Analytics and the Real-Time Revolution

Page 1 of 2 next >>

For more and more organizations, real time is not only the right time—it’s the only time that matters. No longer can decision makers afford to only understand the material events affecting their businesses a month, a day, or even just a few hours after the fact. They need a sense of what is happening in the moment to be able to predict and adjust for events that are arising.

The value of real-time data “is that it provides needed information to accelerate business insights so teams can make better informed decisions and avoid the issues around using older data,” said Tarun Chopra, vice president of product management for data and AI at IBM.

The good news is that technology solutions have caught up with this need. “Advancements have made it easier to collect data in real time or, even better, to analyze it while they’re still in motion,” said Saurabh Mishra, director of IoT product management at SAS. “With this amount of information, the quicker you can act based on the data, the better decisions you’ll make.”

The data is streaming in from a wide range of sources and platforms. It’s originating from CRM, finance, marketing automation, operations, IoT, and products, said Sam Pierson, senior vice president, engineering data, at Qlik. The data also streams from “different locations ranging from the cloud, hybrid multi-cloud, on-premises, and edge devices.”

Critical enterprise applications—from financials to supply chains—are increasingly relying on AI and related advanced analytics for at-the-moment visibility. Concurrently, the two worlds of enterprise IT—user-facing applications and analytics—are converging. To date, “user-facing applications are inherently real time, whereas analytics are more batch-driven in order to handle the volume of data in the enterprise,” said Matt McLarty, CTO of Boomi. “AI will accelerate the convergence of these two worlds. Most organizations are not yet ready for this future but have the fundamental pieces in place. Once the two worlds converge, there will be a realization that there aren’t data sources and targets, which is how data engineering perceives this world. There is a collection of data endpoints that communicate bidirectionally, and real-time data streams are essential in that setting.”


How real is real time? Use cases with high levels of ROI abound, industry observers point out. “The industries that tend to invest in real-time data the most include finance, environmental telemetry that covers weather, working space conditions, and farm animal health monitoring, soil, and crops,” said Ryan Booz, PostgreSQL advocate at Redgate. Additional use cases include “physical asset management that includes predictive maintenance of high-value assets to avoid downtime causing failure and delivery logistics.”

Telemetry, in particular, stands out as an example of real-time data in action. “Whether it’s an extreme case like the data gathered on Formula 1 racing cars in real time to help the pit crews, or just more common tracking of drivers out for delivery, we’re seeing a lot more data growth on this kind of data,” said Grant Fritchey, product advocate for Redgate.

Real-time monitoring and analytics for sensors and machinery is also a prominent use case for real-time analytics. “The evolution of computer vision and neural networks, plus the cost reductions in accelerated hardware, all play a role in how much companies adopt this technology,” said Mishra. “Other valuable real-time data use cases include production quality and optimization, fraud detection and prevention, personalized marketing, and customer experience optimization.” Demand for real-time operational data is seen across many areas of the business, “especially in monitoring asset utilization rates or to predict bottlenecks before they become a problem,” said Lee An Schommer, chief product officer for insightsoftware.

“The CFO and finance team are responsible for communicating metrics and other financial information accurately and effectively to a range of stakeholders. With real-time data, these teams can not only be assured that they are presenting the most accurate data to their stakeholders, but they can also run predictive analytics based on what is happening in the economy that day, week, month, or even year.”

There are numerous other applications that run on real-time data as well: “Digital twins, predictive maintenance, anomaly detection, and real-time process optimization all rely on real-time data to drive critical business insights,” said Mishra. For example, he added, real-time data streams can be used by city authorities to monitor and redirect traffic flow. The data, which is run through an AI model, consists of historical traffic patterns integrated with live data feeds.

Forecasting is another emerging area. One large U.S. food and beverage company “saw that due to unexpected supply chain disruptions, its forecasts weren’t accurately reflecting the actual consumer demand,” said Chopra. By adding real-time data feeds from new sources, it was able to achieve 75% greater accuracy in demand forecasting. “By harnessing new data sources, such as weather, local-focused trends, and mobile usage patterns, as well as internal shipments data, we helped build a forecast for product-specific and regional patterns.”

Another successful use case for forecasting and sales planning is helping companies streamline their workflows, providing end-to-end visibility across the supply chain and correlating data from siloed systems, said Chopra. Additional use cases coming online include “real-time monitoring of supply chain operations that reduce manual decision making, dynamic pricing optimization in the retail industry, fraud detection, along with personalized customer experiences that drive customer retention,” said Doug Kachelmuss, senior director of data and AI for Launch Consulting Group.

Page 1 of 2 next >>