管理層發言
Good afternoon, everyone, and welcome to Enphase Energy's Second Quarter 2026 Financial Results Conference Call. All participants will be in a listen-only mode. Should you need assistance, after today's presentation, there will be an opportunity to ask questions. To ask a question, you may press star then 1 on your touch-tone phones. To withdraw your questions, you may press star and 2. Please also note this event is being recorded. At this time, I would like to turn the floor over to Zachary Freedman with Enphase. Please go ahead.
Good afternoon, and thank you for joining us on today's conference call to discuss Enphase Energy's second quarter 2026 results. On today's call are Badrinarayanan Kothandaraman, our President and Chief Executive Officer; Mandy Yang, our Chief Financial Officer; and Raghuveer R. Belur, our Chief Products Officer. After the market closed today, Enphase issued a press release announcing the results for its second quarter ended 06/30/2026. During this conference call, Enphase management will make forward-looking statements, including, but not limited to, statements related to our expected future financial performance, market trends, the capabilities of our technology and products, and the benefits to homeowners and installers; our operations, including manufacturing, customer service, and supply and demand; anticipated growth in existing and new markets, including the TPO market; the timing of new product introductions and enhancements to existing products; and regulatory, tax, tariff, and supply chain matters. These forward-looking statements involve significant risks and uncertainties, and our actual results and the timing of events could differ materially from these expectations. For a more complete discussion of the risks and uncertainties, please see our most recent Form 10-K and 10-Qs filed with the SEC. We caution you not to place undue reliance on forward-looking statements and undertake no duty or obligation to update any forward-looking statements as a result of new information, future events, or changes in expectations. Also, note that financial measures used on this call are expressed on a non-GAAP basis unless otherwise noted, and have been adjusted to exclude certain charges. We have provided a reconciliation of these non-GAAP financial measures to GAAP financial measures in our earnings release furnished with the SEC on Form 8-K, which can also be found in the Investor Relations section of our website.
Now I would like to introduce Badrinarayanan Kothandaraman, our President and Chief Executive Officer. Badrinarayanan?
Good afternoon, and thank you for joining us today to discuss our second quarter 2026 financial results. We reported quarterly revenue of $291.9 million, shipped 1.59 million microinverters and 113.8 megawatt hours of batteries, and generated free cash flow of $25.9 million. Our Q2 revenue included $84.3 million of safe harbor revenue. We exited the quarter with channel inventory normal for batteries, and slightly elevated for microinverters. On a GAAP basis, we delivered gross margin of 60%, operating expense of 42.3%, and operating income of 17.7%, all as a percentage of revenue. On a non-GAAP basis, we delivered gross margin of 46.8%, operating expense of 27.3%, and operating income of 19.4%, all as a percentage of revenue. Mandy will cover the financials later in the call. Our global customer service NPS was 80% in the second quarter as compared to 82% in the first quarter. Our average call wait time remained approximately 2 minutes. We also made our AI assistant available to 1.5 million homeowners worldwide. This gives our customers faster, personalized system-specific support and makes their energy systems easier to understand and manage, ultimately reducing the number of calls to operations. In the second quarter, we shipped 1.58 million U.S.-made microinverters and battery inverters from our Texas and South Carolina manufacturing facilities and booked the associated 45X production tax credits. We also shipped 43 megawatt hours of IQ Batteries from our Texas facility in the second quarter. We offer IQ Batteries that meet domestic content requirements, helping lease and PPA customers qualify for ITC bonuses. Let's now cover revenue and regional performance. Our global Q2 revenue increased 3% compared to Q1. Our global sell-through was approximately flat as compared to Q1, as growth in Europe offset the softness in the U.S. Our revenue mix was 78% from the U.S. and 22% from international markets. In the U.S., revenue declined 3% sequentially. Safe Harbor revenue increased to $84.3 million in Q2 as compared to $34.5 million in Q1. Excluding safe harbor revenue, U.S. revenue declined primarily due to us under-shipping into the channel. Our U.S. sell-through in Q2 decreased 7% as compared to Q1. Excluding one-time orders in Q1 that did not recur in Q2, the sell-through was approximately flat sequentially. Our Q2 2026 sell-through declined 34% as compared to one year ago, reflecting continued pressure from higher interest rates and transition following the expiration of the 25B tax credit. Third-party market reports suggest that the broader U.S. residential solar market has