The article is kinda weak on details, but you can download the summary of the report they are referencing. It’s probably important to observe that this is the work of a lobbying group, so some skepticism is merited, but they claim to source from actual utility data. So maybe it’s reliable.

In any case, here are some key observations from the report summary if you’re curious:

• The US solar industry installed 43.2 gigawatts direct current (GWdc) of capacity in 2025, a 14% decrease from 2024. The utility-scale sector shrank nearly 40% quarter-over-quarter in the fourth quarter. Revised tax credit timelines and safe harbor dynamics reduced the imperative to interconnect by year-end. They also increased the urgency to begin construction on new projects.
• Solar accounted for 54% of all new electricity-generating capacity added to the US grid in 2025. Combined, solar and storage made up 79% of new capacity in this timeframe. Throughout all of Wood Mackenzie’s US power sector outlooks, solar capacity constitutes roughly half of new capacity added each year through 2060.
• 2025 was a monumental year for the US solar manufacturing industry. New cell capacity continued to expand, and wafer capacity came online for the first time since 2016. Module manufacturing grew more than 50% in 2025, with 65.5 GW of capacity online, up from 42.5 GW at the end of 2024. However, the actual production of these facilities remains considerably below domestic demand.
• In 2025, the residential segment installed 4,647 MWdc of solar capacity, declining 2% compared to 2024. Although module shortages and delivery delays were a concern in the fourth quarter, many installers ultimately received the equipment they needed. However, 2025 volumes weren’t higher leading up to the Section 25D expiration because there simply wasn’t enough time to meaningfully ramp up sales and installations after the passage of the OBBBA.
• The commercial solar segment grew 6% in 2025, adding 2,345 MWdc of new capacity. The pipeline of NEM 2.0 installations in California continued to come online. We expect it to decrease in 2026, but even in the fourth quarter, more than 70% of installations were still NEM 2.0 projects rather than NBT (net billing tariff) projects.
• The community solar segment installed 1,435 MWdc in 2025, down 25% from 2024. Maine and New York saw slowdowns, and no new community solar programs generated growth.
• The utility-scale segment installed 34.7 GWdc in 2025, a 16% decline compared to 2024. Nearly the same amount of capacity came online through the first three quarters of the year as did in 2024. But substantially fewer projects that were originally slated to come online in Q4 were energized. Due to the changes in tax credit deadlines, developers delayed commercial operation dates and focused on safe harboring their pipeline

The point about NEM3 in California is already happening. NBT installations and applications are sharply down from before the MEM3 cutover. I expected that to be a bigger factor in the data than it was. But residential being down only 2% nationwide likely means it’s up in most other states. That’s good news.