Conversational Interface for Product Discovery Using Generative AI (LLM)

Experience a natural, engaging way to explore products with our conversational interface. Powered by a robust, generic foundation model through advanced Generative AI (LLM, e.g., Microsoft Azure OpenAI, GPT-4-Series), the Copilot understands customer queries, intentions, and preferences. This versatility allows for personalized product recommendations and assistance tailored just for the customer.

Integration with SPARQUE.AI Recommendations and Search

Leveraging the powerful capabilities of SPARQUE.AI, the Copilot provides highly accurate search results and personalized product recommendations. This integration ensures that customers find exactly what they are looking for quickly and efficiently, enhancing satisfaction and conversion rates.

Integration with SOLR Search

Quickly introduce and integrate the Copilot with Solr search, offering fast and relevant search results with minimal cost and risk. This ensures a smooth transition for businesses relying on Solr, maintaining high standards of search accuracy and speed without significant investment.

Deep Integration Into the Intershop Progressive Web App (PWA) Frontend, Including UI Actions

Seamlessly integrated into the Progressive Web App (PWA) frontend, the Copilot can interact with various UI elements. This allows for a smooth and intuitive purchase experience where actions such as filtering results, comparing products, or navigating product categories are effortless and user-friendly.

Basket and Checkout Actions

Simplify the buying process with intelligent basket and checkout actions. The Copilot can add, remove, or update items in your shopping basket through conversational commands, making the checkout process more efficient and less cumbersome.

Order Template Actions

Products can be added to an order template for later use. Products can be ordered from order templates.

Access to User Order History

The Copilot provides access to the individual order history of the buyer, supporting fast reordering and reference to previous transactions. Users can specify a time frame to filter relevant orders. This feature is handy for repeat purchases.

Personalized System Prompts Based on User Context

The Copilot supports user-specific system prompts to enable personalized and context-aware assistance. This includes addressing users by name and adapting responses based on account-specific preferences, roles, or past interactions. The result is a more efficient and relevant user experience tailored to the needs of each purchasing professional.

Image-Based Product Recognition

Customers can easily find products by uploading an image. Powered by a versatile generic foundation model, the system accurately identifies and matches products from the image, providing you with reliable product information and recommendations. This feature streamlines the product discovery process, making it more intuitive and efficient for the customer.

Voice Interaction

Customers can engage with the system effortlessly using voice commands. The voice interaction feature understands and responds to spoken queries, offering accurate product information and recommendations. This hands-free approach enhances user convenience, making the product discovery process seamless and accessible for every customer.

Low-Code Customization Options

Tailor the Copilot to meet specific business needs with low-code customization options. This feature allows businesses to modify the AI’s behavior, responses, and integration points with minimal coding, ensuring the solution fits perfectly within their existing ecosystem.

This allows, for example, the integration of extended service functionality such as the scheduling of repairs or other service visits.

Central chat configuration for all channels: A central chat configuration ensures consistent AI settings across all channels while accommodating channel-specific products and content (CMS pages). For an additional fee, a fully customized chat configuration tailored to individual channels is also available.

Conversation Analytics

Gain valuable insights into customer interactions with comprehensive conversation analytics. Track and analyze chat histories, customer intents, and response effectiveness to continuously improve the AI’s performance and the overall customer experience.

Multi-Agent Integration

Enhance collaboration and task execution with Multi-Agent Integration. This feature allows the Copilot to interact and coordinate with multiple AI agents, each specialized in different functions such as customer service, technical support, and sales. By integrating these agents, the Copilot can provide more comprehensive and efficient assistance, ensuring that customer needs are met through the combined expertise of various AI capabilities.

This feature allows the integration of existing chat agents that may already be established in your organization.

Document-Based Query Assistance

Enable users to easily get answers to their questions based on additional documents such as web pages, setup guides, data sheets, or other knowledge bases. This feature allows the Copilot Assistant to respond directly to user questions based on their queries, providing precise answers and relevant information.

Image-Based Install Base Recognition
(additional option)

Identify installed base items, such as machines, by photo. This feature allows the assistant to provide specific actions around spare parts, consumables, and services for the identified install base item.

Identification of Products via External Product ID

When a purchaser provides a known product identifier (e.g., EAN) for an item not listed in the current catalog, the Copilot searches external sources, including competitor offerings, to identify the product. This identification suggests equivalent or compatible alternatives available in the shop. This supports efficient handling of substitution requests.

Intent-Based Filtering with Integrated PWA Results Display

The Copilot interprets user intent expressed in natural language to automatically identify relevant filter criteria—such as technical attributes, compatibility requirements, or usage context—and applies them to the product search. Filtered results are presented directly within the Progressive Web App (PWA) storefront for immediate review and action.

This feature includes Explainable AI capabilities: Users can see why specific products were recommended, including which attributes or filters were applied based on the conversation. This transparency ensures procurement decisions are traceable, prevents misconsulting, and builds trust in the AI-assisted process.