How NxGN Deployed ESG Data Management for ARM
African Rainbow Minerals operates a diversified portfolio of mining assets across South Africa. These include platinum, iron ore, manganese, nickel, and chrome operations spread across four provinces. Each operation had its own approach to capturing sustainability data. Some used IsoMetrix for safety, while others relied on spreadsheets. Environmental data sat in separate monitoring systems. As a result, there was no consistent way to collect, validate, and consolidate ESG metrics at the group level.
The Reporting Challenge
Compiling group-level reports required manual gathering from disparate sources across eight sites. The team had to reconcile inconsistencies and validate numbers collected through different processes. ARM had reporting obligations across 80 ESG disciplines. Each discipline required data from multiple operations, and those operations had different systems and varying levels of digital maturity.
The Trigger
ARM’s sustainability leadership saw what NxGN had built for Anglo American. The Objectives & Targets platform (now productised as NxGN Capstone) manages 3,500 data inputs and 1,650 calculations across 160+ operations. The same ESG data management problem ARM faced manually had already been solved at a much larger scale. Rather than building from scratch, ARM adopted the proven platform. NxGN then configured it for their specific operations, disciplines, and reporting requirements.
Why This Was Hard to Solve
The diversity of ARM’s operations made a one-size-fits-all approach impossible. For example, an underground platinum mine in Limpopo has fundamentally different data infrastructure from an iron ore operation in the Northern Cape. Similarly, a smelter works in KwaZulu-Natal operates differently from both. Some sites had mature IsoMetrix deployments, while others had minimal digital infrastructure. With 80 disciplines, the team needed 80 different data collection structures unified into a single model. Any solution had to integrate with existing systems where available, support manual capture where they don’t exist, and still produce a single consolidated, validated dataset.
Why NxGN
NxGN had already solved this exact ESG data management problem at Anglo American at 20 times the scale. The Capstone platform was proven, and the IsoMetrix integration pipelines were built. Moreover, the distributed operations architecture could handle sites at different levels of digital maturity. ARM was the first commercial customer for a platform that had been battle-tested across 160+ operations.
What Did NxGN Deploy?
Assessment: Understanding What Existed at Every Site
Before deploying anything, NxGN conducted a Digital Transformation Maturity Assessment across all eight operations. The team used a methodology adapted from the Industry 4.0 framework and ran a 2-hour workshop at each site. Workshop participants included system owners and discipline owners across all ESG domains.
The assessment covered eight sites across four geographical clusters. The Northern Cape cluster included Khumani, Beeshoek, and Black Rock. Limpopo covered Bokoni Platinum and Modikwa. Mpumalanga encompassed Two Rivers, Machadodorp Works, and Nkomati. Finally, Cato Ridge Works represented KwaZulu-Natal. For each site, the team identified the system of record, its version, the primary integration mechanism, and any security or environment constraints.
The deliverable was a Business Requirements Specification. It included the as-is landscape, a gap analysis against the Capstone reference architecture, and an Industry 4.0 maturity score for each operation.
Deployment: Site by Site
NxGN deployed Capstone on ARM’s Azure environment and worked through each operation individually. For each site, the process followed the same pattern. The team ran the workshop, gathered the site’s data, validated metrics against ARM’s 80 ESG disciplines, configured the model, trained users, and deployed. NxGN’s actual effort was approximately two weeks per site. However, the elapsed time was closer to two months per operation because of site availability, data readiness, and operational schedules.
Data Integration and Validation
The 1,153 data inputs span 80 disciplines across all eight operations. IsoMetrix integration pipelines, drawn directly from the Anglo American implementation, automatically feed safety and health data into Capstone. Operations without source systems use Capstone’s native data capture interface. This interface includes Excel-like functionality: copy-paste, import/export, data flagging, change history, and a validation workflow.
All data goes through monthly validation and locking before export. This creates an auditable chain from source through to the data warehouse. The consolidated, validated dataset then feeds ARM’s reporting data warehouse for downstream ESG calculations and reporting. Capstone now includes full calculation and reporting capabilities. At Anglo American, it handles 1,650 automated calculations. This means the platform can manage the entire pipeline from data collection through to final reporting within a single stack.
