The AI capability of 'suggested replies' works by analyzing and generating embeddings vectors for each ticket, knowledge base article, and scraped website data. These vectors are stored in a vector database. When a query or new ticket comes in, it is compared against the embeddings vectors to find relevant data. The relevant data is then passed to ChatGPT, an AI language model, along with the original question. ChatGPT provides an answer based on the attached data, which is presented as a suggested reply for the agent.
Use case 1
For landscape design and construction services companies, the AI-powered suggested replies can assist in answering specific project-related queries. For example, if a customer asks about the best plants for a particular type of garden, the system can suggest a reply based on previous knowledge base articles or successful project recommendations.
Use case 2
Another use case is when customers inquire about the estimated cost of a landscaping project. The AI-generated suggested replies can provide agents with accurate and consistent responses based on historical data, ensuring transparency and facilitating informed decision-making for potential clients.
Use case 3
The 'suggested replies' feature can also be utilized to address common maintenance or troubleshooting questions from customers. By leveraging previous ticket resolutions and knowledge base articles, agents can quickly provide helpful suggestions or step-by-step instructions, improving customer satisfaction and reducing resolution times.