stabilized, with industry-wide permits in June increasing 4% from May and upstream sales activity rising 5%. Both remain about 30% below prior year levels. Higher electricity cost markets are performing better while several Sunbelt states remain under pressure. The stronger industry-wide signals are for storage and commercial solar. National residential battery attachment remains near 40% with materially higher levels in key markets. The U.S. commercial solar permit activity increased 36% year-on-year in June. Taken together, these third-party data points suggest that the next phase of U.S. market growth will be shaped by storage economics, commercial demand, financing availability, and utility rates. In Europe, our revenue increased 35% sequentially in the second quarter, while sell-through grew 30% with strong performance across both solar and batteries in multiple markets. The growth was supported by higher power prices as well as accelerating battery adoption. Europe is increasingly becoming a battery-led market as self-consumption, dynamic tariffs, and VPPs gain importance. The company that owns the battery relationship is well positioned to expand over time into the broader home energy system including solar, EV charging, and VPP. In the Netherlands, our battery activations increased approximately 102% from the first quarter as rising export penalties and the planned phase out of net metering at the end of 2026 strengthened self-consumption. In France, lower feed-in tariffs are similarly shifting the market towards self-consumption and driving greater interest in batteries, particularly with new solar installations. Battery activations in France increased approximately 34% sequentially. In Germany, the growth was broad-based with both microinverter and battery activations increasing approximately 35% and 30%, respectively. We are intensifying our focus on battery retrofits in both the Netherlands and France where we have a combined installed base of nearly 900,000 Enphase customers. Building on the success of our initial programs, we have increased the cadence of homeowner events and direct marketing campaigns. Our newly established inside sales team, supported by an improved lead management platform, is helping convert this demand into revenue. We also showcased our fifth generation battery at Intersolar Munich where customer feedback was positive, and we expect initial shipments before the end of this year. Let's now discuss our outlook for the third quarter. We expect revenue of $290 million to $320 million representing approximately 5% growth at the midpoint. Our Q3 revenue guidance includes $75 million of safe harbor revenue. We are currently over 70% booked to the midpoint of our guidance. We expect global sell-through in Q3 to increase 10% as compared to Q2. Distributors remain cautious amid broader macroeconomic uncertainty, including interest rates, and our guidance assumes modest under-shipment relative to sell-through. For batteries, we expect shipments between 130 to 150 megawatt hours as momentum continues to build in both the U.S. and Europe. As reciprocal tariffs have moderated somewhat, we reduced battery pricing in late May, and we expect to take further targeted pricing actions as necessary to improve system economics and support demand. Turning to safe harbor: we have executed year-to-date agreements with third-party owners totaling approximately $1.1 billion — $102 million under the 5% ITC safe harbor method and $878.6 million under the physical work test, or beginning of construction, method. These agreements provide two important benefits. They secure meaningful multiyear volume for our microinverter and accessory business, and they create a strong foundation for future battery attach opportunities as these systems are installed from 2028 through 2030. Moving to financing: PROPEL is entering a new phase of growth. To remind you, Propel is a TPO offering from Sole Source Solutions that combines Enphase equipment financing, loan financing provided by TriBeam Financial through the Concert Finance platform, and national distribution through Green-Tech Renewables. Purpose-built for the long tail of installers, Propel has expanded from four states to six states with recent launches in Pennsylvania and Colorado, and plans to reach a total of 12 states by the end of the third quarter. Installer participation has grown to above 290. Propel originations are running at approximately 200 per week with battery attachment at roughly 75%. We expect this will begin to grow again as installers in new states start to ramp up. Sole Source is targeting 500 originations per week by the end of the year and scaling by securing sufficient warehousing capacity and tax credit buyers. In today's higher interest rate environment, Propel offers homeowners and installers a compelling alternative to conventional solar loans and can help restore a meaningful portion of the cash and loan market affected by 25B expiry. Let's talk about products. Starting with IQ Batteries: we showcased our fifth generation IQ Battery G5 at Intersolar Munich in June where it received a strong response. Built from stackable, AC-coupled, 5 kilowatt-hour modules that