Technical detail: Distributed operations architecture
The core platform comprises a web application, application API, identity server, transactional database, and reporting data warehouse. These deploy as Azure services with no dedicated servers. For operations with on-premises systems that cannot integrate with the cloud, Capstone’s distributed operations architecture deploys satellite instances locally. These instances collect data from site-level systems and synchronise to the central Azure instance on a configurable schedule. IsoMetrix integration pipelines use Azure Data Factory to transform IsoMetrix’s proprietary key-value pair database structure into a relational format. The pipelines run nightly without impacting source system performance. The validated dataset then exports to ARM’s data warehouse for downstream consumption.
What Were the Results?
| Metric | Before | After | Improvement |
|---|---|---|---|
| ESG data collection | Fragmented across 8 operations, different systems per site | 1,153 structured data inputs in a single platform | Single source of truth for ESG data |
| Disciplines covered | Inconsistent tracking, manual compilation | 80 disciplines systematically configured across all operations | Comprehensive ESG coverage |
| Data validation | Ad-hoc manual reconciliation on final spreadsheet | Validation at source with monthly locking and evidence | Audit-ready at every level |
| Integration approach | No consistent integration layer | IsoMetrix pipelines + manual capture + distributed sync | Every operation connected regardless of infrastructure maturity |
| Deployment per site | Months of requirements gathering with uncertain outcomes | ~2 weeks NxGN effort per site (~2 months elapsed including operational schedules) | Repeatable deployment methodology across 8 operations |
| Downstream reporting | Manual compilation before calculations could begin | Validated dataset exported to data warehouse for downstream calculation | Audit-ready data feeding ESG calculations and reporting |
A Single Source of Truth
The platform became ARM’s foundation for ESG data management. Operations across all four provinces now capture and validate sustainability data through a consistent process. This ensures that the numbers underpinning ARM’s ESG reporting are defensible and auditable from source.
Meeting Each Operation Where It Was
The distributed architecture proved essential. Rather than waiting for every site to reach the same level of digital maturity, the platform met each operation where it stood. It integrated with IsoMetrix where available, accepted manual capture where needed, and synchronised everything to the central instance. Consequently, ARM could start reporting from the platform immediately. The option to automate additional data feeds remains open as site-level systems mature.
Proven, Repeatable Methodology
The deployment also demonstrated the value of a proven, repeatable methodology. Because NxGN had already built and refined the platform at Anglo American, each ARM site followed a consistent process: workshop, data gathering, metric validation, model configuration, training, and deployment. NxGN’s actual effort was approximately two weeks per site. The proven platform meant NxGN could work through all eight operations with a predictable process, even when individual sites moved at different speeds.
Frequently Asked Questions
Can NxGN Capstone work across operations with different levels of digital maturity? Yes. This is a core design principle. ARM’s eight operations ranged from sites with mature IsoMetrix deployments to operations with minimal digital infrastructure. Capstone’s distributed operations architecture deploys satellite instances at sites that cannot connect directly to the cloud. These instances synchronise data to the central platform on a configurable schedule. Operations that can only provide manual input use Capstone’s native data capture interface. As a result, every site connects regardless of its infrastructure maturity.
Can Capstone handle ESG calculations and reporting as well as data collection? Yes. At Anglo American, Capstone handles both data collection and calculation. The platform processes 3,500 inputs and 1,650 automated calculations, producing outputs for the CEO scorecard, annual sustainability report, and quarterly performance reviews. The configurable formula language supports complex interdependent KPI calculations across any number of disciplines and operations. Organisations can use Capstone for the full pipeline from collection through to final reporting, or integrate it with existing downstream tools.
How quickly can Capstone deploy across multiple mining operations? Each ARM site took approximately two weeks of NxGN effort. This covered workshops, data gathering, metric validation, model configuration, training, and deployment. The elapsed time per site was closer to two months due to site availability and operational schedules. NxGN worked through all 8 operations in a staggered rollout. The efficiency was possible because Capstone had already been proven at Anglo American across 160+ operations. The platform, integration patterns, and deployment methodology were ready, so each site followed a repeatable process.