Utilities keep warning that the power grid is under strain, especially during brutal summer heat waves and winter cold snaps. ecobee wants to position itself as part of the solution, and a new independent report gives the company some fresh numbers to point to.
According to an evaluation by Demand Side Analytics, ecobee’s Grid Resiliency service delivered about 108 megawatts of capacity during peak demand events in the summer of 2025. That figure came from real-world grid stress events across parts of the United States, and ecobee argues it shows how smart thermostats can quietly act as a kind of virtual power plant when demand spikes.
Grid Resiliency is now baked into ecobee’s eco+ platform, which already handles things like energy savings and schedule optimization for enrolled homes. The idea is fairly simple. When the grid is under pressure, ecobee can make small, temporary temperature adjustments across a large number of participating homes. Individually, the changes are minor. At scale, ecobee says they add up to something utilities can actually rely on.
The Demand Side Analytics report looked at 11 Grid Resiliency events across CAISO, ERCOT, and SPP territories, covering more than 143,000 ecobee devices. During those events, the average reduction worked out to roughly 0.65 kilowatts per-device, with peak reductions exceeding 1 kilowatt in some cases. Participation rates averaged 73 percent across all event hours, which is notable given that homeowners can opt-out at any time.
That opt-out point is one ecobee continues to emphasize. Users are not locked-in during grid events, and comfort is supposed to be preserved even when adjustments are made. Still, the data suggests that most participants stayed engaged, including during longer, multi-hour events. From ecobee’s perspective, that consistency is what makes the program interesting to utilities that are already familiar with traditional demand-response programs.
Demand Side Analytics reached a similar conclusion. The firm found that per-device load impact and participation rates were broadly comparable to what utilities see from more conventional demand-response efforts. The difference here is scale and friction. Instead of custom programs with complex enrollment, Grid Resiliency is designed to ride on top of devices that are already installed in homes.
That ease-of-enrollment matters if the goal is capacity measured in gigawatts rather than megawatts. Based on current adoption and modeled expansion, the report estimates that Grid Resiliency could unlock about 2,585 megawatts of incremental capacity in the United States and another 215 megawatts in Canada. Combined, that works out to roughly 2.8 gigawatts of potential capacity, at least on-paper.
Those numbers deserve a bit of healthy skepticism. They assume widespread participation, consistent performance, and continued cooperation from homeowners during stressful grid conditions. They also depend on utilities being willing to treat aggregated residential load reductions as a serious planning resource rather than a nice-to-have experiment. Still, the fact that 108 megawatts showed up during real-world summer events gives the idea more weight than a purely theoretical model.
There is also a broader industry context here. Virtual power plants have become a popular talking point as utilities look for faster and cheaper ways to address reliability concerns. Instead of building new generation or transmission infrastructure, companies are trying to squeeze more flexibility out of what already exists in homes and businesses. Smart thermostats are an obvious candidate because they control one of the largest loads in many households.
ecobee’s approach leans heavily on automation. Homeowners do not need to respond to alerts or manually adjust settings when the grid is stressed. The system works quietly in the background, which likely helps explain the relatively high participation rates. At the same time, the ability to opt-out acts as a pressure-release valve for anyone who feels uncomfortable during an event.
From a grid operator’s point-of-view, the appeal is speed. When extreme weather hits, there is not much time to negotiate or fine-tune programs. A platform that can quickly dispatch small reductions across hundreds-of-thousands of homes could be useful, especially as heat waves become more frequent and electricity demand continues to climb.
Whether Grid Resiliency ultimately becomes a core part of grid planning is still an open question. For now, ecobee has some real-world data to support its claims, and utilities have another example of how consumer devices might help carry part of the load during peak demand. The summer 2025 results suggest that, under the right conditions, a lot of tiny adjustments can add up to something that actually matters.
