Throughout the 21st century, enterprise resource planning (ERP) systems have been the foundation of construction’s accounting and operations universe. But as the industry evolves, companies are increasingly constrained by ERP-centric structures that limit adaptability.
Enter data centricity: a shift that makes data — not applications — the foundation of a technology ecosystem. Instead of forcing systems to depend on ERPs, data flows freely, allowing for seamless integration, flexibility, and innovation.
Rather than discarding ERP systems, consider positioning them as one component of a broader strategy. The result can unlock new possibilities through greater agility, more accessibility, and lower operational costs.
Data Centricity
In a data-centric model, the emphasis is on liberating data from silos and making it accessible across the organization. At its core, data centricity is about empowering organizations to derive maximum value from their data assets.
With this, you are no longer attached to any particular system or application to hold information. Instead, data serves as the foundation for decision-making and system architecture, providing agility to adapt and scale.
Instead of building rigid systems around ERP platforms, data centricity advocates for connecting ERP systems to a broader data hemisphere. This model removes the constraints of being confined within rigid structures and consolidates your data into a unified, accessible repository, enabling seamless integration and strategic use.
But isn’t this just a data warehouse, data mart, or another form of big data architecture? Not exactly. A data warehouse is just one tool within this ideology; overall, the data-centric approach transcends traditional methodologies.
Embracing the Shift
A true data-centric approach begins by identifying the information required, even before selecting the system to manage or process it. The aim is to establish a singular, authoritative
source of truth within a centralized data warehouse.
Unlike legacy ERP-driven strategies, applications become secondary to the primacy of data; this notion may alarm you because it requires a paradigm shift. This approach demands a unified data environment for reporting, analytics, and decision-making.
By making these choices, you eliminate the limitations imposed by ERP-driven architectures and disconnected applications, as many organizations often rely on a patchwork of tools to make sense of fragmented data, tracking critical details outside of core platforms.
To embrace data centricity, you must shift the focus from applications to the true asset — your data. Data is the only business asset that retains value indefinitely; it does not degrade over time. It serves as the institutional memory of your organization, offering the foresight to anticipate challenges and opportunities.
With the right data strategy, you can identify risks weeks or even years in advance — whether it’s for a project, subcontract, client, or company-wide issue.
This structured approach paves the way for AI, predictive analytics, and automation — not as futuristic concepts, but as practical, actionable tools. The data itself, rather than the systems that store it, must be safeguarded, structured, and leveraged.
Leveraging ERP Systems Within a Data-Centric Framework
Embracing data centricity doesn’t render ERP systems obsolete; rather, it repositions them within the technology landscape. As a part of the paradigm shift, think of them as a planet, not the sun (Exhibit 1 on page 33). With this focus, your ERP will still remain a crucial player in your technology framework.
In the data-centric universe, the ERP functions as a tool, delivering top-notch capabilities through its streamlined and efficient processes. It serves as the cornerstone for ensuring that controls, workflows, and essential tasks are managed with precision.
To truly maximize your investment, deploy the ERP in a way that allows it to maintain control over the things it was designed to do.
Evaluating Your ERP
First, look at it through the eyes of your employees and partners. By approaching your systems with this mindset, you can decide which system should be the interface, or plainly, the place in which the data is entered.
The second important factor in leveraging your ERP is to determine what processes, workflows, and outputs are required.
The third factor is to start a conversation around the analytics platform you will use for the entire organization and begin to talk to your company about freeing your data. This requires that you open data to anyone, anywhere in your company.
The caveat is that some data must remain confidential (e.g., protected information, sensitive client data, bank accounts).
Allowing access to data for everyone is more about eliminating your ERP from being the primary driver for data reporting and setting up future success; all data reporting and analysis will begin in your analytics platform. For example, someone in operations should not need to call someone in accounting to get a cost report.
Although more can be done to leverage your ERP, these are the major first steps to get the best use of your ERP and begin to turn it into a precise and invaluable tool that isn’t synonymous with your data.
The Power of Integration
Integration may not be a new concept, but running integrations as a subscribed service may be. The standard for integrations has previously been point-to-point integrations from one system to another.
For example, if you bought a new safety platform and want to get job data into it, you may begin by writing an integration from your ERP to send new jobs to your safety system. You could also write an integration to bring in employees, which may come from your ERP or from a separate human capital management (HCM) system.
Next, you would bring over cost information from jobs, such as volume, to your new safety system to evaluate incidents to volume or other variables.
These integrations are powerful in themselves; they ensure that the same data is being used across your systems in areas such as employees, vendor identifications, job numbers, and more. On the other hand, they can often be difficult to get working and can add up to hundreds of data integrations.
Each of these integrations are running on events, or timers, and often across your systems at the same time. These are expensive to not only get started, but also to maintain and change as new systems or data are added.
In today’s cloud-centric landscape, tools that weren’t on the radar previously now enable integration with significantly enhanced robustness. While still relying on application programming interfaces (APIs), the real game-changer is the approach to initial integration for the data warehouse.
A Single Point of Integration
In this paradigm shift, transitioning to a single point of integration leads to a centralized source. Every cloud data hosting system offers a platform for establishing this central hub and automating information management across your network.
When setting up and establishing your single integration point, key fields and objects are defined once. From there, you can subscribe a system to the “event” grid.
When a new job is created, every system subscribed to the job change event will automatically receive the pertinent information pushed to it through the event grid. This streamlined approach revolutionizes how data is managed and distributed across the network (Exhibit 2).
Example
Consider the example of managing employee data. In this scenario, the event grid is looking for changes, such as new hires or terminations, as they are pushed from the main application to the warehouse.
When such events occur in the HCM system, the information is automatically sent to the data warehouse. The event grid will also auto-push the latest information to any system waiting for it, ensuring that any updates in employee data are seamlessly reflected in any system using the information.
With this streamlined integration process, whether it’s adding a new user, updating their details, or adjusting reporting structures, the process remains efficient. Changes also only need to be made once, ensuring seamless updates throughout the systems.