Main content

How a modern data architecture supports agility and resilience

Brad Stillwell Profile picture for user Brad Stillwell May 26, 2020
In these times, every business needs maximum agility and resilience. Infor's Brad Stillwell explains how a modern data architecture can help

Data going into bottle, 3D in blue © 3dkombinat - shutterstock
(© 3dkombinat - shutterstock)

Today’s global economic upheaval underscores the critical need for enterprises to be agile and resilient. In most industries, change is constant and dramatic. It can also be debilitating for organizations lacking a flexible infrastructure and tools needed to make decisions quickly and accurately. As businesses worldwide adopt new go-to-market models, heightened supply chain expectations, and flexible workforce policies, their enterprise data architectures should also be examined and updated.

Leveraging a modern data architecture helps enterprises remain competitive in the new economy. Some mandates aren’t new. Accurate data has long been a component of budgeting and planning. But, today, strategic planning is not a once-a-year ritual involving reams of print-outs and hand-plotted charts. Analytics today must be:

  • in real-time, based on a company-wide version of the truth
  • highly consumable in easy-to-read formats
  • on-demand and user-friendly
  • unified, fully integrated to company-wide systems
  • cloud-based for elastic storage space
  • highly accurate and reliable, with clear definitions and sound, well-documented logic

These qualities are essential for building a company culture that is equipped to react with great speed and proactively address changing issues. Kaushal Vyas, Vice President of Product Management at Infor, sums up the issues that companies are facing:

Markets are changing at an unprecedented pace. New challenges are emerging every day that strain our already over-taxed resources and expose weaknesses in manual or cumbersome processes. Often this pushes current infrastructure and technologies to their limits, and sometimes beyond.

He adds that it doesn’t have to be this way. Enterprises can take advantage of analytics that are highly flexible and support important, rapid decision-making processes — at multiple layers in the organization, not just the top tier. He says this helps build resiliency – how quickly businesses can adjust to change, either negative or positive. 

Organizations strive to be future-proof. That means they need the ability to be elastic and bounce back after changes — returning to a ‘routine’ that is in a better position than before the change…

Resiliency means we need to be able to move fast, make decisions quickly, and then pivot to a new normal, rapidly, and with confidence.

Overcoming challenges

Three common challenges typically impede organizations with outdated data architectures:

  • Multiple data sources. The first is the inability to handle the different data sources and use cases that are involved in delivering analytics at scale. Most data platforms have evolved over time and are not prepared to manage more modern sources of information, such as machine to machine (M2M) data. Also, legacy infrastructures often force companies to fit volumes of data into predefined structures, even before the potential application is identified. This is limiting from day one.

  • High cost of scaling. The second challenge is an inability to rapidly scale -- coupled with high maintenance costs. For most legacy solutions, the enterprise pays for capacity, even if it’s unused. Cutting back on capacity, though, is high-risk and can leave the company unprepared. Upscaling when a need arises can take weeks or months to provision the necessary hardware and software. By then, the opportunity may be lost to a more agile competitor.

  • Bottlenecks in reporting. The third common challenge is an inability to discover new insights rapidly. Because of legacy technologies, cumbersome workflows and the need for highly specialized IT resources, creating new reports can be laborious. This can result in long lead times to obtain and ingest new sources of information. Without tools to create reports themselves, users can wait weeks in a queue for IT resources to become available. Bottlenecks in research can delay innovations and slow decisions.

How did we get here?

This slow-moving analytics structure wasn’t adopted intentionally but was a product of evolution. For many companies, the original goal was to build a repository with all the information necessary to make decisions in critical business categories. But, processes and tools were often bolted on, creating a haphazard structure of multiple tools, multiple extracts, multiple data stores, multiple teams working on multiple data silos. Projects were often stalled as stakeholders argue over who has the right data. With long lead times needed to obtain relevant data, decision making was slow. 

Some detours in the journey to modern analytics have plagued some companies as they experimented with short-term solutions. For example, first-generation data lake technology offered the ability to mash-up multiple forms of data — by standing up massive farms of commodity hardware servers. This, too, had limitations. 

User adoption was a challenge. Because the toolsets were primarily designed for developers and data scientist, users had to be highly skilled to generate new reports. The IT team became the gate keepers — or bottlenecks. Also, some early data solutions weren’t built for dynamic needs, frustrating front-line users.

A lack of governance or semantic layer also added complexity. Every new use case required starting at step one, leading to mistrust and teams resorting to their own data tracking, often using spreadsheets. 

The modern solution

Enterprises now have more options. Here are some guidelines to help distinguish a modern solution from an outdated one, which will hinder agility.

