How to wrangle meaning from Internet of Things data

Without careful forethought, you will miss out on critical new insights

This vendor-written tech primer has been edited by Network World to eliminate product promotion, but readers should note it will likely favor the submitter’s approach.

The Internet of Things (IoT) promises to produce troves of valuable, fast moving, real-time data, offering insights that can change the way we engage with everyday objects and technologies, amplify our business acumen, and improve the efficiencies of the machines, large and small, wearable and walkable, that run our world.

But without careful, holistic forethought about how to manage a variety of data sources and types, businesses will not only miss out on critical insights, but fall behind the status quo. Here’s how to get prepared to wrangle and extract meaning from all of the data that’s headed your way:

* Automate the data gathering process:  IoT technologies, like smart monitors, cameras and sensors, often deliver a steady stream of time-stamped and geospatial data that enable us to keep a constant pulse on anything and everything – from our heart rate to the chemical compositions of soil on a farm.

To take advantage of the real-time nature of IoT data, automation will be key to data collection. The sheer volume and speed of IoT data is overwhelming enough, but the varying types of data that companies strive to gather at once –a mix of structured and unstructured, real-time and historical – will require tools that can ingest, process, mash-up and re-deliver data automatically and without manual efforts.

Often this means running simple automated analysis on real-time data before feeding it into a main database in an effort to manage data volume. For example, rather than an activity sensor directly feeding per-second measurements to a main database, measurements could be automatically averaged over 30-second intervals and then input into a primary database for analysis. Nimble data integration systems are able to automatically prepare and process data.

* Add context to IoT data with a diverse data ecosystem: With nearly every machine and object nearing some form of digital connectivity and tracking in the IoT universe, we are guaranteed plenty of new information to analyze and inform business decisions. However, IoT data shouldn’t work alone. Traditional data you’ve been gathering for years, like sales, campaign, marketing and social media data can contextualize information coming from devices and sensors.

When mapping your data framework, plan for data to come from a variety of sources, including relational database systems, raw text files, Excel spreadsheets, emerging IoT data sources and those that have not yet been developed. True IoT value comes from the integration of all of these sources, so look to tools that are strong in marrying a range of data types.

* Make connectivity your default: IoT data achieves its highest value when it connects with a range of data sources. When a factory’s machinery sensors connect with maintenance services, for example, essential repairs and upkeep can be completed proactively in real-time, requiring minimal production interruption.

Accessing and unlocking the multiple external data sources you need to really move the needle when it comes to optimizing operations or increasing sales, however, can be a challenge. Be sure the tools on your side can pull all relevant information in. Research software that offers a broad spectrum of available and effective connectivity options, such as pre-built connectors, APIs and other outside tools. Knowing the cost, quantity and readiness of these options can also speed up implementation of their offerings and, in the long run, reduce overhead.

* Give your entire team access to the data of things: IoT data can inform a range of business functions. A ride-share company can use time and geospatial data to implement peak activity pricing models, location-based marketing efforts and driver recruitment. Find data integration tools that marketing, sales and finance teams can access as easily as your engineers to ensure that you are getting the most value from the framework that you’ve worked so thoughtfully to create. Look for features like drag-and-drop interfaces that offer intuitive means of functionality for non-technical users.

* Prioritize flexibility: As the world around us becomes increasingly connected, the volume and types of data will evolve rapidly. We can’t predict what the future brings, but we can select tools that show a flexible ability to manage custom, unique and technical challenges. Prioritize options that give you the options of creating and managing custom connectors, using specialized coding and accessing a database layer.

It’s clear that IoT technologies are making their presence felt in the marketplace. The average organization using IoT data saw its volume of data grow by 30% in the last year, according to a report from the Aberdeen Group. And yet, 54% of those organizations find their data analysis capabilities insufficient.

The insights available through IoT data will remain obscured in so many data tables unless businesses take a thoughtful approach to their data architecture. Now is the time to invest in finding data integration tools that are prepared, powerful and flexible enough to tackle the challenges IoT is sure to bring.  

TMMData is a data integration and preparation software company.

Join the Computerworld newsletter!

Error: Please check your email address.

More about Aberdeen GroupExcel

Show Comments