Studies show that around 40% of products fail. But what if product designers could understand what features are most and least popular, which components tend to fail sooner than others, and how customers actually use products versus how designers think they use them? And, what if product developers could then utilize these insights to develop products that perform better, potentially cost less and, most importantly, are aligned with actual customer needs?
Innovative product development teams in pretty much every industry are beginning to look at ways to translate enormous streams of real time machine data into actionable information to improve the product development process by understanding where product innovation is necessary, which features are most desirable, and how to lower their overall cost of ownership.
For some time now, many companies have embedded sensors in their products that are producing huge amounts of information about performance. However, this machine data has traditionally been difficult to collect and analyze, due to both the enormous amounts of data involved and the different types. Data comes in a variety of formats, such as text logs, XML, JSON, CSV or SNMP. There are different data class categories, like event messages, configuration blobs or statistical dumps. Data is likely to be in different protocols, such as email, FTP, SFTP, as a stream or as a batch log file.
But advanced analytics companies have developed new solutions that are able to handle the volumes and disparate types of data involved in real time, making machine data analytics practical and affordable for a much wider range of organizations.
Achieving a single point of truth
In order to develop and market new products most effectively, you need to create a "single point of truth," or a body of data and insights that is comprehensive, accurate and timely.
These data and insights will provide all disciplines within the company involved in designing, manufacturing and marketing a new product the information they need to make critical decisions – product features, pricing, distribution and related functions. Essential elements of an information platform to provide this single point of truth include:
- A centralized data repository that can capture terabytes of structured and unstructured operational data
- Analytical tools that can describe and create meaning and relationships between elements in the data
- Reporting capability on how customers are using existing products, performance information on components within existing products, et al.
- Dashboards and similar elements that enable new product development and other teams to create customer analyses and drill downs
Streetline harnesses machine data
Streetline is a good example of how a company is using what some are calling IoT analytics to build out new products. The company recently launched Streetline IoT Gateway, part of the company’s mission to create smart cities and smart campuses. Analyzing the data streaming into the Gateway, Streetline realized they could build a successful business by harnessing smart data and advanced analytics to help cities, universities and campuses provide guidance on available parking and parking enforcement.
The Gateway allows Streetline customers to more effectively capture data from in-ground sensors and video cameras, and make more effective decisions. Better parking is the first solution to come out of the Gateway, and is not only a large revenue generator for cities, but has the potential to decrease congestion, and build a nice revenue stream for Streetline at the same time.
Streetline developed its new gateway jointly with Cisco. The Gateway integrates Streetline’s low-power mesh network and Cisco’s Smart+Connected WiFi solution.
Another example is Trane, which designs and produces HVAC equipment. The company began to insert sensors into its equipment and the data analysis enabled facility managers to think of and operate various HVAC and other devices as a single large system. For example, rather than simply turning on at the beginning of the day and off at the end, a “smart” HVAC system will now operate based on the building’s occupancy. In the future, Trane plans to sell “outcomes” versus equipment; e.g., ensuring the air within a facility remains at 72 degrees.
IoT analytics enables new product teams to completely rethink the solutions they develop and how they are supported. With IoT analytics, the only limit to designing winning new products is the creativity of the new product development team.
At Glassbeam, Puneet and team are focused on a mission to create a new category of Big Data business app around machine data analytics. The premise is that any high technology product, network or cloud is increasingly capable of generating copious amounts of rich but unstructured machine data, that once mined can provide tremendous business and operational insights for its manufacturers and users.