WebWhiz

WebWhiz revolutionizes AI customer support through website-content-driven chatbots, offering businesses an intuitive solution with transparent technical implementation. This comprehensive guide explores WebWhiz's capabilities, from its unique three-step chatbot creation process to its expansive self-hosting options, helping readers understand how this open-source platform transforms web content into precise support solutions.

Overview

WebWhiz's AI-powered customer support solution handles over 1,000 business cases across multiple industries. The platform automatically fetches website content to train its chatbot, requiring only a website URL to get started.

The three-step chatbot creation process begins with adding the website URL and specifying which pages to include or exclude in training. WebWhiz then uses this data to generate responses based on the website's content - a feature that differentiates it from typical chatbots which guess answers. The platform scans websites once per month by default, with optional more frequent scans upon request.

Powered by AI but with a focus on precision, WebWhiz's chatbot provides not just basic support, but also handles complex queries by training on specific website data. This results in reduced support volumes for businesses, translating into happier customers and lower support costs.

Pricing and Plans

WebWhiz offers four subscription plans with increasing capacity and features. The Basic plan starts at $19 per month with 2,000 allowed messages and five projects, supporting up to 100 pages. The Standard plan raises the message limit to 5,000 per month, increases project capacity to 10, and extends page support to 1,000. The Premium tier offers 10,000 messages, 100 projects, and 2,500 pages at $99 per month, while the Enterprise plan provides 40,000 messages, unlimited projects, and 10,000 pages for $349 monthly.

Each message in WebWhiz consumes tokens, with complexity-based token counts for different message types. If users exceed their monthly limit across projects and pages, their chatbots temporarily stop generating AI responses, displaying predefined messages until usage returns to within limits. For businesses requiring more customization, WebWhiz offers a $19 per month add-on that removes the "powered by WebWhiz" branding tag while maintaining compatibility with all plan levels.

Integration Methods

WebWhiz provides three integration options tailored to different deployment needs. The chatbot with launcher icon offers a simple click-to-chat interface, implemented via a <script> tag placed before the closing </body> tag. For more seamless integration, the embedded chat interface employs an <iframe> tag that can be positioned directly within web pages. The public link option allows sharing direct access to the chatbot via hyperlinks, enabling external communication or temporary visitor assistance.

Technical Implementation

The WebWhiz platform requires specific technical infrastructure to operate efficiently. Core components include an API server built with NestJS, a JavaScript Celery worker for handling data generation, and a Python Celery worker responsible for content extraction. The system leverages MongoDB as its primary database and Redis for caching mechanisms, with third-party services like OpenAI providing essential functionality for both chatbot operations and error monitoring.

Software Licensing

As an open-source project governed by AGPL 3.0, WebWhiz allows users extensive modification capabilities while requiring source code release for distributed modifications. For businesses seeking proprietary source code, a commercial license pathway exists. To use WebWhiz, users must meet basic technical requirements, including Docker and docker-compose installation, along with specific version requirements for MongoDB, Redis, and programming languages. The detailed installation process guides users through repository cloning, environment variable configuration, and service deployment, enabling both self-hosted and direct usage options.

Self-Hosting Options

WebWhiz offers both open-source and commercial licensing options for self-hosting, governed by AGPL 3.0. Under this license, users can modify the software for their own use but must release any distributed modifications under AGPL 3.0, including the full source code. For businesses requiring proprietary source code, WebWhiz provides commercial Team or organization licenses, which allow keeping modifications closed source while removing the "powered by WebWhiz" branding tag.

The self-hosted version of WebWhiz requires meeting specific technical prerequisites, including Docker and docker-compose installation, along with version requirements for MongoDB (v6), Redis (v7), and the programming languages used in the platform. Installation follows a structured process that begins with cloning the WebWhiz repository and configuring environment variables. The core system consists of multiple services defined in docker-compose.yml, including the WebWhiz UI, backend API, and required third-party services like OpenAI for chatbot functionality and error monitoring.

The chatbot integration process involves several key steps. Users begin by adding their website URL and specifying which pages to include or exclude in the training data. WebWhiz then employs its three-step chatbot creation process: entering website details, uploading PDF files (if needed), and finalizing chatbot creation. After the chatbot is created, users can access and customize various settings through the WebWhiz dashboard.

The technical infrastructure supporting WebWhiz includes an API server built with NestJS, JavaScript and Python Celery workers for data processing, and MongoDB paired with Redis for database operations. The platform's design enables scaling from small projects to large enterprise deployments, with documented installation procedures that guide users through both the development environment setup and the production deployment process.

Integration and Setup

WebWhiz's integration process begins with adding the website URL and specifying which pages to include or exclude in training. The platform automatically fetches and prepares training data from the specified website paths.

The chatbot creation process consists of several key steps. First, users navigate to the "Create new chatbot" section in the WebWhiz dashboard and select the "Website" tab. They then enter the full website URL, ensuring it uses the https:// protocol, and specify both included and excluded paths using forward slashes and commas.

After setting up the website details, users can upload PDF files through either drag-and-drop or manual file selection. Once uploaded, these files appear below the upload box and are used to further train the chatbot's understanding of the website content.

The final creation step involves double-checking the URL and paths before clicking "Create Chatbot." This triggers WebWhiz to fetch data from the specified paths and use it to train the chatbot for answering questions and handling user interactions.

After the chatbot is created, users receive a confirmation message and can proceed to customize various settings through the WebWhiz dashboard, including the chatbot's name, greeting messages, and response templates. The platform's three-step creation process enables quick setup while ensuring the chatbot receives precise training based on the website's specific content.

Technical Requirements and Installation

The WebWhiz chatbot infrastructure consists of three main components: an API server built with NestJS, JavaScript and Python Celery workers for data processing, and database operations handled by MongoDB paired with Redis caching. The system requires specific technical prerequisites, including Docker and docker-compose installation, along with version requirements for MongoDB (v6), Redis (v7), Node.js (v18 or above), Yarn, Python (v3.6 or above), and third-party services like OpenAI for chatbot functionality and error monitoring.

Installation follows a structured process beginning with cloning the WebWhiz repository and configuring environment variables in the .env and workers/.env files. Users must install frontend dependencies in the root folder and Python worker dependencies in the workers directory. The application is built using yarn, and services are managed with pm2. The system includes both a dashboard UI and backend API, accessible via localhost addresses.

The chatbot can be deployed through three primary methods. The simplest approach is integrating the chatbot with a launcher icon using a script tag placed just before the closing <body> tag. For deeper integration, an iframe can be used to embed the chatbot directly into web pages. Alternatively, businesses can provide direct access to the chatbot via shared links, which open in new browser tabs or windows.

To create a chatbot, users navigate to the WebWhiz dashboard, select "Create new chatbot" and choose the "Website" option. They then enter the website URL, including https://, and specify which paths to include and exclude. After setting up the website details, users can upload PDF files for further training. The final step involves reviewing the configuration and clicking "Create Chatbot," after which WebWhiz fetches and processes the specified data.

The platform's technical design enables scalability from small projects to enterprise deployments, with detailed documentation guiding both development environment setup and production deployment. Users have the option to self-host the software under AGPL 3.0 licensing, allowing modifications with the requirement to open-source distributed modifications. For businesses requiring proprietary source code, WebWhiz offers commercial Team or organization licenses that provide the flexibility to remove the "powered by WebWhiz" tag while maintaining closed-source development practices.