(Wright Studio/Shutterstock)
Will generative AI ultimately yield a satisfactory return on investment for corporations, given the significant sums they’re committing? According to Gartner, approximately 30% of GenAI initiatives are expected to fail within the next 12 months. A newly released report from ThoughtSpot reveals that pioneers of GenAI-powered analytics have begun to experience substantial benefits.
Commissioned SMR Connections and its analytical partner to conduct a comprehensive study, surveying 1,000 enterprise leaders on the adoption and application of generative artificial intelligence (GenAI) in their data analysis processes. Topics were categorised into three groups according to the maturity level of their GenAI projects, with 67% designated as early adopters having already integrated GenAI applications into production, 26% planning deployment, and 7% still in evaluation mode.
Among the numerous early adopters, nearly half rely on a return on investment (ROI) for GenAI functions of at least 100% over three years; specifically, 12% expect an ROI exceeding 300%, while 11% anticipate returns between 200% and 299%. The ROI expectations for this cohort are notably higher than those of the planners, who anticipate a return of at least 100% over a three-year period; in fact, nearly one-third expect returns of 200-299%, while fewer than 1% predict an ROI of 300% or greater.
The report, titled “The Future of Work,” highlights a concerning trend: GenAI is creating an insidious divide between individuals who effectively harness this expertise and those who do not.
Among early adopters, 37% report being significantly ahead of the market and rivals in terms of their GenAI usage, compared to just 11% among the planning cohort. Meanwhile, almost half (46%) of early adopters say GenAI has enabled them to stay only slightly ahead of the competition, a proportion similar to that seen among the planning group (51%).
Cindi Howson, ThoughtSpot’s Chief Information Technology Officer, is enthusiastic about the prospect of GenAI having a profound impact on the world of knowledge and analytics, thanks to these impressive numbers that have piqued her interest.
“With the potential for productivity enhancements and innovative business models now unlocked, we’re just at the starting line,” said Howson. As we navigate the early 1990s era of dial-up internet, people are only just starting to grasp the vast possibilities that lie ahead.
What Data Scientists Can Learn from Business Intelligence
Alternative revenue streams for GenAI include prominent applications such as chatbots and co-pilot assistants, which have gained significant traction since ChatGPT’s debut in autumn 2022. Additionally, agentic AI capabilities are also being explored. In contrast, ThoughtSpot’s corporate utilizes GenAI to amplify the analytics and enterprise intelligence capabilities for its clients’ prospects.
As organisations invest in advanced analytics and business intelligence, they can reap numerous benefits, including increased revenue and productivity through more informed, data-driven decision-making, as well as enhanced enterprise effectiveness and the potential to create valuable knowledge products.
“The benefits are multifaceted,” Howson noted, “including generating novel revenue streams, optimizing decisions surrounding those streams, and subsequently streamlining operational efficacies throughout the process.”
Studies have consistently shown that only around 25% of employees in a typical organization possess the ability to seek answers and tap into its collective knowledge. Only a quarter of employees have access to business intelligence (BI) and analytics in different phrases. ThoughtSpot strives to empower 100% of its workforce with seamless access to analytics, and GenAI plays a crucial role in achieving this goal.
When discussing their organization’s objectives, Howson emphasized, “That’s a vital component of our mission.” While it’s acknowledged that individuals may have limited knowledge literacy skills, the phrase “we’ve got low knowledge literacy” is somewhat ambiguous. It would be more precise to state “many of us lack the necessary knowledge literacy skills.” Additionally, using the term “upskilling” implies a one-way process, whereas developing knowledge literacy is a collaborative effort that requires mutual understanding and engagement. Generative AI’s capacity to elucidate unclear charts and highlight outliers on a webpage is making a significant impact.
GenAI in Analytics
ThoughtSpot leverages GenAI primarily by harnessing the power of natural language queries (NLQ), significantly reducing the need for technical expertise to access knowledge. Utilizing GenAI, various applications are developed to streamline the creation of dashboards and experiences, thereby enabling faster detection of anomalies within data.
“With this new tool, dashboard creators will no longer be bogged down by mundane tasks or tedious work, instead being empowered to focus on higher-value activities.” “For business professionals, this technology enables them to pose more sophisticated queries and cultivate a deeper analytical mindset, rather than operating in ignorance. Ultimately, generative AI has the potential to elevate everyone’s work; however, those who fail to adapt and learn how to effectively harness its power risk being left behind or marginalized.”
According to ThoughtSpot, GenAI can leverage internal and external databases to access relevant data significantly faster than even the most skilled executives or information professionals could accomplish independently. By permitting users to formulate their own queries in plain language, this tool enables individuals to discover solutions tailored to their needs through a conversational dialogue, rather than relying on pre-existing data created by experts with limited business knowledge that may not be applicable in real-world scenarios.
Long before ChatGPT’s debut, ThoughtSpot had already been pioneering the use of Natural Language Query (NLQ) capabilities to revolutionize decision-making processes. When ChatGPT showcased its unparalleled prowess in generating massive language models, numerous companies surmised that these LLMs could effortlessly produce not only Shakespearean sonnets in English and code segments in Java but also coherent SQL statements.
Unfortunately, this assessment does not align with Howson’s findings.
Straightforward translation of plain text to SQL queries simply isn’t feasible. According to her, the highest level of accuracy was around 30%. We’ve maintained a proprietary, patented semantic layer within our marketplace ecosystem for over a decade, complemented by a suite of sophisticated rating algorithms and a robust RAG structure, thereby significantly enhancing overall accuracy. After which, the human operator is re-looped in to further fine-tune the accuracy.
Foundations for GenAI Success
Can’t you just rise above the challenges and decide to revolutionize your business operations with GenAI? As companies have learned from their early experiences with traditional machine learning technology, there are often pre-requisites that must be fulfilled before organizations can effectively leverage cutting-edge AI capabilities.
MIT’s report bares this out. While early adopters of General Artificial Intelligence face numerous hurdles, the top five obstacles include concerns about ensuring safety, navigating strategic implications, resolving model quality and utilization issues, addressing knowledge gaps, and overcoming implementation difficulties. Information administration and general techniques remain significant hurdles?
“One cannot successfully deploy AI without a solid foundation of knowledge, nor can they establish a positive reputation unless their efforts align with the company’s values.” While there’s a distinction between proving ideas versus driving tangible outcomes, it’s essential to highlight how our services can amplify client expertise, streamline operations, and boost productivity by tackling the dashboard backlog and enhancing analyst and business user efficiency. Having these two elements constitute one of the most significant differences.
At GenAI, we’ve mastered the art of revolutionizing the future of artificial intelligence. As Howson noted, synchronizing the roadmaps of both the enterprise and IT divisions is a crucial challenge that cannot be overlooked.
“The tensions between opposing viewpoints are palpable, with a sense of ‘us versus them’ and deep-seated frustration evident on both sides.” The information team’s pace is far too slow. Enterprise will get annoyed. They set off to establish their own enterprise. And GenAI enables seamless discussions about desires, fostering innovative collaborations.
Given the extensive buzz surrounding this development, it is evident that GenAI offers tangible alternatives. While not every instance of this technology may prove successful, initial findings suggest that early adopters are already seeing positive results. As the potential of GenAI is set to surge in the coming years, it’s crucial that businesses seize this opportunity by investing now to position themselves for future success.
“The value we can extract from this lies in its productiveness benefits, enabling innovative business models where we’re just starting out,” Howson said. “We’re at a pivotal moment, reminiscent of the early dial-up era of the internet, where people are just starting to grasp the vast possibilities enabled by GenAI.”
Obtain a comprehensive report from MIT.
