Artificial Intelligence and Analytics | A Changing of the Guard

When the coach starts looking to the bullpen, the pitcher on the mound is not long for the game.

This post is another gem from Tom Davenport.  For me, it’s something of a canary in the coal mine.  Its very title, “The End of Analytics?,” poses a seminal question that, in effect, aptly names this moment in time; that we’re in the final act of the analytics show, where visualization tools like Tableau and Qlik are front and center for decision making.  From now on, we’ll start to see organizations increasingly look towards new cognitive and artificial intelligence solutions that are powered by machine learning, natural language processing, and text analytics to augment, and, eventually, eclipse the likes of the Tableau-types.  “Leapfrog” is the term Davenport uses.  Salesforce’s Einstein has made a great fanfare of late, and IBM Watson has been at it for some time (though laughably so, to some who track the space closely).  And then there’s Magellan from OpenText and even Attivio’s Cognitive Search and Insight Platform.  One thing is for certain: this won’t happen overnight.

But it’s clear to me that Artificial Intelligence is the next great wave.  Thirty years ago, it was ERP.  Then came the internet.  Then Analytics.  Now there’s AI.  There are no clear start or stops in between each; these are movements, not revolutions.  No hard and fast replacements of one with another.  In fact, much of one bleeds into the next.  And none truly disappears.  AI will continue to emerge gradually, while visual analytics is still very much present.  But one small ripple at a time, the movement will occur, like a faint but steady drum beat.  In the moment, most of us won’t even recognize that it’s part of AI’s larger, broader wave of change sweeping the land.  Think Alexa.  Like a swimmer in the ocean, unable to appreciate at any one second the receding of the tide.  But once she emerges from the big drink, sitting on the sand, towel in hand, it becomes apparent to her that the tide is, in fact, going out.

So here we are, on the cusp of a sea change.  One wave pulls back, as another creeps forward.  At times, it’s hard to notice.  Yet there are signs.  Tom’s article is a clarion trumpet.  AI is creeping onto the mainstage.  For now, alongside Analytics.  But it won’t be long.

Traditional BI reporting dominated the software market for a generation.  It was how decisions were made.  For some, their entire job was dedicated to running the same report each week for the C-Suite.  In the earliest days of the movement, it was the CIO and his coterie analyzing reports.  Their executive insights and directives were to trickle down throughout organization, one hierarchical layer at a time.  As word passed from the mountain top, their decrees would be enacted at the lower levels.  But, as it turned out, this centralized vision of decision making never truly materialized.  The reports were static and rearview-focused.  The whole process wasn’t agile enough.  Gradually, the dark arts of data reporting were demystified and, in time, discredited.  New vendors, like Tableau, popped up to realize the vision and – suddenly! - the entire process of data-driven decision making was truly democratized.  Now everyone was empowered.  And it spread like wildfire.  It was almost akin to the Gutenberg printing press, as data was read and interpreted at the individual level, versus only at the foot of a high priest or data scientist.  At this point, the data-driven organization has been powered by Tableau and a host of increasingly visual analytical tools for almost a generation.

And now there’s AI.  Somehow, it seems to have even greater potential to radically alter how decisions are made.  It seems pregnant with the possibility to affect our daily lives in myriad ways, large and small, to wash over all of society in one way or another.  The use cases are legion.

But like the mainframe and tape backups, which both are still around decades after their demise was foretold by many, analytics will be here for the long term.  AI is not its death knell.  Nonetheless, as Davenport suggests, organizations are starting to look towards the early manifestations of AI – the cognitive stage - as a way to skip over an obsessively graph-centric approach towards decision making.  This quote, in particular, from Davenport’s article seems to sum it up: “We’ve had enough bar charts. Nobody has time to digest them anyway. We want our analytics to tell people what to do.”  Indeed.

It will be awhile before we see this vision meaningfully realized, and – who knows? – maybe we’ll never get there, as whatever innovation is next on the horizon pulls us off our charted course towards AI into another direction entirely.  But what is clear is that there’s a changing of the guard afoot right now.  Make no mistake.  Cognitive solutions, the hallmark of the early stage of AI, are augmenting, for now, analytically-driven approaches towards decision making.  In time, they may very well do more than just augment.