For decades, utilities have measured grid investment largely by what could be seen and built: transmission lines, substations, transformers and power plants. Those assets remain essential, and utilities will continue to need more of them.
But as demand grows and distributed energy resources (DERs) reshape local grid conditions, physical infrastructure alone will not be enough to meet the pace of change.
Utility planning is entering a new era. According to the International Energy Agency, US electricity demand growth in 2025 reached its second-highest level this century, driven by data centers, industrial activity and evolving environmental conditions. But for utilities, rising demand is only part of the challenge. New load is appearing unevenly across the distribution grid, often faster than traditional planning processes can identify and address.
Electric vehicles (EVs), rooftop solar, battery storage and localised data centre growth are changing when and where electricity is consumed. An EV charging cluster can create a sharp evening peak in one neighbourhood, while another exports solar power during the day and draws heavily from the grid after sunset. These patterns vary from feeder to feeder and, in some cases, from one transformer bank to the next.
As a result, utilities need a more precise understanding of how local grid conditions are changing and which investments will deliver the greatest value.
Static forecasts were built for a different grid
Traditional distribution planning relied on relatively stable assumptions about demand. Historical consumption, population growth, weather patterns and economic development provided a reliable foundation for forecasting future needs. That approach worked when electricity largely moved in one direction, from centralised generation through transmission and distribution networks to customers.
Now, variability exists at a much more localised level than traditional planning models were designed to accommodate.
Behind-the-meter resources have changed the shape of local load curves. Midday solar generation may reduce net load without indicating that customers are using less energy. Evening peaks may be driven by EV charging, battery behaviour or building electrification rather than traditional residential consumption. A feeder that appears healthy in monthly averages may still experience recurring stress during specific hours of the day.
If utilities cannot see and account for these shifts, they face difficult tradeoffs. They may invest in infrastructure before upgrades are truly needed, increasing costs for customers, or they may delay action where localised assets are already approaching operational limits, creating reliability risks. The challenge is no longer simply forecasting how much demand will grow. It is understanding where growth is occurring, when it creates constraints and which response is most appropriate.
IoT infrastructure is expanding the field of view
The modern distribution grid is increasingly becoming a large-scale IoT environment. Smart endpoints, line sensors, transformer monitors, automated switches, DER controllers and other connected devices are generating data from parts of the grid that were once difficult to observe.
These devices provide insights into voltage, loading, power quality, outage conditions and customer-side activity. Together, they help utilities distinguish between systemwide demand growth and localised constraints. That distinction matters because two areas with similar load growth may require very different responses. One may need a transformer upgrade, while another may benefit from managed charging programs or other forms of load flexibility.
Supporting this visibility requires communications infrastructure capable of connecting a growing mix of devices with different technical requirements. As utilities deploy more connected assets across the grid, flexible network architectures become increasingly important for moving data where it is needed and supporting new applications at the grid edge.
Connected infrastructure can help utilities translate device-level data into operational insight that supports planning, investment and asset management decisions.
Edge analytics turn device data into operational intelligence
More connected devices means more data. The challenge is identifying which signals matter and translating them into actionable information.
By processing data closer to where it is generated, utilities can identify emerging issues more quickly and reduce the volume of information that must be sent to centralised systems for analysis.
Grid-edge analytics can surface conditions that often remain hidden in traditional planning models, such as:
• A transformer that experiences repeated stress during a narrow evening window, even though daily averages appear normal.
• A cluster of solar installations creating voltage fluctuations under specific weather and load conditions.
• A group of EV chargers creating short-duration peaks that accelerate wear on local equipment.
These insights help utilities move beyond broad assumptions and understand how assets are actually performing. A utility that can identify precisely when and where a constraint occurs may be able to defer an upgrade by managing demand during a limited peak window. In other cases, the same data may confirm that infrastructure investment is the most practical solution.
Utilities can use these insights to prioritise upgrades, target investments and improve asset utilization based on measured operating conditions rather than broad system averages.
DER planning must move from interconnection to coordination
DERs such as rooftop solar, batteries, EV chargers and smart thermostats are becoming increasingly important components of the modern grid. Historically, utility planning focused on whether these resources could be interconnected safely. Now, the question is how they can be coordinated to support reliability and help address local constraints.
Grid-edge data provides visibility into where flexibility exists and where constraints are emerging. But visibility alone is not enough. Utilities also need systems that can help translate planning assumptions into operational action.
A distributed energy resource management system (DERMS) can help utilities understand available flexibility, evaluate local conditions and coordinate customer-side resources in support of grid needs.
For example, managed EV charging may reduce pressure on a constrained transformer during evening demand peaks. Elsewhere, battery dispatch may support peak reduction or help maintain service during an outage. In other cases, infrastructure upgrades may still be the right solution because flexibility alone cannot address the constraint.
Infrastructure investment remains essential, but utilities need greater confidence that these investments are targeted where they create the greatest value.
Every megawatt of capacity unlocked through better use of existing infrastructure is capacity customers do not have to pay to build unnecessarily. And when new infrastructure is needed, utilities can move forward with greater confidence that they are addressing the right problem at the right time.
Planning from the grid edge inward
Utilities will continue to invest in substations, transformers and new grid capacity. But as electrification, DER adoption and demand growth accelerate, they will also need to get more from the infrastructure already in place.
Connected devices, communications networks and grid edge analytics provide a way to identify constraints earlier, understand how assets are performing and evaluate alternatives before committing capital. Combined with DER coordination, these technologies help utilities make more informed decisions about when to invest, where to invest and whether flexibility can solve a problem before new infrastructure is required.
In a future shaped by electrification, distributed energy resources and increasingly complex demand patterns, software and operational intelligence will play a larger role in helping utilities deliver reliable and affordable service. The industry has spent more than a century building a bigger grid. The next chapter is about building a smarter one.
