Define Your Business Case for Streaming Analytics
For a successful streaming analytics project, first you need to identify which business processes can benefit. Use this list of questions to frame your business case.
- By David Loshin
- August 29, 2016
Today's enterprise data arrives continuously from a spectrum of machines, sensors, mobile devices, social media feeds, weather reports, telecommunications network devices, and utility network meters, among other sources. Organizations are beginning to realize the potential business value of analyzing data in motion.
Streaming analytics is an integral aspect of a predictive analytics capability -- it allows you to ingest and analyze a broad range of data streams in real time while leveraging captured historical data.
The success of a streaming analytics program is critically bound to establishing a proper business case. To design and develop a streaming analytics project, you should start by identifying the business processes that can benefit, such as optimizing manufacturing production, increasing production quality, increasing sales revenue, or improving logistics.
Your goal is to identify which parts of a business process sit at the nexus of sets of data streams and determine whether the introduction of analytics can influence the process's performance.
Common Factors for a Streaming Business Case
Finding a scenario where continuous analysis of streaming data can positively impact your business depends on recognizing a number of common factors.
Reliance on continuous data: The business process continuously generates and communicates information.
For example, power utility companies rely on a network of devices that monitor energy transmission and use, linking the source of power generation, the methods of distribution, and the use at the end point. In addition, each node within the power grid can communicate its status on a regular basis. These data points are necessary to deliver power to users and ensure their consumption is properly logged for invoicing.
Opportunities for improvement: There are recognized deficiencies in the outcomes of the business process, and those weaknesses can be corrected using predictive analytics.
As a simple example, a trucking business can be impacted when the drivers are not aware of weather or traffic incidents along the route. This can lead to delays in deliveries.
Potential use of analytics results: The business process could improve outcomes in real time with the knowledge provided by real-time analytics.
To return to the power utility example, integrating information about the current state of all aspects of the energy transmission route can to enable real-time analysis of the fastest routes, enabling preemptive rerouting of power deliveries so to not overload circuits or minimize outages due to high demand.
Ask These Questions to Define Business Context
To frame it another way, identifying a strong business case requires answering these questions:
- What aspects of the business can be improved? In our shipping example, there are two aspects: faster delivery time and better prediction of delivery time.
- What information can influence improved performance? In the power utility example, that includes knowledge of capacity and potential high loads along transmission lines and data about alternative routes.
- How would that information be integrated into the business process? For our trucking example, accumulation of the real-time weather and traffic data at a central point allows analysis to be performed and results to be communicated in real time to the truckers on the road.
Once you identify the business context, your organization can determine what data streams are available, how you will manage and process that data, and what kinds of analytics to perform. Proper evaluation of the business context helps ensure that the delivery of analytics results can be directly aligned with your business process improvements.
David Loshin is a recognized thought leader in the areas of data quality and governance, master data management, and business intelligence. David is a prolific author regarding BI best practices via the expert channel at BeyeNETWORK and numerous books on BI and data quality. His valuable MDM insights can be found in his book, Master Data Management, which has been endorsed by data management industry leaders.