How to Use Generative BI for Self-service Analytics and Fill the Data Literacy Gap

Learn how Generative BI can help business users get quick insights from data without needing to write SQL queries or learn BI tools.
How to Use Generative BI for Self-service Analytics and Fill the Data Literacy Gap

These days there is too much buzz around Generative AI. Everyone is trying to figure out how to use ChatGPT and similar chat-based AI tools for their business. I reviewed different use cases that you can potentially enable using Generative AI chatbots. Most of them fall into the category of content generation.

Those are good for productivity in areas like sales and marketing. However, when it comes to getting insights from data, I think Generative BI (Business Intelligence) must be on top of every data leader’s mind. Many different BI vendors are trying to build some sort of AI capability into their tools with varying degrees of success.

What is Generative BI?

Generative BI is a way to generate business intelligence dashboards and reports that are key to getting insights based on your business data. Traditional generative AI relies on LLMs that are trained on large public data. For an enterprise, the higher value use case is to get insights based on your data.

You don’t want to put your company’s private data into the public domain. As such, each company needs to figure out how to use its private data and get insights based on it using a private model.

With the popularity of cloud-based data warehousing platforms that separate storage from computing, a lot of Business Intelligence tools enabled self-service. The business users can be isolated from each other, and they can perform their BI analysis independently without impacting other departments.

However, traditional BI is time-consuming and requires specialized training to build dashboards and reports that can provide actionable insights. The skills and data literacy gap is very real, and most end-user organizations struggle to hire the right skills in business teams that can realize the dream of self-service.

Business users just aren’t familiar with how the data is organized in your data warehouse.

This is where Generative BI can come to the rescue and help fill the gap in data literacy and skills. With  Generative BI, you don’t need to write SQL queries or learn BI tools. The natural language processing capability of Generative BI can help average business users author dashboards and reports.

Also Read
How Should CEOs Think About Generative AI? – Delivering Business Value
How to Use Generative BI for Self-service Analytics and Fill the Data Literacy Gap

This will not only speed up the time required to build the dashboards and reports but will also shield business users from low-level technical details that they may not know.

Generative BI will be able to figure out which tables to use, in certain cases, there may be complex SQL queries that need to be written to get data from tables. These queries could be generated as well.

Once the right data is selected, the next challenge is to visualize this data. The generative BI tool can select appropriate visualization charts as well. This looks like a great win for average business users who want to get quick business insights from data and generative BI just helps them with that.

The next level challenge for business users is when they present these types of visualizations to executive leaders who have questions about the charts. These types of questions can also be answered by generative BI.

Imagine typing questions, “At what rate are my sales growing quarter-over-quarter?” and the tool comes back with an answer in the form of a chart based on your data. That is the future we are pursuing. There can be additional complex questions based on initial simple questions.

This is where we almost get into a conversation with the tool and data. However, the tools are not there yet. All the different BI vendors are trying to build this. Some of them claim that they have it already, but it is not ready for prime time yet.

The responses from the tool will have to be validated by a human to make sure we are not falling into the hallucination problem a lot of AI tools seem to fall into.

However, with a proper strategy for incorporating these modern tools into data-driven decision-making processes, organizations can delve into the modern AI-driven era that seems to promise many exciting and cool capabilities.

About the Author:

Kiran Kanetkar is Vice President - Data and Analytics at Pendulum Therapeutics. He is also a member of the CDO Magazine Editorial Board.

Kanetkar is a seasoned technology executive leader with over two decades of experience in the field. He is adept at leading teams and establishing a data-driven culture. Previously, he was Senior Director of Data Engineering, Business Intelligence, and Analytics at Petco, where he led the company's digital transformation and improved analytics capabilities.

Before Petco,  Kanetkar worked as Enterprise Program Manager at Warner Bros. Entertainment Group of Companies and as Enterprise Integration Architect at Abbott Laboratories.

He is passionate about using modern agile technology architecture to solve business problems, deliver business value, and remove technical debt.

Related Stories

No stories found.
CDO Magazine