can scale up to 30 kilowatt-hours in one stack, the G5 uses 100 ampere-hour prismatic cells and is designed to deliver 50% higher energy density than our fourth generation battery at roughly 40% lower cost per kilowatt-hour. When shipments begin in the fourth quarter of 2026, we believe the IQ Battery G5 will stand out as one of the few truly stackable AC-coupled battery platforms in the market. Its combination of lower cost, flexible sizing, strong performance, high quality, and high serviceability should make it highly competitive across the U.S., Europe, and Australia. We are also making good progress on our commercial battery, called IQ Vault, targeted for both 3-phase 208 and 480-volt markets. The first product, called IQV-80, is an 80 kilowatt-hour battery, 3-phase, 480 and 208 volts with 40 kilowatts of continuous power — basically a two-hour battery. Each outdoor cabinet uses five field-serviceable 16 kilowatt-hour LFP modules built with 314 ampere-hour prismatic cells and up to 25 cabinets can scale the system to 2 megawatt hours. The 480-volt 3-phase configuration is designed for larger commercial buildings, while the 208-volt 3-phase configuration will address small commercial and multifamily properties including applicable California projects driven by Title 24 requirements. The distributed architecture provides module-level fire suppression and is designed for self-consumption peak shaving, time-of-use, VPP, and backup. We have completed the functional system demonstration in the last quarter and we expect to open preorders soon, with initial shipments planned for Q1 2027. Turning to microinverters: we launched our GaN-based IQ9 residential microinverter across the U.S. and key European markets in June, followed by Australia and New Zealand earlier this month. Also gaining traction in the U.S. commercial market with several promising national opportunities advancing with large retail customers. During the second quarter, we began shipping the IQ9s 3-phase microinverter, our highest power microinverter to date, 480 watts based on gallium nitride, for 480-volt systems. This is designed to support solar panels up to 770 watts. With U.S. manufacturing, domestic content eligibility, and FIAT, we believe our commercial business is well positioned for continued growth. We recently opened preorders for our smart thermostat, a new control point for the Enphase energy system. By bringing HVAC into the system, Enphase can optimize one of the home's largest energy loads alongside solar and batteries to improve savings, preserve backup capacity, and support VPP. An integrated display on the device also gives homeowners a simple way to view their solar, battery, and home power live from inside the home. We expect shipments next month. Moving on to EV charging: we are making strong progress on the DC-based IQ bidirectional EV charger, which we showcased at Intersolar Munich. Built on our GaN power platform, this is designed to support both 400-volt DC and 800-volt DC EVs and deliver up to 11.5 kilowatts of bidirectional power. The ISO 15118-20 standard enables standardized communication between the vehicle and the charger, while our expertise in utility interconnection, grid code compliance, and distributed energy management supports V2H backup, V2G, and use cases like green charging. We are collaborating with three leading automotive OEMs in the U.S. and one in Europe, with additional engagements underway. Subject to the successful completion of applicable compliance testing, we expect to begin pilot shipments in the fourth quarter alongside vehicle launches from one U.S. OEM and one European OEM. Finally, let me provide more details on our IQ Solid State Transformer, or IQSST. The rapid build-out of AI infrastructure is reshaping data center power architecture as rack densities rise from approximately 15 kilowatts today towards 1 megawatt and beyond. Delivering power at that scale will require a fundamentally more efficient, responsive, and reliable way to move medium-voltage power directly to the computer. IQSST is designed to meet that need by converting 13 kV and 13.8 kV or 34.5 kV medium-voltage AC directly to 800 volt DC through a modular single-stage architecture. At the core of the platform is our IQSST power module, which utilizes our predictive control enabled by custom silicon and GaN, enabling high-frequency switching and innovation in medium-voltage transformer design. Built on more than 20 years' experience in distributed power electronics, we are targeting approximately 98.5% efficiency, five-nines reliability, and sub-millisecond response time. That response time is a key differentiator. AI workloads can create rapid swings in power demand and IQSST is designed to respond in real time to help stabilize the load as seen by the data center power system. This could allow most of the energy storage to be centralized in a BESS located in the data center's black space rather than placed beside every compute rack, freeing up valuable white space. This configuration would utilize a second SST for the BESS, effectively doubling our data center opportunity. For customers that still require storage near the rack, the same platform can