Most articles about the internet of things and its impact on business intelligence are targeted towards an IT audience and focus on real-time data ingestion, big data technology and analytic solutions for data scientists. And while these are critically important elements of the IoT/BI landscape, a discussion about delivering direct business value is often missing.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
It’s important to shift the focus of discussion to the solutions required to drive clear, tangible business value with IoT and BI. Deriving business value by achieving greater efficiencies, creating revenue streams and increasing profit for an enterprise can only be achieved if these solutions are focused on business users and end customers. IoT driven BI solutions must be embedded into the business user’s everyday work experience and be designed for an entire enterprise ecosystem.
There is much talk in the industry about how the wealth of IoT data can, for example, provide early warnings to enterprises to better serve their customers, a relatively simple task if the IoT device is reporting an error condition. But more likely, the IoT device is reporting telemetry about usage, operating conditions and performance. This is great data, but how will it drive better business actions that result in direct business value? This is especially difficult when the data is delivered to a small set of data scientists and analysts versus the people who have the ability to take action to implement change.
Placing this burden on a business user to examine the data for potential problems or opportunities and take action is clearly unreasonable. A business user cannot possibly process the tremendous amount or diversity of data, nor will they be able to identify the barely perceptible patterns that could be that ever important early warning sign.
This is a perfect scenario for machine learning. Machine learning needs to augment an IoT-driven BI solution to process and identify those hard-to-perceive patterns to guide the business user towards action. In fact, machine learning can be applied to deliver suggestive and prescriptive analytics to directly influence better business actions.
But machine learning alone will not ensure that this augmented insight will lead to a better business action if it is not embedded at the point-of-work. Ultimately, an actionable insight is dependent on “location, location, location.” When it comes to business users, the insights to drive better business actions must be embedded where that business user works. When done properly, the combination of embedding and machine-learning capabilities can deliver in-context automation, recommendations and insights to help the business user drive tangible benefits for their company.
Consider, each New Year many people make resolutions and sign up for new gym memberships which go unused or are cancelled by midyear. Gyms of course have access to when one enters the gym but have little other data around what the member actually accomplished during his visit. If gyms were to take advantage of the wealth of IoT data increasingly coming from their fitness equipment, wearable technologies and facilities, machine learning augmented analytics could proactively identify patterns indicating customer churn so that targeted recommendations and interactions can be made to increase “customer stickiness.”
Consequently, the business intelligence platform must be designed specifically for application integration and embedded delivery to ensure that the IoT-driven insight is contextually delivered at the point-of-work. Otherwise, if the insight is located elsewhere, such as in a standalone dashboard or delivered solely as an email alert, there is a high probability that the IoT-driven insight, no matter how valuable, will not result in a better business action and business value will be forever lost.
With the tremendous amount of IoT data being collected, the enterprise is in a perfect position to become an “insights as a service” provider and deliver data products to monetize their IoT data. For example, a manufacturing company can provide IoT-driven benchmarking solutions to its ecosystem of service providers.
Let’s further our example of the gym by considering a manufacturer of fitness equipment. In the manufacturing of fitness equipment, there are potentially hundreds of products made of thousands of parts that could potentially fail inside each gym. Since fitness equipment manufacturers are beginning to add IoT to their products, they can track how often they are used, how they perform and when they fail. The fitness equipment manufacturer could provide benchmarks around performance, lifespan and failure based on usage and environment to provide competitive advantage for service providers. Such information would allow the service provider to focus on maximizing longevity, quality of experience and availability of fitness equipment for gym members with proactive servicing and repair.
But, being an insights as a service provider is more than just creating an API endpoint to access the raw data; it needs to be curated, governed, secured, semantically consistent and accessible to business users. In other words, the business intelligence platform needs to be designed for secure, scalable, analytically consistent and easy-to-use distribution. This is not “your father’s typical BI solution,” but requires a new breed of business intelligence.
A BI platform designed for the enterprise and the enterprise ecosystem must consider deployed analytics, standalone or embedded, as a product. And just like any product delivered as a service, it must have means for provisioning, user management, security, lifecycle management, billing and continuous improvement.
There is much opportunity for IoT and business intelligence to drive direct and indirect value for an enterprise, but it requires the thoughtful application of solutions focused on business users and end customers and less on “techy” solutions focused on the analytically elite.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.