The Evolution of Insight Advisor – Continuing the path forward with Augmented Analytics and AI

[QlikProductUpdates]

Introducing LLM-driven language generation and automated authoring

Insight Advisor in Qlik was introduced in 2019 as a first step into AI-driven analytics.  At the time, users could auto-generate visualizations and analytics using selections and keyword search.  That kicked off a journey of rapid innovation that resulted in full natural language-driven insight generation in over 10 languages, conversational analytics through Insight Advisor Chat, natural language insights on dashboards, a variety of advanced analysis types such as correlation, clustering, forecasting, key driver analysis, etc., and predictive analytics with Qlik AutoML.  We introduced a rich business logic layer to customize natural language processing and insight generation, and a full set of augmented analytics APIs to extend and embed AI-driven capabilities.

[QlikProductUpdates]

But in 2023, the arrival of ChatGPT and generative-AI changed the game.

Qlik was quick to respond, but not in haste.  We introduced a new AI new strategy coined Qlik Staige, which is comprised of three pillars – a trusted data foundation for AI, AI-enhanced analytics, and self-service AI solutions.  The foundational aspect of our AI-enhanced analytics strategy is to take Insight Advisor to the next level, modernizing the architecture and language model to take advantage of the latest generative-AI capabilities.  We are pleased to introduce the first step on that journey – LLM driven language generation.

Insight Advisor now features generative-AI driven language generation as a private preview feature, both in Insight Advisor Chat and in-app search experiences.  Users will now get more "chat GPT like" answers when asking questions, with human sounding narrative for improved readability and a wider range of observations, summarizations, and additional insights.  Qlik has partnered with Amazon Bedrock to utilize state of the art LLM technology, through a fully built-in solution that takes advantage of Qlik's security and governance.  And of course, all analytical calculations are still generated by the trusted Qlik engine, with the LLM being utilized to enhance and improve output.  Going forward, we will continue to modernize the language model in Insight Advisor, improving intent recognition and enhancing the overall capabilities and experience.

In addition, in the coming months we will be introducing automated authoring.  Business analysts might be great at building visualizations, but it’s a time-consuming and complex process.  Automated authoring is a new AI-driven creation experience that speeds and simplifies authoring in Qlik.  With a wide variety of analysis types to choose from, you can now drag-and-drop AI-generated analyses directly onto your sheets, select applicable data from a list of suggestions, and let Qlik do the rest.  In a few clicks you get a sophisticated analysis that would have taken far longer to build manually.  This enables business analysts and authors who know what they want to gain efficiency, shorten time-to-value, and spend more of their effort actually analyzing the business.

Both of these exciting new features were demonstrated at our annual user conference, Qlik Connect, and received great feedback.  The new LLM-driven language generation capability is now offered in private preview, and automated authoring will be generally available later this summer.

Stay tuned as we continue to lead the way in AI-powered analytics – lots more to come.