  • Results-driven. A modern data architecture is business centric, not IT focused. It isn’t about technology for the sake of technology. It is about driving better business outcomes. It is focused on the needs of the business over the technology that enables business success.
  • Automated. Next, a modern data architecture leverages automation. It seeks to augment and automate the most manual tasks to ensure we are not building brittle processes.
  • Flexible and elastic. A modern system should be flexible enough to address the use cases, which aren’t even envisioned yet. A modern data architecture is also elastic, leveraging the power of cloud computing to provide instant, on-demand scalability, ensuring that capacity is always available.
  • Adaptable. The modern solution should be able to adapt to the changing needs of the business and the changing landscape of the enterprise. With a semantic layer that can be updated, the company can add definitions and parameters as the needs of the company expand. This means the company isn’t locked into the way work is being done today.
  • Smart. Today’s solutions should leverage the power of artificial intelligence (AI) and machine learning (ML) to operationalize automated insights. AI-driven functionality can help users discover previously undetected insights about data, spotting trends and patterns that can easily be missed by humans.
  • Secure. Modern solutions, of course, must be secure, ensuring governance across the entire information supply chain. The systems should not only protect from outside infiltration but should control internal access. Users only should be allowed to access and use the information that is appropriate to their roles.
  • Collaborative. A modern data architecture should be collaborative, supporting sharing of information across organization boundaries, departments, or even outside the enterprise, ensuring everyone is working from the same data.

With these characteristics in place, users are free to pursue the analysis their business requires, says Vyas:

A modern data architecture is focused on enabling any user to take their analysis in any direction.. It allows for multiple tools, with multiple use cases, to easily and securely access any data object. It supports users through guided self-service to a single, semantic definition of all the enterprise assets, plus provides a version of the truth that is simple to access and trusted company-wide.

Case studies of results

Infor has several case studies of modern data architectures working for customers and transforming outcomes. Here are some findings and specific examples.

Speed pays off. Using implementation accelerators and a business-centric approach based on proven applications, organizations are achieving 66% higher implementation success rates, driving adoption out to the furthest parts of their enterprises. Automated data refinement technology is also enabling customers to build out analytic models quickly. Customers are reporting as much as a 70% reduction in the amount of time it takes to ingest and model data.

The modern data architecture is a pre-integrated, end-to-end solution, delivered in the cloud, with nothing to install. This yields deployment times as much as 40% lower than legacy approaches. But even more importantly, 61% of customers use the Infor data and analytics platform as their only solution, eliminating dozens of legacy tools and technologies, significantly simplifying their technology footprint.

Digital supply chain at Citrix

Citrix, a company that provides a digital workspace platform, was able to completely architect a digital supply chain using Infor’s modern data architecture. This approach integrated data from more than 400 internal and external data sources, on-premise and cloud, into a single end-to-end view of their supply chain. Citrix now has global and near real-time visibility connecting, aggregating, and transforming data every seven minutes — managed by a single IT person. With a digital supply chain, Citrix can leapfrog competitors and achieve an industry-leading 5X increase in inventory turns and 99% on-time delivery. Citrix also was able to achieve this is 73 days.

Consolidated purchasing at Schneider Electric

Schneider Electric provides digital solutions for homes and businesses. The company had more than 200 legacy, on-premise ERP systems. The company needed to consolidate the purchasing activities across 200 systems to achieve a holistic view of spending. Leveraging Infor’s modern data architecture, it was able to bring the systems together in a 360-degree view of purchasing. Transformational benefits were achieved in just more than 90 days. Schneider Electric is now able to save $500 million annually on non-production purchases that provide it the necessary capital to weather constantly changing conditions in its extremely competitive industry.

Miller Industries moves from spreadsheets to dashboards

Miller Industries is the world’s largest manufacturer of towing and recovery equipment. The company struggled with complex revenue pipelines and multiple data silos across multiple tools, including four legacy on-premise ERP systems. Miller Industries had more than 20 years of data that needed to be maintained across multiple databases and disparate data formats. The lack of a single integrated view of the current and historical manufacturing business made it difficult to forecast for future demand. Hundreds of manual spreadsheets had made automation and agility seem impossible.

The organization adopted Infor’s modern data architecture. Now, it has been able to replace hundreds of manual spreadsheets with information pulled into a centralized repository consumable through a few dashboards. This was delivered in just 90 days. The integrated solution has enabled Miller Industries to:

  • Dramatically simplify its analytics and reporting stack across every department
  • Significantly reduce its historical systems licensing and infrastructure overhead
  • Improve forecasting and pattern analysis through advanced analytics

Now the entire company, from the CEO to the front-line workers, can answer 80% of its business questions through personalized mobile dashboards. This has generated significant value for Miller, including reducing past-due account receivable balances by 40%.

These savings and the elimination of manual, broken processes enabled Miller Industries to reallocate resources to move its entire ERP footprint to the cloud. This was a strategic initiative that had long been delayed due to limited time and resources. With the new modern data architecture, Miller Industries also can tackle more initiatives and innovative ideas than ever before. 

Final takeaways

Enterprises today need to unify complex data across their organizations and bring it into a single integrated view of the business. Reporting needs to be fast and easy so the organization can manage the ever-changing needs of their industry and customers. Agility is essential today, and resilience is just as important. Only modern data architectures can offer companies the insights they need to adapt and stay relevant.

A grey colored placeholder image