also support a DC configuration that charges and discharges a local high C-rate battery to help manage dynamic AI loads. U.S. manufacturing and a FIAT-compliant supply chain add another important layer of differentiation. They give customers greater confidence in supply continuity, product traceability, and the ability to deploy at scale without relying on restricted foreign entities. For hyperscalers and data center operators making long-lived infrastructure decisions, we believe domestic manufacturing, resilient sourcing, and a clear path to high-volume production will be as important as product performance. Our new and existing customer engagements continue to deepen. We have advanced a few of these opportunities to the RFI/RFP stages representing potential demand totaling multiple gigawatts. These engagements are directly shaping our roadmap across power level input voltage, footprint, cooling, battery connectivity, and serviceability. Importantly, we have been able to address evolving customer requirements without changing the fundamental IQSST power module, underscoring the flexibility of our platform. We have also made substantial technical progress over the last three months towards a fully working system later this year. Our team has now grown to about 120 people. We have begun testing the second revision of the IQSST power module and results give us confidence that the next revision can become our production candidate. We have completed the build-out of our medium-voltage lab and validated the medium-voltage transformer design. We are now optimizing it for manufacturability and cost. This work has already generated meaningful IP, particularly around the transformer. At the system level, our power modules are connected in series on the medium-voltage input side and in parallel on the regulated 800 volt DC output side. Managing stability and balancing power across the series stack are mission-critical. Through modeling and hardware experimentation, we have demonstrated that our proprietary True Control architecture can robustly manage the series stack and maintain balanced power across modules. Specifically, we have demonstrated 15 IQSST power modules operating in series and are now advancing the complete first-generation system including the thermal architecture, rack-level controls, and mechanical design. The first-generation platform is designed to scale from 1.25 megawatts to 2.5 megawatts across 13.8 kV and 34.5 kV configurations. We remain on track for a fully working system later this year, customer pilots beginning in 2027, and commercial shipments in 2028. Beyond AI data centers, we are evaluating the broader applicability of the IQSST platform across utility-scale solar, storage, and DC fast charging. In each of these markets, we believe IQSST can connect directly to medium-voltage AC, eliminating the need for a conventional transformer and simplifying the overall power architecture. This can reduce the number of stages, system complexity, footprint, and cost while preserving the same core advantages of high efficiency, fast control, and modular redundancy. While these applications are at an earlier stage, we believe the same underlying platform can ultimately support a much broader set of power conversion markets. Let me conclude. Our next phase of growth starts with residential energy systems. Across the U.S. and Europe, IQ9 microinverters, our upcoming fifth generation battery, and the IQ bidirectional EV charger significantly expand the value of the Enphase home. Together, they position us to win new battery-led systems, deepen engagement with our installed base, and address standalone bidirectional EV charging. In the U.S., prepaid lease programs like Propel add an important financing lever to support that growth. Beyond residential, we are expanding into small commercial energy systems. Our 3-phase microinverter portfolio now spans both 208 volts and 480 volts applications. The IQV-80 with the 80 kilowatt-hour battery adds commercial storage and our EV charging portfolio broadens the opportunity further. Together, these products give us the foundation for an integrated small commercial energy system spanning solar, batteries, EV charging, controls, and energy management. The next frontier is data center infrastructure with IQSST. The same architecture can extend into utility-scale solar, battery, and high-power DC fast charging. These markets require the same fundamental capabilities: direct medium-voltage connectivity, high efficiency, fast control, modular redundancy, compact design, and competitive system cost. Our expansion from residential to commercial to data centers and ultimately utility-scale is built on the same core technology foundation: single-stage power conversion, custom silicon-enabled control, high-frequency GaN switching, and innovation in transformer design. We believe this positions Enphase to compound growth across progressively larger markets while leveraging the same differentiated architecture, technology, and execution capabilities that established our leadership in residential energy systems. With that, I will turn the call over to Mandy for her review of our financial results. Mandy?
