Can we fix the plague in analytics with AI?
Every Business Intelligence (BI) and analytics vendor is integrating a form of artificial intelligence (AI), machine learning algorithm (ML), and natural language generation (NLG) into their products. 'Augmented analytics', is the hot new topic and full of hype right now, but can it fix the fundamental flaw that has plagued BI tools for decades - adoption?
A Gartner Survey Note suggested the adoption of analytics products was about 32% of employees in an organization. In my previous analytics life, this percentage was a lot smaller.
Clearly, the adoption of analytics is a huge problem to be solved because it highlights just what a small percentage of business decision making is based on quantitative evidence. Without data, businesses cannot optimize for efficiency and will quickly fall behind organizations that build a culture of decision making on empirical data.
The technologies behind the rise of the new wave of analytics - AI, ML, and NLG - have been around for decades, but we can already see that their application to analytics will fundamentally change how analytics is used in the enterprise.
"Augmented analytics, an approach that automates insights using machine learning and natural-language generation, marks the next wave of disruption in the data and analytics market.
It will transform how users interact with data and how they consume and act on insights."
Gartner, Augmented Analytics is the Future of Data and Analytics
Organizations that have a centralized data-driven approach throughout all departments are incredibly rare (despite the claims of many businesses). A lack of analytics skills among business users, the time lag from request to delivered insights, and several other crucial factors mean that non-analysts see BI tools as another delay to them just getting their jobs done. So adoption remains low. But can augmented analytics fix this?
Augmented analytics doesn't come without its own problems though. This blog series will take a look at the current state of analytics, opportunities to increase adoption through AI-driven technology, and the challenges that this new wave of analytics will bring.
The state of analytics today
In many ways, BI has not really changed in decades. It still centres around dashboards and requires enormous amounts of manual data preparation.
If you have a good BI tool, you will have access to analytic dashboards (operational, analytical, and strategical), the ability to drill horizontally and vertically across data sources to get detailed information, and you will have sharing and broadcasting across devices for multiple users. On top of those, you should have the ability to set up alerts built on rules and get actionable data insights. Some of these are newer features that have not always been available, but everything comes back to the dashboard - the bedrock of BI.
Despite most leading BI tools having the aforementioned features in different forms, there still is not the wide-spread adoption of analytics that the self-service analytics tools expected and promised.
The data on analytics usage in businesses today
Computing Research surveyed the analytics usage of 110 businesses ranging from less than 250 employees to many thousands across multiple industry sectors. Some of these results were published in 'Business Analytics in the Machine Learning Era', February 2018 and can be seen below. They highlight the current state of analytics adoption and advancement in business today.
The survey revealed that 68% of businesses are either following or falling behind everybody in their use of analytics. This reveals that there is a long way to go before businesses realize the potential of analytics, full stop.
What best describes your organization's use of analytics
Highlighting the issue of adoption, the figures below show that analytics is 'definitely not' or 'probably not' being used to its full extent in 77% of business today. When such a large percentage of businesses is still relying on gut feel rather than data, it becomes clear that adoption is still dire.
Is your analytics being used to its full extent?
So how are we going to see increased usage of data with this new wave of analytics if the amazing tools we have today are still not being fully utilized? We first need to have a high-level understanding of how each analytics era has evolved, solved old issues, and brought new challenges with them. Then we can begin to understand the cause of low adoption rates and how augmented analytics could help or hinder.