管理層發言
My name is Sioban Hickie, SunPower's VP of IR, and I would like to welcome everyone to the First Quarter 2026 Earnings Call. I will review a few housekeeping items before turning the call over to our CEO, Dr. T.J. Rodgers. All lines have been placed on mute at this time. This call is being recorded, and a replay will be available within the Events section of SunPower's website. Please note that today's presentation may contain projections and other forward-looking statements. These statements are subject to known and unknown risks and uncertainties that may cause actual results to differ from those expressed or implied in our statements. Also on today's call, we may discuss certain non-GAAP financial measures. A reconciliation of any differences between those non-GAAP financial measures and the most directly comparable GAAP financial measures are available within our press release. Lastly, we will be holding a question and answer session after the end of formal remarks today. For those watching via the webcast, you may submit a written question at any time via the submission box located on the right-hand side of your screen. For those joining our live Q&A, please click the raise hand icon located at the bottom of your screen to enter the queue. With that, I will turn the call over to T.J. Rodgers, SunPower's Chairman and CEO.
Good morning. We've got the Q1 '26 results to show you this morning and answer questions. First, top lines: Q1 '26 revenue was $72.8 million. That was down 9% from our guidance. Our latest guidance was $80 million. So the market closed softer than we thought it would. Not catastrophic. A 9% decline quarter-on-quarter is not bad, but it was weaker than we expected. This revenue alone would have impacted our operating income by $1.8 million, but our non-GAAP operating income was minus $12.9 million. That is a one-time event because we added $9.9 million of spending during the quarter. We had anticipated, and still do anticipate, a great Q3, and we started hiring 86 people last quarter. Now we've turned it around. We've gone from plus 86 hires to minus 115. Our cash was flat. We raised $41 million during the quarter and used all of it to pay off debt, except to keep working cash at around $10 million. Since the beginning of May, we've cut our costs by $9.9 million a quarter. That included 115 employees who were RIFed. We went from 86 hires to 115 RIFs. We implemented an across-the-board four-day workweek through September. The theory on a four-day workweek comes from my prior life in semiconductors: it's extremely difficult to build up a good workforce, and the last thing you want to do in a yo-yo economy is lose your good people. We did have a layoff focused on overhead and redundancy among our four startups, but in sales, fulfillment, and installation we moved to a four-day workweek. That means you work four days a week and get paid for four days a week. Another way to look at that is a 20% pay cut. But when you come out of it and start working the fifth day again, the people you've got remain in place and you have less of an arduous climb back. So that was the rationale for the four-day workweek. We've cut our inside sales group. We had a large call center and have reduced it to the level needed to maintain our pipeline because, paradoxically, call center sales have a lower profit margin and worse cash flow profile than our conventional sales force. Our conventional sales force has now grown to 1,552 members and represents 90% of our revenue. The inside sales group relied on many purchased leads from the market, and we wanted to eliminate that expense. We will continue the inside sales function but at a reduced scale focusing on the top producers. We reduced finance and administrative costs, which had ballooned—not for any bad reason, but because we went through a very arduous audit and had allowed hiring of contractors and employees without restriction. Now we've brought that back down. You've seen this graph pretty much all the time; it's my proxy and the way I run the company. It's a metric I can understand and others can understand. We actually run it on dollars, but I report it on headcount. Pre-merger there were 3,500 SunPower and Complete Solar people. We picked 1,280 of them to start the company. A year ago, Q4 was our first quarter, and we successively dropped that target over time. We have been targeting 820, which is very lean, and we bounce up and down around 820. As we acquire companies, we bring in people—for example, over 100 from Ambia—and that pops us up, and then we work it back down with synergies. This quarter we've dropped our target to 700. We think that's doable, and we're currently at 710. I wanted to cover that before going into the 2026 forecast. The cuts reduced our operating expense by $9.9 million a quarter. That's done. They were too late to make Q1 '26 better, hence the roughly $12 million loss. But they will be in effect for about 60% of the second quarter and will have a significant positive impact in Q2. Our current Q2 '26 revenue estimate is $75 million, up $3 million from last quarter but still anemic; the market is still weak. However, our acquisitions are starting to allow us to bounce off the bottom. The operating loss will be reduced to $3 million based on the cuts I described. So we're going to have a reasonable quarter, but a loss this quarter. I rarely forecast more than one quarter ahead, but in Q3 we believe we're going to exceed $96 million. I'll explain that in a little while. At $96 million in that quarter, we will be profitable and cash flow positive. So we're going through a mildly weak period on our way to a plan we've had all year. This version of the plan is the one we used to raise money; we raised $41 million in the last quarter. The plan shows revenue through the next three years. The guidance means what I'm telling you, what I'm planning to achieve, and what I expect to be criticized for if I don't meet it; out here is a model. Our mission is to reach $1 billion in revenue, and that run rate will be achieved in Q3 of '28. That's still on target. We still expect a big jump in revenue in 2026, and you can see a nontrivial gap in the model. As we've shown and posted on the website, that is because our acquisitions—Ambia, Sunder, Cobalt—and the recovery of New Homes from the bankruptcy are all kicking in, and that's what we expect to give us a big jump in revenue in 2026. We've done careful calculations: our current operating income breakeven revenue is $76 million, and our current cash flow breakeven revenue is $96 million. That $20 million gap, multiplied by various yields through the P&L, is what's required to pay down our debt. We still anticipate big growth in Q3. Dan McCranie is here; he's running sales and marketing for us on a daily basis. I'll let him talk about that now. Dan?
