Guide for WooCommerce AI Chatbot

Woocommerce AI Chatbot, powered by the unification of OpenAI SDK Compatible LLMs integrates seamlessly into your Woocommerce Store, providing exceptional customer support and assistance.

Woocommerce AI chatbot becomes an integral part of your store’s front end, offering a user-friendly and interactive interface for customers to engage with.

Customers can explore your product offerings, ask specific queries about shipping details and discount coupons, and receive relevant information.

Backed by the Unification of OPEN AI SDK Compatible LLMs, the Woocommerce AI Chatbot excels in understanding and responding to customer inquiries.

Utilizing advanced RAG techniques, the chatbot analyzes customer inputs and their sentiments then delivers comprehensive answers using AI Agents.

Elevate your customers’ shopping experience with the Woocommerce AI Chatbot powered by large language model.

If you also want to provide the customers on your WooCommerce store to search products using images and phrases, then you can check WooCommerce AI Image & Semantic Search.

We provide a production version of the plugin. If you want a development version then you can get it at an additional cost. Please contact [email protected] for it.

Check the below – mentioned video for an overview of the workflow of the plugin:

• Upload WooCommerce product data to the Vector database to train the chatbot.
• Show the AI chatbot on multiple WooCommerce storefront pages.
• Conversational AI and Natural Language Processing (NLP).
• OpenAI’s latest GPT models and Gemini Flash 3 provide more contextually relevant responses.
• The plugin now supports all the latest OpenAI SDK-compatible LLM Models but outperforms on ChatGPT, GEMINI, and Cerebras models.
• Customize the WooCommerce chatbot appearance with your brand.
• Admin can create AI-generated content for the posts, pages, and products from the backend. 
• Admin can view chat history between the Chatbot and the users and guest users by enabling guest history on or off.
• Product long descriptions will be created by the admin while adding new products.
• Admin can see a live dashboard that delivers real-time insights into its performance and the most-asked product.
• This module is compatible with WooCommerce’s new High-Performance Order Storage (HPOS) feature.
• It offers three types of embedding models: OpenAI, GEMINI, and ONNX embedding models (e5-small) for local embedding generation.
• It also supports two types of vector databases, Pinecone for hosted and chromadb for local vector DB.
• Now they also respond about shipping details and discount coupon availability.

The user will get a zip file, which he has to upload to the “Add New” menu option in the WordPress admin panel.

For this login to the WordPress Admin Panel and under the Dashboard, hover your mouse over the “Plugins” menu option. This brings out a Sub-Menu and then selects the “Add New” option.

After that, you will see an option on the top of your page that is “Upload Plugin”, click the option to upload the zip file.

Then click on the “Upload Plugin” option, below that you will see a button “Choose File” Click on the button to browse for the zip file as per the snapshot below.

After browsing the file, click the “Install Now” button to install the plugin as per the snapshot.

Now, when the plugin is installed correctly, you will see the success message. Additionally, an option to activate the plugin.

Click on “Activate Plugin” to activate the installed plugin.

To learn how to translate the module, you can click here.

NOTE: To generate JSON from your translations, save your translation file in the default WordPress languages directory.

Then run the following command in the same directory: Note `xx_YY` refers to the locale (eg. `en_US`)
“`wp i18n make-json ai-chatbot-for-woocommerce-xx_YY.po –no-purge

The admin has to flush the permalink by navigating to Settings > Permalinks so that the links to the chatbot store work correctly.

1) Pincone Settings

To configure the PineCone, the admin will navigate through Webkul WC Addons > Chatbot > Settings > Vector DB Selection> PineCone Settings. Below are the steps to generate the PineCone keys.

To create an account, the admin can continue with their Google, GitHub, and Microsoft account. Otherwise, enter your email to create an account on Pinecone.

After creating an account on Pinecone, to create a new API key click on API Keys > Create API key as shown in the below screenshots.

Enter the key name and click on the create key to create the new API key.

Copy and enter the environment and value key at the admin end. Please refer to the below image for a better understanding.

