4 Steps to Data Modernization that Drives Intelligent Business

Traditional data warehouses can no longer accommodate the volume and variety of data your organization is collecting

data modernization
iStock

As companies transform large portions of their business, they are prioritizing data modernization as a key step to drive intelligence and agility into operations and decision making.

A new IDG/Cognizant Explorer survey found data modernization and intelligence to be an important driver of cloud adoption. Data modernization and intelligence was the second-ranked priority for cloud migration efforts, trailing only experience transformation.

Data modernization is a central pillar of digital transformation because traditional architectures like data warehouses and data marts can’t accommodate the volume and variety of unstructured data being generated from email, social media, and Internet of Things (IoT) sensors and other devices. Moreover, the pre-defined reporting that characterizes traditional data warehouse initiatives delivers an historical view, as opposed to the real-time intelligence that is crucial for today’s fast-paced digital business environments.

“We’ve moved from huge repositories holding transactional data to an era where companies need machine learning algorithms, robotics, and linguistics to run their businesses and create new customer experiences,” says Michelle Wallig, data modernization transformation practice lead at Cognizant. “The old rigid data structures don’t work for that anymore.”

While traditional infrastructure won’t be wholly replaced, organizations need to transition away from operational data stores and business intelligence (BI) reporting tools to a more modern, cloud-based data architecture that encompasses graph databases, ingestion layers, scalable storage, visualization tools, and machine learning-powered analytics. For example, AWS’ Redshift data warehouse can access data directly from the data lakes without duplicating data, avoiding costly data redundancy.

In addition, enterprises should consider the following recommendations to ensure successful data modernization:

Don’t skimp on strategy. While it’s tempting to orchestrate a quick “lift and shift” migration to the cloud, Wallig contends a rushed approach will create challenges, especially if the effort simply transfers outdated or inefficient processes to the new platform. “Moving bad processes into the cloud is a waste of money,” she says. Instead, companies should take the time to understand the current landscape and lay out a strategic, multi-year migration plan.

Be clear on the intended business outcomes. Data modernization, in the context of a broader cloud migration initiative, needs to tie back to specific short- and long-term business objectives. It’s critical, therefore, to involve key business stakeholders during the planning stage.

Elevate business intelligence. Machine learning will help organizations move from traditional descriptive and diagnostic analytics activities to more advanced predictive and prescriptive capabilities. These advanced capabilities can then be applied to improving the business across three core areas: customers, products and services, and operations.

“Data modernization has to align with all of those areas, from strategy to implementation to intelligence elevation,” says Wallig. One advantage of AWS Redshift Spectrum is its use of S3, which provides access to all AWS services, including advanced business intelligence capabilities.

Emphasize education and change management. Successful data modernization will do more than improve the business; it will change the lives of employees by helping them to be more effective in their roles. Instead of spending time finding and gathering data, they will be able to quickly apply data to make better decisions that impact company performance or foster customer engagement. Employees may not fully grasp this benefit without a concerted educational and change management effort that showcases how their roles and responsibilities will change for the better.

Data modernization can take years to fully execute, but with the right planning and focus, companies can move methodically and iterate strategically, capturing plenty of benefits along the way. Here’s one example of how a leading multinational mining, metals, and petroleum company benefited from an AWS-based data and analytics platform:

Business drivers:

  • Determine significance and impact of various levers on tonnage by analyzing equipment downtime, throughput rate, inventory levels, and connected processes.
  • Analyze cause-effect relationship between downtimes and reasons for delay by using rank ordering and forecasting techniques.
  • Use Set-Point to perform diagnostic analysis of key drivers of performance, such as throughput rate analysis.

Solution highlights:

  • Preparation of data and selection of variable: Developed regression models to identify the impact of variables on daily tonnage and prioritized throughput and delays based on the predicted impact and not the size of the throughput or delay.
  • Variable impact analysis: Identified scheduled and unscheduled delays having significant impact and optimum direct load to achieve maximum tonnage.
  • Root cause analysis: Drill-down analysis to determine delays having maximum impact on tonnage.
  • Technology stack:
    • ETL - Redshift Scripts
    • File Layer – S3
    • Data Warehouse – Redshift
    • Reports – SpotFire

Business outcomes:

  • Elimination of operational bottlenecks and identify ROI impacts on company’s assets
  • 30% improvement in utilization at equipment level
  • 20-25% reduction in scheduled/non-scheduled delays
  • 7-10% increase in port inflow

Learn more about how Cognizant and AWS can help with your data modernization efforts.

Related:

Copyright © 2019 IDG Communications, Inc.