Thanks, Badrinarayanan, and good afternoon, everyone. I will provide more details related to our second quarter 2026 financial results as well as our business outlook for the third quarter of 2026. We have provided reconciliations of these non-GAAP to GAAP financial measures in our earnings release posted today, which can also be found in the IR section of our website. Total revenue for Q2 was $291.9 million. We shipped approximately 725.2 megawatt DC of microinverters and 113.8 megawatt hours of battery, above the high end of our battery guidance. Q2 revenue included $84.3 million of safe harbor revenue. As a reminder, we define safe harbor revenue as any sales made to customers who plan to install the inventory over more than a year. Non-GAAP gross margin was 46.8% in Q2 compared to 43.9% in Q1. GAAP gross margin was 60% in Q2 compared to 35.5% in Q1. GAAP gross margin was positively impacted by 15.6 percentage points for the tariff refunds received. Reciprocal tariffs negatively impacted gross margin by 2 percentage points in Q2. Non-GAAP operating expenses were $79.8 million for Q2 compared to $77 million for Q1. The increase was driven by higher investment in R&D spending. GAAP operating expenses were $123.5 million for Q2 compared to $130 million for Q1. GAAP operating expenses for Q2 included $39.7 million of stock-based compensation expenses and $4 million of acquisition-related expenses and amortization, restructuring, and asset impairment charges. On a non-GAAP basis, income from operations for Q2 was $56.7 million compared to $47.3 million for Q1. On a GAAP basis, income from operations was $1.5 million for Q2 compared to loss from operations of $29.6 million for Q1. On a non-GAAP basis, net income for Q2 was $61.5 million compared to $62.3 million for Q1. This resulted in non-GAAP diluted earnings per share of $0.46 for Q2 compared to $0.47 for Q1. GAAP net income for Q2 was $36.1 million compared to GAAP net loss of $7.4 million for Q1. This resulted in GAAP diluted earnings per share of $0.27 for Q2 compared to diluted loss per share of $0.06 for Q1. We exited Q2 with a total cash, cash equivalents, and marketable securities balance of $937.7 million compared to $930.6 million at the end of Q1. In Q2, we generated $40.3 million in cash flow from operations and $25.9 million in free cash flow. Capital expenditure was $14.4 million for Q2 compared to $19.9 million for Q1. As of 06/30/2026, after monetizing the PTCs generated in 2025 and Q1 2026, we had approximately $193.5 million of PTCs on our balance sheet. This included $108.3 million related to U.S.-made microinverters shipped to customers in 2024 and $85.2 million related to shipments in the first half of 2026. We elected direct pay for the 2024 PTCs, which are expected to be refunded through our 2024 tax return filed in April 2025. However, we have limited visibility into the timing of receipt of the $108.3 million due to IRS processing. As a reminder, in May 2026, we revoked our direct pay election. Going forward, we plan to sell PTCs on a regular basis to better align cash inflows with expenses. We expect these sales to be part of our normal course of business and the impact of this approach is included in our quarterly gross margin guidance. We announced a tax credit transfer agreement to sell $150 million of PTCs generated in 2026 to a leading financial institution with four quarterly payments from April 2026 to January 2027. We received tariff refunds of approximately $41 million from U.S. Customs and Border Protection in the second quarter with another $11 million received after the quarter end. Second quarter GAAP results were impacted by $52 million, of which $45.4 million was recognized as an increase to gross profit, $1.6 million was recognized as GAAP interest income, and $5 million was capitalized as cost of inventory as of 06/30/2026. We have submitted additional refund claims that remain subject to CBP's review and validation. Now let's discuss our outlook for the third quarter of 2026. We expect Q3 revenue to be in the range of $290 million to $320 million including shipments of 130 to 150 megawatt hours of IQ Batteries. For the remainder of 2026, we anticipate recognizing $136.2 million of safe harbor revenue, with $75 million in Q3 and $61.2 million in Q4. We expect GAAP gross margin to be within a range of 42% to 45%, including approximately 2 percentage points of reciprocal tariff impact. We expect non-GAAP gross margin to be within a range of 44% to 47%, including approximately 2 percentage points of reciprocal tariff impact. Non-GAAP gross margin excludes stock-based compensation expenses and acquisition-related amortization. We expect our GAAP operating expenses to be within a range of $120 to $124 million, including approximately $44 million estimated for stock-based compensation expenses, acquisition-related amortization, and restructuring and asset impairment charges. We expect our non-GAAP operating expenses to be within a range of $76 to $80 million. With that, I will open the line for questions.
分析師問答
Ladies and gentlemen, at this time, we will begin the session. To ask a question, you may press star then 1 on a touch-tone telephone. If you are using a speakerphone, we do ask that you please pick up your handset before pressing the keys. We do ask that you please limit yourselves to a single question and one follow-up. To withdraw your questions, you may press star and 2. Again, that is star and then 1 to join the question queue. At this time, we will pause momentarily to assemble the roster. Our first question today comes from Praneeth Satish from Wells Fargo. Please go ahead with your question.
Okay. Thanks. Good afternoon, everyone. Maybe on SST, recognizing it is early, but just conceptually, how are you thinking about balancing margin capture versus market share adoption? Based on our understanding, if you include 45X credits, the IQSSTs could potentially support very, very high gross margins. But then on the other hand, you have talked about ultimately selling the product into other markets like utility-scale solar, which presumably would imply setting a more competitive ASP. So just trying to understand at a high level how you plan on navigating that.
