Digital Twin
Updated
Digital Twin is an LLM-powered chatbot that helps you resolve customer questions instantly and accurately by training on your personalized learning content, thus freeing up your team for high-value conversations.
Let’s have a look on how to setup Digital Twin on Sprinklr Self-Serve. Here we have listed down 6 steps:
Step 1: Personalize your Digital Twin
Define the personality of your Digital Twin by adding the following details below:
The list of fields in the Setup are listed below:
Click here to learn more on what all a Digital Twin can do.
Step 2: Setup Audience
Store and manage different types of audience to customize customer handling. Define the target audience for your Twin, ensuring that it caters specifically to their needs when engaging with customers from that particular group.
There are two types of Audience Segments:
Static: Static audience is a convenient feature that allows you to effortlessly upload your audience list in the form of a static Excel file.
Dynamic: Dynamic audience helps in generating an audience segment by utilizing the conditions specified in the Contact Properties. This segment continuously updates itself as new customers, belonging to the same segment, engage with your business.
Step 3: Train your Digital Twin
Importing Supported Files
Training a Digital Twin involves providing it with the necessary data, examples, and instructions to learn and improve its conversational abilities. In Sprinklr, you can import content from external files, public URLs, or your internal KB. You can import internal files formats such as PDF, DOCX, TXT, HTML, PPTX, XLSX, ZIP Up to 25.0 MB.
Adding quality content to the Digital Twin makes interactions more engaging and valuable for users. It provides relevant information, answers questions, and guides users through various tasks, ultimately improving their overall experience with the chatbot. Click here to learn more.
Importing External Content
You can train your Digital Twin by importing external content from public URLs or websites and uploading it into Digital Twin. Click here to learn more.
Sync from Sprinklr Knowledge Base
Syncing your Knowledge Base articles with the Twin involves integrating the content and information stored in the KB with the Digital Twin’s functionality. Digital Twin provides users with instant access to relevant information, streamlines support processes and enhances the overall user experience. Click here to learn more.
Setup Custom Answers
Custom Answers in Digital Twin are designed to provide users with accurate, helpful, and informative responses based on the specific context of their inquiry. These Custom Answers can be used to resolve specific questions a customer asks at the start of a new conversation, and Twin will prioritize them over its AI Answers. Click here to learn more.
Step 4: Tasks and Guardrails
Tasks which the Digital Twin will perform within the specified Guardrails.
Tasks are pre-established workflows that your target user follows when engaging with your Digital Twin. These workflows are tailored to your industry, such as the financial sector, as exemplified by the process of purchasing home loans.
Guardrails are predefined boundaries or limits set within the platform to guide users and ensure compliance with policies, regulations, or best practices. These guardrails serve to protect the integrity of data, maintain consistency in operations, and mitigate risks associated with misuse or improper configurations. Click here to learn more.
Step 5: Deploy your Digital Twin
Deploy your Digital Twin across Custom Domain, Social and Messaging channels and Live chat.
Set up your custom domain
Setting up a custom domain for an AI chatbot involves configuring the necessary infrastructure and settings to host and deploy the Twin on your own domain. Click here to learn more.
Deploy Twin on Social and Messaging account
Once you have trained your AI with relevant KB articles, external URLs, imported files, and added custom answers, you can deploy your Digital Twin across various social media and messaging channels for your potential customers. Click here to learn more.
Deploy on Live Chat
Sprinklr Live Chat, a customer service capability, helps brands provide immediate support to visitors on their website and mobile applications. It allows end customers to interact with brands via a messaging interface in real-time. Using Live Chat, contact centers can provide immediate customer support, resolve customer requests instantly, and improve customer relationships. Click here to learn more.
Step 6: Routing
Define routing to users when bot is unable to answer customer queries.
Routing refers to the process of handing over a user's conversation to the appropriate human agent when the chatbot is unable to handle the query or when the user requests to speak with a person.
This routing is often based on various factors such as the nature of the query, the availability and expertise of the agents, and sometimes even the user's history or preferences. You can define the routing conditions which will assign the tickets to an agent under defined circumstances. Click here to learn more.