain or external services to pay for. If you need help designing conversation flows or fine-tuning your setup, get in touch.