I am not going to give you actual numbers, but I will tell you how we are thinking about it. We are going to be extremely competitive but will clearly focus on our value drivers. There are many competitors developing SSTs, so our focus is what Enphase does differently and better. Our value drivers include fast response times. Because the IQSST can respond within sub-millisecond, we think battery storage can move to the facility space, the data center black space, which will be a key differentiator for our solution. In addition, our modularity, redundancy, reliability, and U.S. manufacturing are other value drivers. You are correct on the 45X PTC: we are finalizing those details, but that will also help us to be highly profitable.
Got it. And then maybe just shifting gears on PROPEL. Last quarter you said originations were running at roughly 200 per week, and it sounds like they are still tracking at around that same level today. Should we interpret the relatively flat sequential trend there as a function of financing capacity or other supply-side constraints? What is going to be the driver that gets you to 500 originations a week by year end? Is that based on demand increasing or your partners increasing financing capacity?
We were conservative when we started the Propel pilot, which began in four states. Sole Source is responsible for scaling and is securing financing so they can scale deliberately and properly. It is primarily a function of how many states we are in. For example, in the first four weeks of this month, the numbers are running a bit higher than 200. We will scale to 12 states: Propel is in six states now and Sole Source plans to scale it to 12 states by the end of Q3. We expect a more aggressive ramp in Q4 and their target is to exit the year with 500 originations per week.
Thank you.
Our next question comes from Brian Lee from Goldman Sachs. Please go ahead with your question.
Hey, everyone. Good afternoon. Thanks for taking the questions. Badrinarayanan, the safe harbor revenue is pretty significant both in Q3 and Q4 and it seems to be tracking higher than you have been guiding to. Is this market share gain amongst TPOs? Can you speak to what is driving that momentum? And then I had a follow-up.
We have strong relationships with many of our TPO partners. Some are well-capitalized and supported by parent companies, and there are also some new TPO partners. It reflects confidence in pursuing either the 5% safe harbor strategy or the physical work test safe harbor strategy. To date, we have executed approximately $1.1 billion in agreements: about $102 million under the 5% method and $878.6 million under the physical work test. It is a function of strong relationships with the TPOs.
Fair enough. And related to that, you sounded more positive on a return of growth in residential even in the near term. But if we adjust for the under-shipping in Q2 and safe harbor, you are implying flat revenue from Q2 to Q3 while still under-shipping. Why ship in Q3 when demand is improving? How should we think about Q4 seasonality? Do you expect to still be under-shipping into Q4? Should Q4 revenue including safe harbor be higher than Q3, including safe harbor?
Our Q3 guidance at the midpoint is $305 million. Out of that, $75 million is safe harbor, so core revenue is $230 million. We expect sell-through to be 10% higher in Q3 and we forecast sell-through around $245 million. We are assuming a modest under-shipment of about $15 million to maintain healthy channel inventory. We focused in Q2 on improving channel health: batteries are fine and microinverters were slightly elevated, and we'll bring microinverter inventory down. On an apples-to-apples basis, core revenue excluding safe harbor from Q2 to Q3 increases by approximately 10% plus. In Europe, Q3 typically has summer seasonality; we expect to be flat sequentially there. The growth is coming from the U.S. Third-party reports show a 5% increase in permits and our internal platform, SolarGraph, shows an increase in proposals in Q2 compared to Q1. Triangulating these data points, plus Propel, we expect about 10% growth in sell-through for Q3.
Our next question comes from Philip Shen from ROTH Capital Partners. Please go ahead with your question.
Hey, everyone. Thanks for taking my questions. First, when you released Q2 results, the FCC announced they are working on a plan to ban Chinese inverters in the U.S. I wanted to check in on your views. It does not really impact your residential segment much given limited exposure, but how much share do you think you could take in your commercial business as a result? How much Chinese inverter penetration do you see out there and how much share could you grab?
There are two opportunities: small commercial and utility-scale solar. I'll focus on small commercial because we have not yet introduced products for utility-scale. We are seeing lots of opportunities with large retail providers for both small and large installations. I expect small commercial U.S. revenue in Q3 to be approximately $10 million and I expect that to grow. We introduced two products in the last six months: IQ9N in December (a GaN 3-phase 480-volt product) and in June the IQ9s 3-phase product at 480 watts. From a product portfolio standpoint, we are covered. Small commercial storage is a significant opportunity; the market is diverse and could be anywhere from 1 to 2 gigawatt-hours. Our IQV-80 product is ideal: an 80 kilowatt-hour cabinet can be scaled and 25 cabinets can go to a site, potentially delivering up to 2 megawatt-hours. For example, our Fremont building will have about 1 megawatt-hour of storage comprised of twelve 80 kilowatt-hour cabinets. With FIAT compliance, domestic content, and U.S. manufacturing, we have the portfolio and expect to ramp into 2027.
