Tech Talk: How Decision intelligence Takes the Heavy Lifting Out for Data Scientists


Ajay Khanna, CEO and founder, Tellius, joins Neha Pradhan Kulkarni to discuss the key differences between business intelligence and decision intelligence. Ajay discusses how the adoption of augmented analytics and decision intelligence has helped to democratize and take the heavy lifting out for data scientists.

In this edition of Tech Talk with Spiceworks News & Insights (SWNI), Ajay shares some of the real-world use cases where decision intelligence adds value. He also talks about key areas to focus on before investing in this technology.

Key Takeaways on How Decision intelligence Democratizes Decision Making for Enterprises:

  • The analytical outputs of decision intelligence are key drivers, segments, correlations.
  • Decision intelligence’s cloud-native architecture allows for distributed, elastic compute to parse the full dataset for insights.
  • Decision intelligence is applicable for both small and large companies

Here are the edited excerpts from our exclusive interview with Ajay Khanna, CEO and founder, Tellius:

SWNI: Tell us about your role at Tellius and what are the key areas you have been addressing since you joined the company?

Ajay: My passion has always been to build new technologies and disruptive business models. Two areas that have been constants throughout my career have been automation and data analytics; in fact, I had been thinking about automation and its potential impacts even before I came to the U.S., and it led to an obsession with data. That—combined with my mission to build something that would keep up with the ever-evolving demand for new technology—is what inspired me to found Tellius.

“Tellius is the first decision intelligence platform that combines AI- and ML-driven automation with a natural language search interface. Making analyzing vast and complex amounts of data easy and accessible for everyone has been the key focus since Tellius’s creation.”

We now empower global organizations across industries – including multiple Fortune 500 companies – to augment human expertise with AI/ML for better and faster data-backed business decision-making.

See More: How Spotify’s Voice Recognition Device Is Making In-Car Listening Better 

SWNI: While the need to evolve BI tools and strategies is not a new trend, there are now more modern, robust tools – such as decision intelligence – that are helping organizations to move beyond non- contextually data analysis. Can you tell us about the key differences between BI and decision intelligence? Why now is the time to evolve traditional BI?

Ajay: Decision intelligence differs from traditional BI in 5 areas: outputs, experience, analysis types supported, data processing techniques, and scale. 

Analytical outputs from BI are primarily dashboards and reports in a very “pull”-based mentality, i.e., an analytics consumer is expected to seek what they need. The analytical outputs of Decision intelligence are key drivers, segments, correlations, and patterns in both a push and pull-based approach, automated by proactive notifications that go beyond simple threshold-based alerts (i.e., using ARIMA models).

The data analysis experience from BI tends to involve browsing dashboards, and reports, or performing manual slice-and-dice hypothesis-based testing which can introduce human bias. Decision intelligence relies on AI-Powered approaches, relevant to the user.

“The types of analysis supported differ as well between BI and Decision intelligence. Whereas BI tends to focus on answering “what happened”-type descriptive questions, decision intelligence goes beyond to provide rapid answers to “what happened” (descriptive), “why it happened” (diagnostic), and “how to proceed” (predictive and prescriptive).”

Another area the two differ is the data processing technique utilized. BI tends to rely on SQL queries. Decision intelligence supports SQL/Python, as well as more sophisticated statistical/ML/AI-based processes, as well as natural language query (NLQ).

Finally, the data scale differs between the two. Traditional BI relies on pre-aggregated data and predetermined drill paths due to limitations in processing power unable to query the full unaggregated data, whereas decision intelligence’s cloud-native architecture allows for distributed, elastic compute to parse the full dataset for insights rapidly.

See More: How TripAdvisor Is Leveraging Voice Technology To Boost Business: Q&A With Christine Maguire

SWNI: How has the adoption of augmented analytics and decision intelligence gone up among enterprises and how does it help to democratize and take the heavy lifting out for data scientists?

Ajay: The global augmented analytics market size will be worth $62.5 Billion by 2028, according to data from Verified Market ResearchOpens a new window . More organizations have started to adopt these tools because they allow for critical components of insight generation to be automated, improving data governance and accuracy and increasing productivity at lower costs. Augmented analytics and decision intelligence software allow anyone in an organization to access data and work with it—even those without a traditional data science background.

“As the data scientist shortage continues to grow—and as AI becomes more ubiquitous in the enterprise—companies must invest in tools that democratize advanced analytics and empower every person to make better, data-backed decisions.” 

Taking the heavy lifting out of the hands of data scientists not only allows more tasks to be completed during the workday but also frees up teams to spend time gaining deeper insights from the data. As a result, companies can grow and push more boundaries, ultimately becoming more profitable.

