Sisense Further Democratizes Data Insights with the Beta Launch of its GenAI-Powered Analytics Chatbot and Enhanced Narration Capabilities

Sisense, the leader in API-first embedded analytics that transforms enterprise data into a competitive edge, today announced the beta launch of the Sisense Analytics Chatbot and enhanced natural language narration. These embeddable conversational analytics tools, powered by Generative AI, allow users to ask natural language queries about their data and gain additional insights with suggested follow-up action through natural language narration.

The new Analytics Chatbot interface answers natural language questions, suggests queries for users to start exploring analytics, provides narrative explanations for greater context through data visualizations, and suggests actions consistent with these insights. Creators can easily embed the Analytics Chatbot or the new narration capabilities within Compose SDK for Fusion without necessitating a chat-like, close-ended UI in a dashboard. Compose SDK delivers customized data experiences using a code-first, scalable, and modular software development kit to enable data monetization and improve revenue generation. Additionally, the chatbot can output code snippets for developers, empowering them to create more innovative end-user experiences, and bringing them to market faster.

“This new transformative tool can redefine how data insights are accessed and utilized within applications,” said Ayala Michelson, Chief Product Officer at Sisense. “With Sisense Analytics Chatbot, we are meeting the demand to integrate GenAI into applications, empowering businesses to build advanced data experiences for everyone with new visual analytics embedded within the context of end-users' work. For developers, the chatbot represents another way to seamlessly integrate customized analytics into their products, bringing added value to their users.”

Analytics chatbot and narration capabilities include:

• In-app analytics:  End-users simply ask questions using natural language to uncover actionable insights at the right place and the right time within the context of their work with suggested actions consistent with these insights. Builders can seamlessly integrate this capability within their application interface using multiple LLM models in combination with outside-world data.
• Automatically generated narratives: Bring data storytelling to life using GenAI to create explanations for charts, graphs, and dashboards and provide narratives for greater context understanding.
• Contextualized data experiences: Creators can quickly embed GenAI-powered capabilities and take advantage of the Sisense Knowledge Graph that interprets user input within the context of their data giving end-users broad flexibility in how they structure their questions.

“Sisense's innovative analytics chatbot and narration features meet a crucial need for more intuitive and actionable data insights,” said Mike Leone, Principal Analyst of Analytics and AI at ESG. “By addressing customers' demands for seamless, integrated, and user-friendly analytics solutions, Sisense is empowering ad-users to explore data on their terms and make insight-driven decisions with ease."

Sisense is leveraging Azure OpenAI or the bring-your-own-LLM-key method as provided by the user, incorporating a proprietary RAG (Retrieval-Augmented Generation) framework and based on a business intelligence universal knowledge graph. When a user asks a natural language question, the prompt is enriched with a few examples of YAML strings composed of Q&As, and other supporting knowledge to generate the query.

"These examples are stored in a vector database as vectors/embeddings, and Sisense uses a similarity search algorithm to find questions similar to those posed by the user," explained Senior Director of AI, Avi Grushka. "Sisense also extracts the user's schema, and runs the augmented prompt. The next step involves receiving a response in the form of a secure data structure. This is translated into SQL using a proprietary tool and executed on the user's data."

With Compose SDK for Fusion, developers can utilize Sisense’s analytics platform, enabling the creation of dynamic queries, charts, and filters directly from application code without any predefined dashboards or widgets. Using flexible React components and APIs, developers can mix and match AI-powered components to create their own tailored analytics experience within their applications, now available to all Compose SDK users.