News Stay informed about the latest enterprise technology news and product updates.

Internet of things data proving to be a hard thing to use

Despite high hopes for IoT data, most connected device developers and manufacturers are having a tough time using it.

If the value derived from internet of things data is so beneficial, why aren't more companies using the data to...

its full potential?

In a study conducted by Forrester Research and commissioned by Xively, researchers found 51% of companies focused on manufacturing or developing IoT devices were collecting data for future use, yet only one-third were using the data for anything at present.

While this statistic is low -- even if it is an improvement from McKinsey's oft-cited statistic that only 1% of all IoT data collected is ever used -- it proves there is a wide gap between the potential of IoT and companies' ability to seize that potential.

"It's a concrete example of how there's opportunity in IoT, but if you approach it the wrong way, you're going to be putting a lot of work and exercise into something that's not going to have a really good payoff," said Ryan Lester, director of IoT strategy at Xively.

Internet of things data: Goals versus reality

There's opportunity in IoT, but if you approach it the wrong way, you're going to be putting a lot of work and exercise into something that's not going to have a really good payoff.
Ryan Lesterdirector of IoT strategy, Xively

When asked about their motives for creating connected products, survey respondents said competitive differentiation, new product features and expanding revenue were top business drivers. They also cited four key use cases of connected product data: integration with existing systems; profiling and segmenting customers; providing feedback to improve product features; and providing insight to internal teams, partners and suppliers.

However, despite the well-defined goals, the majority of the 232 companies polled weren't using the internet of things data as planned, which can lead to a great deal of data waste -- and wasted opportunity.

Of those using the data, 34% claimed it was to profile or segment customers, 33% to provide data-driven insights to stakeholders and partners, and 32% to personalize customer interactions.

Why the gap?

"Everyone is drowning in data," said Mike Gualtieri, principal analyst at Forrester. "The ability of devices to connect and generate data far outpaces the ability of organizations to analyze it."

Bruce Sinclair, president of Iot-Inc., attributed the gap to not having a clear plan for internet of things data. "A value plan qualifies and quantifies the value you will be creating with your product," Sinclair said. "Once done, you will know the information you need to create, and from there, you will know the data you need to capture. It's a romantic notion to think that you can collect data, and then go back to it later and analytically mine it for gold. You need a plan upfront."

A lack of talent also holds many companies back from prospering from the insight of IoT device data. "If they are capturing data without at least getting advice, they are wasting time and money," Sinclair said. "The data will just pile up."

Preventing IoT data waste

One of the survey respondents said, "In the next 10 years, [IoT] will focus around who owns the data. The market will be shaped in this way. We will start to see an expected value from all data we are collecting."

The ability of devices to connect and generate data far outpaces the ability of organizations to analyze it.
Mike Gualtieriprincipal analyst at Forrester

However, to get there, a few things have to change.

"Company leadership must insist on a business plan before building their project," Sinclair said. "Don't let a vendor drive your decision of what data to collect. This needs to be a top-down process, not one that starts with the sensors and what data they can capture."

In addition to knowing how you are going to use the data to begin with, companies must also think of how other departments may be able to leverage the collected information in future use cases. But this leads to several questions, Lester said, including how to make the data available to others and integrate it into siloed teams' tools.

"You don't know how you're going to want to use the data down the road, so looking for platforms that offer predefined integration into other tools you're using today -- ERP, CRM [customer relationship management], billing programs, etc. -- having those customer integrations and APIs that allow you to get the data out and put it into others tools, that's really important," Lester said.

Gualtieri noted the specialized skills and platforms needed to properly analyze internet of things data aren't always available in-house. "Advanced analytics such as machine learning shows the most promise to enable a system of connected devices to learn over time, but it requires data science and machine learning tools," he said. Finding vendors and services that support a perishable insight framework, embed analytics to automate real-time decisions and support human decision making are critical to making sure collected data pays off for the collecting companies, Gualtieri said.

Sinclair agreed. "Companies must have the talent to transform the data into value," he said. "If it does not have it in-house initially, the company needs to find it externally."

An appliance manufacturer that responded to the survey shared the sentiment. "The future of the IoT world depends on where you draw the partnership line. We can't and don't need to do everything ourselves."

Next Steps

Looking at IoT data and analytics

Why IoT is more than just connected devices

A guide to enterprise IoT project success

Dig Deeper on Internet of Things (IoT) Data Management

Join the conversation

2 comments

Send me notifications when other members comment.

Please create a username to comment.

How does your company use internet of things data?
Cancel
I really like this quote...
It's a romantic notion to think that you can collect data, and then go back to it later and analytically mine it for gold.

In the area of IoT that we work (industrial wireless sensors), we see companies collecting huge amounts of data from their machines - and never using it. Or, giving it to PhD computer scientist that have never stepped into a factory.

We have had success by listening to the pain points of the people that call us that are looking for a solution. For example, a motor failing and causing a machine to fail at a critical time. By working with the person that maintains that machine, we can quickly "dial in" what sensing data is needed and set alerts based on the maintenance person's familiarity with the machine. We can then use the on-board "edge computing" power of the sensor to transmit data only: when an alert sensor reading is exceeded (often from multiple sensors) or during a once-a-day heartbeat to assure that the sensor is working. This keeps the data way down but provides high value.

This approach comes from my early career as a maintenance manager at a large factory. It didn't talk long before I could tell a failure was coming by the way the floor was vibrating or the temperature of a machine surface that I touched as I walked by each day.

We have worked with several large organizations that have hired large staffs of computer scientists to try to "tease out" impending failure from huge amounts of data - and that approach rarely works - for many reasons including: computer science people don't often understand what an industrial motor does before it fails.

Our simpler and practical approach to work with the people that understand the process and then put the data filtering in the smart-sensor really helps with acceptance because the database stays a very reasonable size and when an alert is sent, it is taken seriously by the people that helped create it. Often we implement our "small data" system in parallel to established systems to get them implemented quickly and showing a fast ROI (and to get the person that called us quickly out of hot water).

Scott Dalgleish
CEO, Phase IV Engineering
Cancel

-ADS BY GOOGLE

SearchCIO

SearchSecurity

SearchNetworking

SearchDataCenter

SearchDataManagement

Close