5 min read Myth 1: The Internet of Things is just about connecting things IoT [...]
IoT analytics can be confusing. The Internet of Things (IoT) has become an over-marketed term and as a result, its importance is often classified as hype. But the fact is, product manufacturers in every industry are using IoT more and more to increase revenue, reduce costs and improve aftermarket services. By aggregating and analyzing sensor data from their connected products, product managers can better understand how/when/where their products are being used and, as a result, build better products.
Prior to Internet of Things (IoT), typical product lifecycles tracked by traditional PLM mostly covered pre-production states – manufacturing, distribution, shipping, purchasing, etc. Any relevant active life (operational) states such as system errors, failures or product issues that occurred after the product has been deployed were manually reported by the customers. Traditional tactics to alleviate this challenge included focus groups, surveys, field trials and after-the-fact observations. This lack of visibility in product usage data during the entire active life of the product resulted in delays in issue resolution, unnecessary truck rolls, avoidable system failures, loss of business, and ultimately, customer (dissatisfaction) churn.
IoT-enabled products have dramatically changed this process – now, with connected objects, the PLM process extends from pre-production all the way to live deployment. The data from connected products enables actionable lifecycle insights that allows product management, sales, marketing, operations, and C-level to gain end-to-end visibility on product performance and usage, build a better understanding of customer’s actual engagement and interactions, optimize their aftermarket service programs, and focus their future product roadmap.
The IoT is hence not about connecting things, it is all about analyzing data from these connected things.
Looking at the use of analytics in the web, mobile application and social spaces – it is evident that data-driven decision making is a crucial factor in determining the competitiveness and relevance of today’s companies. Most websites, mobile applications and social engines – new and existing – have analytics as an integral part of their product and business strategy.
But like with everything else, using the right tool for the right problem is the key to success. Web analytics is not the same as App analytics that is not the same as IoT analytics. To understand these differences, the following table compares four analytics domains from two angles, ie the process that they track and the KPIs that they analyse:
|Domain||Process tracked||KPI’s analysed|
|Web analytics (Google Analytics, Sprint Metrics Woopra,…)||Visitors > Visits > Pageviews > Events||Referrals, Bounce Rates, Exit Pages, Conversion Rates…|
|Mobile/App analytics (MixPanel, App Annie, Interana…)||Users > Sessions>Events||Session Lengths, Time in App, Acquisitions, Screen Flows, Lifetime Values…|
|Logs analytics (Splunk, Glassbeam…)||Machines/Applications> Data Logs||Compliance with security policies/audits, system troubleshooting, security incident response…|
|IoT analytics (Mnubo)||Devices>Events>Timeseries||Product usage, Brand Engagement, Supplies/Services Monitoring, Process Analysis & Optimization|
Organizations often underestimate the time and resources that go into building analytics. Even when object manufacturers and service providers possess most of the necessary software and hardware skills in-house, they rarely have experience with real (deployed) IoT projects. In most cases, they have limited experience working with the technology elements that are unique to the IoT. Designing, developing and deploying analytic insights in the world of IoT/sensor data requires skill sets that are different than most traditional software projects. These skills include (but are not limited to):
Recruiting some of these in-demand profiles could take several months. If there is no ramp up period at all (which is unlikely) and considering that software built in-house usually takes at least 12-18 months to get off ground, a very optimistic time-to-insight (i.e. some analytic insights in production) would be between 1.5 to 2 years! It is increasingly apparent that for connected object manufacturers/OEMs to operate as quickly as possible with a feedback loop between their data and relevant insights, ready-to-use IoT data analytics and data science capabilities is essential. It may even determine how quickly these organizations can grow or succeed.
IoT projects are only successful when a number of different departments within a company work together. Organizations as a whole must embrace a data-driven mentality to lead the smart product market. IoT projects break-up
silos within a large enterprise, empowering cross-department AI collaboration and a 360 degree view on their products and services.
While Big Data is all about more data, Analytics is about offering more actionable business insights. Given this, collecting data without a plan or strategy about how to use it drives useless increases in costs and complexities. More importantly, if data is incomplete or out of context, it can produce flawed insights and lead to decisions that could do more harm than good.
Before diving into any IoT data strategy, product manufacturers must ask themselves:
1. What endpoints will provide the data?
2. What data points should be collected?
3. Which analyses will generate strategic insights?
4. What additional services do I need to offer?
Find out how Mnubo is tapping into the benefits of product data by transforming raw sensor data into actionable business insights. Read more here!
Thanks for reading! If you have an opinion, issue or simply wish to elaborate on the article, feel free to participate in on Twitter!