Evolution of analytics marked by wider dissemination of data

With BI platforms getting easier to use by the year and enabling more self-service users in process, the value of analytics is being realized by a growing audience.

This is the second in a two-part series looking at the evolution of business intelligence over the past two decades. Part I examined where BI was 20 years ago and where it is today, while Part II looks at how we got from the analytics platforms of 1999 to the ones of 2019, and what BI might look like another five years from now.

An information revolution has transformed the business world over the past two decades, dislodging data from the domain of a select few and disseminating it to a wide array of business users, but the evolution of analytics didn't happen all at once.

It's been gradual, though the process has sped up far more in the last 10 years than the previous 10.

Data is now widely available throughout organizations to anyone in finance, marketing and sales departments -- and beyond. It's for anyone who needs data to do their job.

It's in places like museums, which use data to plan exhibits and drive membership; sports, with teams and players learning patterns in order to predict potential outcomes; and urban planning projects, with traffic patterns and effects on communities all looked at before ground is broken.

It's used in just about every industry, and it's available nearly in real time.

Data lockup

Two decades ago, data was locked up. The technology of the time was rudimentary compared with that of today, and it necessitated that IT professionals oversee an organization's data, from collecting it through managing it  and cleansing it to disseminating it upon request.

The technology of the time also required a significant amount of time between the point of initial request and the delivery of a report.

"It was only available to the people it was absolutely necessary for them to have -- they couldn't do their jobs without it," said Marge Breya, senior executive vice president and chief marketing officer at MicroStrategy, whose software industry career spans more than 25 years. "It was for finances, for CFOs, for anyone with order or forecast needs -- production. If there was planning, production, payroll. It was very basic, and a lot was not automated."

A lot has changed.

The evolution of analytics from there to here, from 1999 to 2019, was characterized by technological innovation well beyond the world of business intelligence -- mobile, for example -- but also innovation by BI vendors taking advantage of new technology. And it was characterized by a changing of industry leadership with the progressive vendors of the time being surpassed by a new generation.

Meanwhile, as the evolution of analytics takes BI into the future, it's still being characterized by innovation beyond the world of business intelligence that is being adopted by BI vendors. New technologies like Augmented intelligence and machine learning are being incorporated into BI platforms with a young crop of vendors leading the way as the former innovators scramble to catch up.

Key innovations

BI remained largely unchanged as the first decade of the new century progressed, though technological inventions advanced the evolution of analytics in small increments.

The widespread use of 64-bit computing beginning in 2003 was a significant development that allowed for the advent of in-memory processing in the middle of the decade.

The evolving landscape of business intelligence timeline
The evolving landscape of business intelligence

Microsoft, meanwhile, began including Excel in its operating systems in the mid-1990s, which at least opened up the idea of data analysis to a wide array of business users; the tech giant added Power Pivot to Excel in 2010.

By the late 1990s, mobile phones were becoming common. It wasn't until 2007, however, that Apple introduced the iPhone. BI vendors developed mobile apps to take advantage of the new capabilities of mobile phones, but even now they're still struggling to figure out how to best use mobile technology for BI purposes, a significant problem being the size of the screen.

The broad adoption of wireless technology, meanwhile, opened up the ability to conduct business nearly anywhere at any time.

And then there's the cloud. Amazon created Amazon Web Services in 2006, and two years later Microsoft first revealed Microsoft Azure, though it wasn't released for another two years.

"Many of the changes came from outside the BI world, like mobile technology and the cloud, which has enabled easy and large-scale deployments," said Donald Farmer, principal at TreeHive Strategy in Seattle and formerly an executive at both Qlik and Microsoft.

Then came a turning point in the evolution of analytics.

Data visualization transformed the entire analytics landscape, moving data consumption away from IT-generated spreadsheets to dashboard displays and any computer user's desktop.

There's no precise date about when data visualization helped move the evolution of analytics into a new era. And if IT-centric reporting characterized the first generation of BI, data visualizations and self-service analytics characterized the second. The process took place gradually, starting late in the first decade of the 21st century and continuing early in the second.

"The next set of innovators were the ones with data discovery tools -- Tableau and Qlik," said Rick Sherman, founder of Athena IT Solutions and a veteran of more than 30 years in the software industry. "Then Microsoft eventually became nimble with Power BI. They had a huge base but weren't burdened by a legacy BI suite and turned it around."