SWNI: Data has occupied the power seat for most AI-led decision-making at enterprises today. What are some of the real-world use cases where decision intelligence adds value?

Ajay: Decision intelligence makes it possible for users to analyze millions of data points in seconds to uncover hidden connections and obtain predictions and clusters that they would not have otherwise seen. This can be useful in helping retailers understand issues like customer satisfaction and cart abandonment in real-time, allow banks to assess a potential lender’s credit risk, and empower pharmaceutical companies to build successful market access strategies with better commercial insights. Decision intelligence is applicable for both small and large companies to help them achieve greater insight into their consumer data and beyond.

See More: AWS Director of Enterprise Strategy on How To Deliver Digital Transformation

SWNI: As more companies are turning to decision intelligence/augmented analytics to support their businesses, what are the key areas to focus on before investing in this technology?

Ajay: Before investing in augmented analytics, the most successful companies typically do three things. 

First, they ensure their data infrastructure is in good shape. This could include modernizing their data stack from a primarily on-premises setup to a cloud data warehouse or data lakehouse. Quality data infrastructure is important because, like all forms of analytics, augmented analytics relies on solid data pipelines. Augmented analytics is ideal for the modern data stack as queries can be pushed down for cost-effective zero data movement.  

Second, successful firms foster a collaborative and open environment for exchange between analytics creators (i.e., analysts and data scientists who do the analytical work) and analytics consumers (i.e., business users who typically benefit most from analytical outputs). 

“Augmented analytics is an accelerator of analytics collaboration. However, organizations should think through team dynamics, communications, and team formation, laying the groundwork for collaborative analytics before investing in tooling.”

Finally, the most successful firms invest time in identifying business-impacting use cases for augmented analytics. Successful companies spend the time to internally identify key stakeholders, the potential impact on business, and build up the business case for the technology before making investments in pure technology.

SWNI: Going forward, which are the key trends in AI-led data analytics to directly impact organizations in the next year?

Ajay: In the coming years, we expect a few key trends in AI-led data analytics. These trends will primarily focus on functionality, BI and ML, and industry-specific solutions.

As the underlying technology that powers augmented analytics matures (e.g., NLP, NLG, AI/ML, etc.), we expect to see more analytics functionality and more powerful/flexible functionality. Predominantly available in cutting-edge big tech firms, these types of functionalities appear “under the hood” of leading augmented analytics solutions, and thereby offer a greater reach and impact to more people and organizations.

Soon there will be an even greater blurring of lines between BI and Data Science/ML offerings. This is already a reality in leading augmented analytics platforms which offer descriptive, diagnostic, predictive, and prescriptive analytics in one tool — but we expect there to be continued innovation in the analytics experience surrounding these capabilities which will further democratize these modes, making them accessible to more people to drive even greater business value.

We should also expect more industry-specific accelerators or ‘pre-canned solutions’ to emerge in the marketplaces of augmented analytics providers, which will further speed up the time-to-value of investments. For example, it is easy to imagine a retail-specific accelerator pack consisting of augmented segmentation (RFM and ML-based models), CLTV forecasting, automated market basket analysis, and other such capabilities bundled together so business users have a seamless plug and play experience rather than facing a blank canvas from the start.

About Tellius

Tellius is an AI-driven decision intelligence platform that enables anyone to get faster insights from their data. The company’s platform combines AI- and ML-driven automation with a search interface for ad hoc exploration, allowing users to ask questions about their business data, analyze billions of records in seconds, and gain comprehensive, automated insights in a single platform. 

About Ajay Khanna

Ajay Khanna, CEO & Founder of Tellius, a company disrupting business analytics with AI, is a tech entrepreneur who has a passion for building disruptive enterprise products with an awesome user experience. Before starting Tellius, Ajay was CTO & Founding member of Celcite, a telecom analytics and solutions company that was acquired by Amdocs in 2013. Ajay has over 25 years of extensive experience working in various technical, business, and consulting roles. He holds a degree in electronics and communications engineering from the Thapar Institute of Engineering & Technology.

About Tech Talk

Tech Talk is an interview series that features notable CTOs and senior technology executives from around the world. Join us as we talk to these technology and IT leaders who share their insights and research on data, analytics, and emerging technologies. If you are a tech expert and wish to share your thoughts, write to [email protected]Opens a new window

What does your organization view decision intelligence in the long run? Share your findings with us on LinkedInOpens a new window , Facebook,Opens a new window  TwitterOpens a new window .