Okay. Thanks. Shifting to the core U.S. residential market, the root cause of weakness is weak capital flows and challenges with TPOs driven by tax equity and FIAT and effective control guidance. Treasury guidance has not yet been issued and may be delayed. Is there a scenario where the U.S. residential outlook could still be challenged even as we get through 2027? How do you manage through that?
There is limited visibility on Treasury guidance, but the market is adjusting. FIAT guidelines are reasonable and TPO partners are becoming more mature. Tax equity is tight but we expect it to improve. For us, the opportunities are Propel, which offers an alternative to traditional loans, and our better battery offerings, including the upcoming fifth generation product in Q4 at reduced cost. We have taken pricing actions to drive volume with our fourth generation product and have broad utility meter collar qualifications at 69 utilities, including Canada, which is the highest of any supplier. The fifth generation battery will be a big differentiator for standalone bidirectional inverters as well. We are continuing with innovative financing, new products, expansion into commercial, and pursuing data center opportunities rather than waiting for external improvements.
Our next question comes from Colin Rusch from Oppenheimer.
Thanks so much. Badrinarayanan, can you talk about the elasticity of demand on the batteries? You mentioned dropping prices a little bit. How much volume do you feel you can drive as you make those pricing adjustments?
There are two actions we have taken. In Europe we were relatively high priced earlier in the year and we adjusted pricing. We are also changing the business model in Europe to generate organic demand from our installed base in addition to B2B sales. In the Netherlands and France, homeowner events generate strong leads; we are running about six homeowner events a week in the Netherlands with roughly 150 attendees each and yields around 50%. We pass the leads to installers and use lead management to ensure a smooth process. Pricing is part of the equation, but product improvements like the meter collar qualification at 69 utilities, Propel, and the G5 product are also important. Propel has about 75% battery attach, which will drive battery volumes. The G5 has 50% higher energy density and about 40% lower cost per kilowatt-hour versus the prior generation, which should further drive demand.
Thanks. That is helpful. Regarding the data center opportunity, it sounds like you are influencing facility designs. How mature are pricing conversations and how mature are the designs that would embed the Enphase solution?
On the technical side, we are making strong progress. Our team is about 120 engineers; we are building the power module and finalizing the design. We demonstrated a 4.16 kV AC series stack with 15 power modules in series and achieved meaningful milestones on medium-voltage transformer feasibility. On the customer side, we are engaged with hyperscalers, neoclouds, colos, EPCs, and the broader ecosystem, and we are participating in RFIs/RFPs which have provided significant learning. Our modular architecture has allowed us to adapt requirements without changing the power module. Because IQSST is fast and modular, we can support centralized BESS in the black space or a DC configuration with high C-rate batteries near the rack. Our next big milestone is to build a full product by approximately the end of the year, likely November, and show it to customers to open the gates for pilots and larger conversations. We are on track for that.
Our next question comes from Eric Stine from Craig-Hallum. Please go ahead with your question.
Hey, Badrinarayanan. Can you expand on Europe? Last quarter you had cautious optimism about green shoots, and you sound more optimistic here. How do you view Europe for the remainder of 2026 and going forward?
In Europe, revenue increased about 35% in Q2 and sell-through grew 30% across both solar and batteries. The markets I'm most excited about are the Netherlands, France, and Germany. In the Netherlands, with the phase-out of net metering at the end of 2026, there is significant interest in batteries and our activations increased about 100% from Q1 to Q2. We have staffed an inside sales team of about 10 people to manage homeowner event leads, and we are running six homeowner events a week. We have a base of roughly half a million Enphase homes in the Netherlands which enables scale. In France, we have about 400,000 Enphase homes and while feed-in tariffs for new installations are low, there is strong interest in energy independence and the model we use in the Netherlands is working in France as well. Germany is attractive and we have partners there; we expect to grow with the fifth generation battery. We have a strong management team in Europe and are very bullish.
Our next question comes from Dylan Nassano from Wolfe Research. Please go ahead with your question.
Hi. Thanks for taking my question. Any updated views on the shape of the cash flows from that $880 million physical work test backlog? For modeling, should we be amortizing that over the next couple of years, or is it more back-end weighted? Also, in the forward guidance for Q3 and Q4, how much of the safe harbor is 5% rule versus physical work test?
