• In studying buyer expectations and experiences from more than 1,000 IoT practitioners, GlobalData found that the majority of users rely upon basic reporting mechanisms as found in business intelligence (BI) systems when analyzing IoT data.
• This points to a significant mismatch between opportunity and expectation with enterprises setting the bar far too low when it comes to using IoT as a means of moving beyond basic operational efficiencies and into the realm of intuitive, self-governing business systems.
Not long ago, 4,000 was just about the right number of features to make Microsoft Office the dominant end user business application. But then came Google Gmail and a whole host of cloud-based, single-use apps, which proved that operational efficiency trumps bells and whistles. These highly focused applications didn’t try to solve every problem imaginable. They married just enough functionality with unbeatable scale, cost, performance and most of all promise to forever change our willingness to accept what was “good enough” right now, so long as there was the promise more somewhere down the road.
And now it appears the same thing is happening with analytics in general and with IoT in particular. We are quite willing it seems to accept a modicum of value in trade for speed, accessibility and simplicity. Case in point: The reliance upon hardcore BI software as a corporate necessity has given way to a number of smaller, more discrete means of deriving value from enterprise data, be that a direct SQL query, an auto-generated data discovery chart, or a live, interactive executive dashboard. I’m not suggesting that this is a negative trend that will prove ultimately harmful. On the contrary, it proves that data has been to a great extent democratized and is something now readily consumable across the enterprise.
However, this trend also points out an apparent problem with our current expectations of data and analytics as seen through the lens of IoT. In studying buyer expectations and experiences from more than 1,000 IoT practitioners, GlobalData found that the majority of users rely heavily upon basic reporting mechanisms as found in those aging, do-it-all business intelligence (BI) systems when analyzing IoT data.
Is there a significant mismatch between opportunity and expectation with enterprises setting the bar far too low when it comes to IoT? Isn’t the instrumentation of “things” supposed to let us anticipate and respond to the unknown? How can it do that if our primary means of understanding this instrumented data comes from pre-built dashboards or reports that show us nothing beyond what we know or expect to discover? The answer to these questions ultimately lies within the realm of artificial intelligence (AI), something we’ll return to in a future post. But before any of that, before an instrumented fleet of trucks can run itself, IoT needs to create transparency. That is the first step and the only means of improving the efficiency of an instrumented business.
It’s no wonder then that 43% of those surveyed in our 2017 IoT survey cited the improvement of operational efficiencies as the number one reason for investing in IoT. Stakeholders simply want first and foremost to know what’s actually happening. That’s the beginning of optimization. Using those aging, feature-laden, centralized BI platforms, businesses can easily apply a little bit of predictive analytics and some basic AI such as machine learning (ML) to continuously analyze and then improve the entire system according to well-defined KPIs.
That’s exactly what 41% of our survey takers indicated as their preferred means of plying AI – not as a means of leaping into the realm of intuitive, self-governing business systems but instead to simply make their businesses run better. As far as starting points go, that’s certainly good enough for now.