top of page
Laptop keyboard, coffee, sticky notes, and pencils on wood background
Search

Will Business Intelligence be powered by AI soon?

Over the years Business Intelligence has become synonymous with dashboards. In the Data Vault User Group’s latest meet-up, Wayne Eckerson – president of the Eckerson Group – presented “Is there a future for Business Intelligence? Key trends you need to know”. Wayne’s presentation sparked a lot of conversation and questions amongst the attendees, and we found his view on the impact of Artificial Intelligence (AI) on Business Intelligence (BI) extremely interesting.


Business Intelligence limitations

Wayne stressed that over the years, people in the industry have pointed out some common Business Intelligence limitations. They are:


  • Too reactive

  • Too Generalised

  • Too Manual

  • Too Descriptive (not analytical enough)

  • Too Inflexible

  • Too hard to use.

  • Doesn’t turn insights into actions.


It has been said that “BI has a last mile problem”. Meaning that there is a gap in the metaphorical bridge between data and action. This has raised questions from many people about whether BI tools are delivering sufficient value to your business.


The evolution of intelligence

Business Intelligence has come a long way since its inception in the 1960s. Initially, businesses relied on simple reporting and data analysis tools to help them make informed decisions. However, as the amount of available data grew, so did the demand for more sophisticated tools to manage and analyse it. In the 1990s, data warehousing emerged to store large volumes of data from various sources in a centralised location, which made it easier to analyse.


In the 2000s, businesses gained access to powerful data visualisation tools, which enables them to explore data in new and more intuitive ways. Today, with the emergence of artificial intelligence, BI has become even more powerful – allowing businesses to not only analyse past data, but also to make predictions based on that data. As technology continues to grow and evolve, we can expect to see even more advancements in BI that will help businesses to make even more informed decisions and stay ahead of the competition.


Three Eras of Intelligence

Furthering on from the evolution of intelligence, Wayne highlights three main eras of intelligence. Of course, some characteristics of each era overlap, but they can be broken down in three main titles – IT-driven, user-driven, and model-driven.


IT-Driven – 1990 to 2005

  • Data Warehousing

  • Data Integration

  • Data Quality/ MDM

  • Standard KPIs and Dashboards


User-Driven – 2005 to 2020

  • Data Refinery

  • Analytics workbench

  • Data Literacy

  • Data and Report Governance


Model-Driven – 2020+

  • Data Science

  • Streaming data

  • Data Pipelines

  • MLOps


Is AI going to replace all the tools and people we have?

The presence of AI in today’s world is growing exponentially as it is fed more data. If you pay attention to economic forecasts, it looks like we’re heading towards a future of tools like ChatGPT being used by most businesses daily. But we believe that it isn’t a case of an “AI take over”. Instead, we see it as a tool to supplement the most important part of any business – people.


AI tools, such as ChatGPT are still very young in their lifecycle, so they come with their own limitations that you need to consider before adopting the tools. They are:


  • It only works with text – not images or other types of media.

  • It doesn’t have DBMS connectors.

  • It doesn’t always get things right.

  • It doesn’t display visualisations.

  • It requires 10,000+ citations to achieve accuracy.


In Wayne’s presentation, he mentioned how ‘Sisense’ – a business intelligence software company in New York – has built an extension that leverages ChatGPT. They have leveraged the AI tool which converts questions asked in it and displays it in ‘Sisense’.


Summary

In summary, it was great to attend the February 2023 Data Vault User Group meet-up with Wayne Eckerson. The president of Eckerson Group presented an interesting topic, and it’ll be interesting to see the development of AI in Business Intelligence.


At Datavault, we see the emergence of AI as an evolution, rather than a revolution. No one is going to be able to stop the emergence of AI tools such as ChatGPT, but we believe that the best approach will be to harness it and add it to our toolbox to deliver data solutions.


Keep an eye out on the Data Vault User Group meet-up page and social media pages so you don’t miss out on any future meet-ups!

bottom of page