We already provided some details. In Q1 2026, we recognized approximately $34 million of safe harbor revenue and $84 million in Q2 2026. In Q3 2026, we are guiding to $75 million and in Q4 we expect about $61.2 million of safe harbor. Regarding the physical work test, we have $878.6 million signed this year plus one agreement from last year. We have not recognized any revenue from the physical work test shipments yet. Revenue recognition for physical work test arrangements is likely to begin in 2028 because these systems are being installed from 2028 onward; the tax credits remain available through the end of 2027 and systems installed from 2028 will utilize those PWT inventories. It is difficult to predict exact timing, but we expect recognition to begin in 2028 and be reasonably linear as systems are installed and customers request components and battery attach.
Okay. Thanks. Quick follow-up on tariff impacts in the guidance, specifically for batteries. In the past you talked about shifting your cell supply. Have you completed that? Are there still cells coming from China?
Tariff impacts have moderated. The reciprocal tariff impact has reduced from about a 5 percentage point gross margin impact to about 2 percentage points due to recent rulings. Our microinverter supply chain has diversified and we can move supply if tariffs change by region. We have a non-China cell source in production that we can leverage if Chinese cells face higher tariffs, and we are also evaluating U.S. cell sources as we move into commercial batteries and future battery generations. Several suppliers are interested in our business.
Our next question comes from Corinne Blanchard from Deutsche Bank. Please go ahead with your question.
Hi. Good afternoon. First, on the SST, what feedback have you received from customers and what kind of customer partnerships are you pursuing? Second, can you talk about the European Cyber Act and how you view the impact on Enphase?
We are talking to the entire data center ecosystem including hyperscalers, colos, neoclouds, EPCs, and some server providers to ensure we deliver an end-to-end solution from medium-voltage to rack. The feedback has been quite positive. We have a unique solution with a fully distributed architecture comprised of hundreds of power modules, which provides strong reliability. We also point to our history of nearly 90 million microinverters shipped to date with a 500 DPPM failure rate as evidence of reliability. On cost, our components are largely off-the-shelf commodity parts including GaN, which reduces cost. We use soft switching which yields a light EMI footprint and enables packaging in a plastic enclosure, further reducing cost. Because each power module is only 4 kilowatts and highly efficient, thermal management is easier. Combine these technical advantages with high-volume manufacturing experience from our microinverter business and we have intrinsic cost advantages. The combination of reliability, sub-millisecond response time, and cost advantages resonates with customers. Also, U.S. manufacturing and FIAT-compliant supply chains are positive differentiators.
Thanks. On the European Cyber Act, what is your view and how might it impact you?
We are fully tuned into the evolving cybersecurity requirements. We have a person focused exclusively on compliance and are meeting current requirements in Europe and the U.S. We understand inverter-based resources will be classified as critical infrastructure and will have additional scrutiny regarding communication layers and device management. We are ahead of the curve and are prioritizing compliance.
Our next question comes from Vikram Bagri from Citi. Please go ahead with your question.
Hi. Ted on for Vik. Thanks for taking the questions. I wanted to go back to guidance. The Netherlands storage activations were over 100% this quarter. What does guidance assume for activations in Q3? And regarding the under-shipments, what is the source of that caution? Is it EU demand, seasonal slowdown, interest rates, or something else?
Typically in Q3 there is summer seasonality in Europe, so we expect more or less flat performance from Q2 to Q3 in Europe. We expect Q4 to be stronger, particularly in the Netherlands where net metering expires and retrofits are expected. We do not break out volumes by region yet, but we are excited about the retrofit opportunity given our installed base. Regarding the under-shipment assumption, our Q3 sell-through forecast is approximately $245 million and we assume a modest under-shipment of about $15 million to maintain healthy channel inventory. That is the cautious assumption reflected in guidance.
Thanks. And on the SST product line, any clarity on 45X eligibility? And to clarify timing, are pilots in 2027 and volume in 2028?
We expect pilots beginning in 2027 and volume shipments in 2028. We are finalizing details on 45X eligibility and will share more when we have a clearer understanding in upcoming quarters.
It is showing no additional questions. I would like to turn the conference call back over to Badrinarayanan Kothandaraman for any closing remarks.
Yes. Thank you all for joining us today and for your continued support of Enphase. We look forward to speaking with you again next quarter. Bye.
The conference has now concluded. We do thank you for attending today's presentation. You may now disconnect your lines.