Now, the evolution of analytics has moved into a third stage, this one characterized by the cloud and AI tools such as machine learning and natural language processing.

The vendors

Two decades ago the vendors most associated with analytics were companies such as BusinessObjects, Cognos, Hyperion, MicroStrategy and SAS Institute.

A decade later, with the evolution of analytics moving into the data visualization stage still to come, those vendors were still the ones providing the vast majority of enterprise BI platforms. The difference, however, was that in 2007 Cognos was acquired by IBM, Hyperion was bought by Oracle and BusinessObjects was purchased by SAP.

Consolidation altered the vendor landscape, but not the technology one.

"In the first wave there was BusinessObjects, Cognos, Hyperion, MicroStrategy -- they were leaders for 10 years," said Sherman. "Anytime a new technology came up by a smaller vendor they would get acquired by the big three. It was a feeder system -- the technology was developed by smaller companies whose end game was to get acquired. But it was still an IT-centric mindset with those vendors. You could transform and visualize -- all that was paved by legacy vendors -- but it was IT-centric."

Still, according to Farmer, the vendors were simply taking advantage of the evolution of technology rather than inventing new tools on their own.

"It would be right to recognize that companies like MicroStrategy, SAS, Information Builders consistently delivered over that time," he said. "But BI was reactive to things outside of analytics -- it was rarely innovative. It was technology that was taken up by BI companies."

Data visualization changed that.

And with the rise of data visualization, so too came the rise of Tableau and Qlik.

"In [recent] years there was data visualization, beautiful charts and graphs to interpret data, and that was led by Tableau," Breya said.

Meanwhile, with Cognos part of IBM, BusinessObjects part of SAP and Hyperion part of Oracle, those vendors that had once been agile were held back somewhat by the size and scope of their acquirers.

The term legacy vendors cropped up to describe them, and they appear to have responded only in the last couple of years.

"At some point they got so big that it was tough for them to innovate in the next wave," said Sherman. "It's tough to innovate when you're part of a behemoth. They had a huge customer base that was IT-centric, and they were looking down their nose at cool new vendors. Eventually, their BI suites got to be unwieldy."

Now, it's Tableau (acquired by Salesforce in June) and Qlik that could be at risk of falling behind as the evolution of analytics moves BI into a third generation.

Instead of being viewed as the innovators, they're scrambling to keep pace with a crop of new vendors – many of them exclusively cloud-based -- like ThoughtSpot, Domo and Salesforce, whose focus is on AI for BI.

The world of 2024

Just as the BI platforms of five years ago are different than the platforms of 2019, the evolution of analytics will result in vastly different platforms five years from now.

Just as the evolution of analytics has taken BI from analytics in the hands of few to the benefit of only a few more to analytics in the hands of anyone who needs data to do their job, it will continue to make data more widely available, and available in new ways.

"The big difference we'll see is that BI as a practice of bringing data together in a single source will be over," Farmer said. "We're going to find it much more distributed and there will be many more ways of interacting with the data."

Farmer compared the way BI will be consumed to word processing, noting that 30 years ago word processing was a task all its own, while now it is infused in the electronic devices countless people around the world use every day.

Similarly, Breya said that in five years data largely won't be consumed on desktop dashboards. It will also be proactive rather than reactive, fueled by augmented intelligence capabilities.

The notion of someone going to find data will be going away. There will be tools to get intuitive information that is needed at that time. It will be predictive and experiential. ... Every product will anticipate what is needed to take the next step.
Marge BreyaSenior executive vice president and chief marketing officer, MicroStrategy

"The notion of someone going to find data will be going away," she said. "There will be tools to get intuitive information that is needed at that time. It will be predictive and experiential, so people or products don't have to look for it. Every product will anticipate what is needed to take the next step."

Sherman, however, said that so much has happened in a short span -- the explosion of mobile capabilities, the emergence of the cloud and the rise of AI -- that over the next five years there might be more adjustment than innovation.

"There will be more of a balance, a blend in engineering and science focus," he said. "There will be a wave of digestion. People will figure out data lakes, data prep, data integration."

All the while, the evolution of analytics will continue to result in the dissemination of data to a broader audience, giving more and more people the power of information.

Dig Deeper on Data science and analytics