Philosophers have struggled for centuries to articulate a definition of the word “intelligence.” Aristotle considered it the highest of human virtues, and the root of rational thought, and indeed, the meaning of its original Latin word, intellectus, translates to understanding.
Another definition, often attributed to the late astronomer Stephen Hawking, describes intelligence as “the ability to adapt to change.” It’s a view that adheres closely to that of Greg McStravick, president of database and data management at SAP when he gets to talking about the intelligent enterprise: “In the same way that humans adapt and learn from experience and information, businesses can learn and adapt in order to grow.”
But whether you’re talking about a person or a business, there’s a catch: The only way to understand something is if you have the right information about it, the right data. And in business, getting the right data isn’t easy.
“The only way to change and grow is through data coupled with emerging technologies such as artificial intelligence and machine learning,” McStravick says. “A solution that helps companies manage large volumes of data from disparate sources to gain valuable insights is the foundational requirement for the intelligent enterprise.”
Businesses gather a massive amount of data about every aspect of their operations. Study the data close enough, the thinking goes, and eventually, patterns emerge that lead to insights that in turn lead to more informed business decisions. And those decisions usually aim for one of two outcomes: increase revenues or reduce costs.
So it’s no coincidence that data has been called the oil of the 21st century. The comparison runs deep. Before it’s useful, oil must first be refined into valuable items like gasoline or jet fuel. For it to have any value, McStravick says, data must also be refined and analyzed: “You can’t fuel an intelligent enterprise without data. It’s like a car with no gas.”
The comparisons end there. Access to valuable, actionable data is difficult for many reasons. For one thing, it’s usually spread all over the place. Most enterprises store their data in a fragmented manner in six to eight different cloud systems, each running different applications. This “data sprawl” leaves companies without the meaningful insights and understanding of their customers, suppliers, and even their own products. And that leads to uninformed business decisions.
This sprawling data landscape is also hard to govern and to protect. Both problems add a new layer of risk and liability: New regulatory schemes such as GDPR in the E.U. impose stiff financial penalties for data misuse.
SAP’s vision is to provide an enterprise with a common data model that brings together all data types from many sources without moving anything.
“Once the data has been captured and processed, it becomes trustworthy, and only then can it be used for analytical and computing purposes,” McStravick explains. “The elegance is in combining large volumes of data outside of core systems … to allow computation on data at the point where it resides without having to replicate it.” That makes it easier to perform analytics or to feed the data into a machine learning system.
An example of enterprise intelligence
Companies in a range of industries are on the journey to intelligence. “Some energy companies are already leveraging geospatial capabilities [of an in-memory platform] to identify all impacts on their grid and to manage the subset of the grid infrastructure on a graph basis,” says McStravick.
He goes on to explain that to achieve the first layer of intelligence, a company would load all data on customers, plants, and homes using their utility services by longitude and latitude into the in-memory system to pinpoint exactly where their assets are. Events like service outages can be immediately identified by location and managed.
“That would normally take hours; with [an in-memory platform] it takes sub-seconds,” McStravick says. “In the second layer, the company could start to leverage other sources of data to predict the likelihood of outages, based on earth observation and weather data.”
It is this capability that would enable them to intelligently and proactively predict outages, structure outage teams, or schedule predictive maintenance.
A phased approach
One big issue facing enterprises on the journey to intelligence is the fact that their enterprise resource planning (ERP) systems are aging, and their master data is not harmonized, making it difficult to access and understand information coming from disparate sources.
For example, any company that has dozens of manufacturing sites worldwide generates huge data volumes in a complex landscape. Achieving standardized reports is a challenge because often tables are stored in different systems, and a lot of manual tools are needed for mapping them. Addressing this challenge is a central data repository and a unified stream of clean data, allowing enterprises to analyze and process to make decisions.
Intelligent enterprises are on a long-term journey that usually involves a phased approach. First, they need an operational platform to automate processes and allow access to all data, which may have been physically and logically dispersed at one time. Next, they can start making better and quicker tactical decisions using trustworthy information. And finally, they can proceed to advanced business decision-making for strategical impact.
Find out about SAP HANA Data Management Suite; providing the data framework, data insight, and data governance to help your company understand itself, with everything your enterprise needs to learn and grow, built in.
Discover data governance solutions.