Data Fabrics Bring the Power of Data to the People
Editor's note: This article originally appeared in Forbes.
In today’s fast-paced world of competing business priorities, the capacity to enable self-service data analytics with right-sized data governance is key. This ability removes the structural barriers between IT-managed data environments and true, businesswide data-driven decision making.
Data fabrics—AI-based data management designed for federated environments—are the connective tissue between data, infrastructure, and software. Data fabrics support data discovery, linkability, quality, and governance, while also assuring agile, trusted information business use at scale.
The problem data fabrics are designed to solve
As business leaders laud the value of Data Cultures and set their North Stars to data-driven everything, they eventually face an unfortunate reality: Their enterprise data environments are outmatched by the demands on them. There’s too much of too many different kinds of data stashed in too many different places to support a meaningful data transformation. It takes specialized skills to extract data from an environment like this—and even then, you can’t trust it.
This is such a common situation that more than half of respondents to recent IDC research said they were overwhelmed by the amount of data they work with, while almost as many—44%—said they don’t have enough data to support decision making.1 The 83% of CEOs who want their companies to be more data-driven need look no further to know why their efforts fail.2
This is the problem that data fabrics are designed to solve. Data fabrics are a modern data management approach built on the assumption that data proliferation and decentralization will continue, so traditional methods of managing data through centrally managed repositories are doomed to fail. Instead, data fabrics assert federated governance and leverage AI to intelligently and dynamically connect disparate data sources across an enterprise, index them, and make them available for data analytics use as needed.
Data fabrics integrate with existing architectures while being nimble enough to connect new data sources as they emerge. Coupled with a robust data analytics platform, data fabrics open up the potential of self-service analytics, empowering everyone to use data with AI-powered predictions, what-if scenario planning, guided model building, insights, and other data science techniques—all with clicks, not code.
Data processing at the speed of business
Data consumers spend most of their time searching for data and prepping it, with little left over to explore data and find fresh insights. It’s a significant and common problem: According to a report from Wakefield and Elastic, more than half of American office professionals say they spend more time searching for files than working.3 In other words, your employees are stuck doing the work that data platforms should be doing.
Data fabrics are equipped to automate some data prep and discovery processes, including pipeline management and intelligent data discovery.
The Lake County Health Department and Community Health Center near Chicago developed a first-generation data fabric to drive self-service analytics and reduce the amount of engineering, computing, and planning resources necessary for producing reports. The center connected 20 data sources and 36 applications, reducing the report maintenance burden to almost nothing while expanding self-service analytics to more than 900 employees.
Lake County Health Department officials found that the streamlined reporting process native to Tableau’s platform enabled faster decision-making and data project development. By shrinking production time and rapidly delivering new solutions, users grew confidence in the process and in leveraging data in their jobs. The program rapidly gained momentum, fueling productivity and excitement around analytics and fostering a strong Data Culture within the organization.
“It's made us completely rethink how we design our data architecture—moving away from a warehouse strategy and more toward ad hoc, where the analysts can manipulate and prepare the data in more agile ways that respond to business needs more quickly and effectively,” explained Jefferson Mcmillan-Wilhoit, health informatics and technology director at the Lake County Health Department and Community Health Center.
Future-proofing through collaboration
Data consumers and IT data managers typically worked independently from one another, effectively decoupling the business’s data needs from IT’s governance and security rule-making. This type of isolation is becoming outdated quickly, as businesses become increasingly invested in data management and governance.
According to Gartner, “By 2023, organizations with shared ontology, semantics, governance and stewardship processes to enable inter-enterprise data sharing will outperform those that don’t.”4
Steven Hittle, vice president and BI innovation leader for JP Morgan Chase (JPMC), said he worked closely with his business stakeholders and designed the JPMC data management architecture around their needs and with the goal of enabling self-service.
“I would rather create a platform that allows the business to solve their own problems, because we [IT] will never know them all. That’s how I’ve approached our Tableau usage—simply because it allows them to connect to and analyze their own data, which they were already doing,” he said.
This is data fabric at its core: uniting business users and IT departments around a shared vision of high-quality, integrated data for all, regardless of where data is located, without sacrificing governance and security.
Get more information on how you can use data fabric to synthesize your data, and read the Forrester Report to determine how to select a data fabric provider.