2) Chroma DB Settings:

To configure the Chroma DB, the admin will navigate through Webkul WC Addons > AIChatBot > Settings > Vector DB Selection> Chroma DB. Click on the Click Here to know how to setup chroma db locally.

Embedding Models Settings

1) Gemini Model Settings

To configure the Gemini Embedding model , the admin will navigate through Webkul WC Addons > AIChat Bot > Settings > Model Settings > Select Model Type > Gemini Model >. Click on the Click Here to know how to generate the Gemini API keys.

2) Open AI Model Settings

To configure the Open AI Settings, the admin will navigate through Webkul WC Addons > AIChat Bot > Settings > Model Settings > Select Model Type >OPEN AI. Below are the steps to create OpenAI keys.

To create an account click on the get started button, or to be login into your account click on the log in button.

Click on the API option to integrate OpenAI models into your AIchat Bot.

Click into your account. Now, click on view API keys as shown in the below screenshot.

To create a new Open AI secret key, navigate through view API keys > user > API keys > create new secret key.

Enter the name of the secret key or you can skip the key name. Click on Create Secret Key to generate to new secret key.

Copy the Secret key and enter it at the admin backend. Please refer to the below screenshots for a better understanding.

3) Onnx Model Settings

To configure the Onnx Embedding model, the admin will navigate through Webkul WC Addons > AI Chatbot > Settings > Model Settings > Select Model Type > Onnx Model >. Onnx model is already loaded with our plugin , so there is no need to configfure anything.

To configure the chat Box, the admin needs to navigate through the webkul wc Addons > AIChat Bot > Settings > chat box settings.

• The admin can allow voice assistant or animation in the chatbot.
• Set the primary and secondary colors of the chatbot box.
• Set the chatbot name.
• You can also enable or disable the guest history.
• Upload the chatbot avatar, background image, and AI logo as per choice.
• Allow the background image for the admin conversation history.
• Admin can set the right or left position of the chatbot on the store.

Data Training

The framework allows the admin to upload their Woocommerce product data into the Selected Vector database and then train your AI Chatbot.

Also, whenever the admin updates the product, the data also gets updated through selected Models.

Admin navigates through dashboard > products > add new product can also generate product long descriptions with AI Chatbot while adding new products to the store.

• Admin has to enter the name of the product.
• Click on the long description with the AI chatbot under the product description to generate the description of the product.
• After the product description is generated by an AI chatbot, the admin can edit the generated description.

AI ChatBot Dashboard

Admin can access the real-time dashboard. Where the admin can view user and AI weekly and monthly messages over chatbot in line/bar/area/scatter format.

Admin can see the daily users over the chatbot and most asked products as shown in the below screenshots.

Conversation History

The admin can access the chat history between customers and AI to gain insights into what customers are searching for after landing on the storefront.

• The admin must enter at least three characters of the user’s name to retrieve the chat history.
• The admin can also select the date from which they want to view the conversation.
• They can download the chat history if required.
• All the guest user history you can find using the guest username
• Additionally, the admin can block a user by clicking the block option, as shown in the screenshots below.

If you want to facilitate your customer to ask order-related queries to the admin, then you can make use WooCommerce Order History Communication.

Content Creation

The module helps to generate accurate responses based on the provided WooCommerce data.

Admin can creative and high-quality content generation. Configured model will generate human-like text, answer questions, and translate languages.

Admin can also generate custom content for a number of topics. Admin has to enter the topic name and can use this module for topic suggestions.

Similarly, use this module to generate sections, content excerpts, and meta descriptions for posts, pages, and products.

After creating the content, admin can the view and edit the generated post in post > all post.

The customer can explore your product offerings, ask product-specific queries, and receive relevant information about the product.

WooCommerce AI Chatbot provides real-time answers to customer’s product queries and displays the answers in the chat window as shown in the below screenshots.

The customer can use the voice assistant to ask questions about the product.

The AI Chatbot understands and responds to customer inquiries as shown in the below screenshots.

If you also want to support your customer over WhatsApp, then you can make use of the WooCommerce WhatsApp Chat Plugin.

Thus, that’s all for the WooCommerce AI Chatbot  module. If you still have any issues, feel free to add a ticket and let us know your views on our Webkul Support System.