Australian industry won't hit net zero if organisations hoard their data

For most of us, checking our phones each morning is routine. Will we need sunscreen or an umbrella? The apps giving us that answer rely on a system where 193 countries standardise and share measurements from ocean buoys, balloon radiosondes and orbital satellites, all feeding into the World Weather Watch. No single country could build an accurate forecast alone. Shared, standardised data turns fragmented local readings into reliable global predictions.

Industrial operations face an almost identical challenge. Individual mines, refineries and manufacturing plants generate vast quantities of data from equipment sensors, process controls and asset trackers, but that data typically stays onsite. An energy anomaly at one site could inform optimisation at another, but nobody can see it.

This matters now more than ever for Australian industry. The reformed Safeguard Mechanism requires roughly 215 facilities emitting more than 100,000 tonnes of CO₂-equivalent per year to reduce their baselines by 4.9% annually through to 2030. In the first year of the reformed scheme, covered emissions fell by 2.7 million tonnes. But the government’s own emissions projections show the path from Australia’s 2030 target of 43% below 2005 levels to its 2035 target of 62–70% will demand faster, deeper cuts across industry.

Some of Australia’s largest operators are already demonstrating what connected data can deliver. BHP, which reported a 32% reduction in operational greenhouse gas emissions by 2024 against its 2020 baseline, connects sensor data from drilling equipment, conveyor systems and processing plants into models that predict equipment failures, reduce unplanned downtime and cut fuel use. BHP’s operational emissions dropped to 9.2 MtCO₂-e, achieved through onsite abatement rather than carbon credit purchases.

Australia’s electricity grid faces the same coordination problem at a national scale. AEMO’s draft 2026 Integrated System Plan projects that total generation and storage capacity in the National Electricity Market needs to triple from 92 GW today to 297 GW by 2050. Managing that system, where rooftop solar on 1400 homes a day joins grid-scale wind farms, utility batteries and aging coal plants across five states, is fundamentally a data-sharing problem. AEMO’s 2025 Transition Plan for System Security identifies eight transition points linked to coal plant retirements where real-time coordination between generators, storage operators and network businesses will determine whether the lights stay on. The technology to connect those assets exists. The barrier is getting separate organisations to share operational data quickly enough for the system to respond.

These examples are encouraging but they remain exceptions. The World Economic Forum estimates that existing digital technologies could enable industries to cut emissions by up to 20%. Schneider Electric, whose industrial software customers span mining, energy and manufacturing operations in over 100 countries, reports that digitally connected operations across its customer base have saved more than 734 million tonnes of CO₂ since 2018.

Most of Australia’s safeguard-covered facilities are nowhere near that potential. The Clean Energy Regulator’s data from the scheme’s second year shows that 140 facilities exceeded their baselines by a combined 13.4 million tonnes, and the majority of compliance so far has relied on purchasing carbon credits rather than onsite abatement. When two-thirds of emission reductions under the scheme come from offsets rather than operational changes, the data infrastructure gap is part of the explanation.

Closing that gap requires more than buying software. Three things need to line up: reliable, standardised data flows between organisations, clear rules for how operational decisions get made from that data, and governance frameworks that give operators, vendors and grid managers enough trust to act on shared information. Many operators still treat process data as proprietary. Interoperability standards for industrial control systems remain fragmented. And the skills to interpret connected data at scale are in short supply across Australian industry.

Weather forecasting became reliable only after decades of institutional cooperation, common standards, agreed data formats and trusted exchange protocols. Industrial data sharing in Australia is at a much earlier stage. But the Safeguard Mechanism is tightening every year, and the facilities that cannot see their own emissions in real time will not find the reductions the scheme demands. BHP’s results show what connected operational data can deliver at site level. AEMO’s transition planning shows how urgently the same principle needs to scale across the grid. The question for the rest of Australian industry is whether it starts building that infrastructure now, or scrambles to buy offsets when baselines tighten further in 2027.

Image credit: iStock.com/metamorworks