Skip to Content
The Makinari API repo now includes an MCP Server — connect your AI models directly. View on GitHub →
GuidesWebsite: Chat & Tracking

Website: Chat & Tracking

Integrating Makinari into your website allows you to track visitor behavior, identify leads, and provide automated customer support via a live chat widget.

When to use

  • You want to see which pages leads are visiting before they buy.
  • You want an AI support agent to answer customer questions 24/7.
  • You want to trigger outbound campaigns based on website behavior.

Step 1: Install the Tracking Script

The Makinari tracking script is a lightweight JavaScript snippet served from our CDN.

  1. Locate your Site ID in the Makinari dashboard.
  2. Add the following snippet just before the closing </head> tag on all pages of your website:
<script src="https://files.uncodie.com/tracking.min.js" data-site-id="YOUR_SITE_ID"></script>

Once installed, the script will automatically track page views and session duration. You can view this data in the Dashboard or via the Visitors API.

Step 2: Configure Allowed Domains (Important)

For security, Makinari’s APIs use strict CORS validation for browser requests. The chat widget and tracking script will fail with a 401 Unauthorized or CORS Error if your domain is not authorized.

  1. Go to Settings > Channels > Website in the Makinari dashboard.
  2. Add your website’s domain (e.g., https://www.yourcompany.com) to the Allowed Domains list.
  3. Do not include trailing slashes or subpaths.

Step 3: Enable the Chat Widget

  1. In the same Website settings panel, toggle the Chat Widget to Active.
  2. Customize the primary color, welcome message, and agent avatar to match your brand.
  3. The tracking script you installed in Step 1 will now automatically render the chat bubble in the bottom right corner of your site.

When a visitor interacts with the chat, the messages are routed to your Customer Support agent, who uses your Brand Context to answer questions.

Identifying Users Programmatically

If your website has a login system, you can attach the Makinari session to your known user data.

// Call this after the user logs in window.MarketFit.identify({ email: 'user@example.com', firstName: 'Jane', lastName: 'Doe' });

See the Visitors API Reference for more details.

Last updated on