IoT in the Enterprise: Who’s Doing it, What’s Happening and What’s Working?

Whether in the mainstream media, the tech press or in business circles, a lot has been made of the Internet of Things (IoT). But IoT is about a lot more than internet connectivity for far-flung machines and devices. It’s about collecting sensor readings and other data from those “things.”

Why is providing value from that data so important? A confluence of factors in technology and business has given rise to IoT’s utility and the fascination around it.

 

Trends Giving Rise to IoT

The combination of cloud computing, streaming data, Big Data analytics, artificial intelligence, and the broader prospect of digital transformation have created a readiness in the market for generation, consumption, and analysis of time-series data that is machine- and sensor-produced.

While many of us may think of IoT in the consumer realm, conjuring up images of home automation and turning lights on and off with our personal digital assistants, it’s the industrial realm where IoT really shines.

For the most part, it’s sensor devices found in engine rooms, factories, elevators, automobiles (both self-driving and conventional), turbines, farms, and other industrial settings, that are leading the charge.

Sensors can report all types of conditions including temperature, moisture, pressure, the number of people in close proximity, the working order of a given component, or raw images and video.

Companies can monitor what these sensors are reporting, and can do so repeatedly, over very short intervals. While sensors used to just tell us if conditions were within a normal range, now they can do much more. Instead of just reading current conditions, this data can be saved and analyzed so that historical conditions can be correlated with certain phenomena.

Doing this facilitates the construction of predictive models so that certain sensor-reported conditions can be used to forecast specific phenomena. For example, patterns in the temperature of a piece of equipment may provide tell-tale signs of an impending failure.

And it probably goes without saying, being able to forecast such breakdowns before they happen can vastly reduce stress and revenue shortfalls stemming from lapses in operational continuity.

The above provides an operations-based explanation for why IoT is such a phenomenon, but what about the business reasons?

There are several. For one, companies are shifting their computing infrastructure from a mostly on-premises approach to a hybrid approach with some assets in the cloud. When analytics is one of the workloads that move to the cloud, things get teed up nicely for IoT.

Not only does cloud storage allow for arbitrarily large data sets, but it allows them to grow over time, even geometrically, as necessary.

And since sensor data originates from remote locations, storing it in the cloud works well. There’s no reason to store data on-premises that doesn’t originate there.

On the other hand, data that does originate from and is stored on-premises may be needed in the analysis work, along with the sensor data. That means the analytics system in play must be capable of both blending data from different data sets and working in a hybrid on-premises/cloud modality.

In fact, modern cloud-based analytics systems can do this very well, further facilitating the IoT workflows.

Perhaps the most important precondition for IoT, though, is the yearning businesses have to work with data in real-time, acting on new conditions as soon as they arise. IoT data—and the patterns for processing it—are completely aligned with that principle, often referred to as a data-driven culture.

Further, because IoT sensor readings are formatted as time-series data (successive point-in-time recordings of the sensor readings), they lend themselves well to AI and predictive analytics, both of which work extremely well with time-series data, as the predictions are often just additional, extrapolated data points in the series.

All of these technological developments and breakthroughs combine to serve as the cradle for digital transformation, a broad term that describes the journey to data mastery and data-driven outcomes.

In fact, when the essence of IoT is considered, digital transformation is what it’s all about: getting real-time data on the operations of a business, performing immediate descriptive and predictive analytics on it, and acting on the results.

 

Background on the Survey

With all of that in mind, Gigaom surveyed a large group of Enterprise customers (300 Enterprise IT Executives) to learn about the IoT initiatives in their organizations.

Our goals were to determine the inspiration for IoT initiatives, the business units responsible for pushing them, the groups that provided the funding, as well as those units that led them.

And beyond the quest for general knowledge about which business units play which role, we wanted to know how IT (information technology) and OT (operations technology) collaborated for IoT initiatives.

We also wanted to gain insight into the relative maturity of organizations’ IoT projects, and the organizations’ IoT maturity overall.

Also, as with all Gigaom surveys, we asked several categorizing questions—for example, size of the organization and the industry it works within—up-front, so that we could drill down to, and correlate answers to the other questions with these subgroupings.

Beyond learning who is driving IoT and what relative maturity of organizations doing IoT, we also set out to determine the scope of the projects, the general rate of success in their implementation, and how different companies implementing IoT approach vendor selection, as they proceed from planning to project implementation.

Like IoT itself, the overall goal of this survey was to discern things both descriptive and prescriptive. We wanted to see what customers have done, how much they’ve covered, how they’ve done it, and what impact these choices may have had on their success.

Read more: gigaom.com