Hey, thanks, T.J. Can I get the graph up, please? Thanks. This graph shows total bookings from Q4 '24 through Q1 '26. A brief word on our definition of bookings: this is not just a signed home improvement contract from a customer; it is a signed contract plus completion of the design and funding approval, so it is a robust, high-yielding booking that we use for our forecasting. From Q4 '24 through Q3 '25, the numbers were hovering around 1,500 to 2,500 jobs per quarter. You can see a step-function increase. In Q4 '25, remember we acquired all three of our major acquisitions, Sunder, Ambia, and Cobalt. Beginning in the second half of '24, we started seeing the results of their bookings, and in Q4 '25 over 4,000 jobs were created. In Q1 '26, we had a record 4,446 jobs. There is about a three-month lag between a booking and revenue in this industry, so bookings made now in Q2 are for the first stages of our Q3 revenue plan. T.J. showed very robust Q3 numbers, a step up from about $75 million to $130 million. T.J. said he is guaranteeing at least $96 million. We are currently on track in Q2 with the bookings we have so far across all departments to meet that $130 million number. We're happy with the way bookings are going. It is predominantly the Sunder and Ambia turn-on that is occurring, particularly in the spring when the contracts get much larger compared to winter. So going forward in Q2 '26, you will have a record in bookings, and we believe those bookings will allow us to deliver Q3 revenue well in excess of T.J.'s $96 million.
Dan used two words: guarantee and well in excess of $96 million. I can tell you right now that our lawyer in New York has just had a myocardial infarction and is lying on the floor. We'll call people to help him. I want to make one other point here. These jobs, I don't deal with solar backlog. Solar backlog is like oatmeal; it's just not firm. You can't tell where you're at. So we have a definition in the company: it means you have a signed contract — the customer signed up to begin with; then you design his home for him, you show him the pictures and he signs off on that; and then you tell him the funding was approved. We've gone through our funders and he was approved. Today, third-party ownership, or TPO, is the way people fund it, meaning the funder is going to pay for the house and own the installation, and his house will be part of a pseudo-utility later on for that funder. This is really good news. Normally, if I weren't talking about a $12 million loss, I'd be bragging about this and talking about big things in the future, and I still feel that way. What?
Bernard.
Oh, Bernard. Yes. What's wrong is what we changed. Boys, I'm sitting here looking at it. I worked till midnight last night on this thing, and there are pages missing. So I need to take about a two-minute break and bring out a memory stick and load a computer with things I forgot, like our new CFO. I put it in my briefcase. I could tell you that I had planned this in advance, but that wouldn't be true. All right. Now I go to PowerPoint. I'll give you the first page right now. Okay. That one I gave you. No? I talked about the convertible note offering. Bookings are a record, we talked about that. We saw the detail. We filed our 10-K, and it was a difficult audit. I'm going to talk about that audit and what happened. The audit required restatements, three restatements covering three quarters, and they're going to happen on time. That means within the next week. Here's the P&L. In Q4 2025, the last quarter, we did $90 million in revenue with $3 million in profit. Footnote 4 says we restated 10-Q results consistent with adjustments in the 2025 10-K. So we've got a 10-K. That's filed. Everything is going to be consistent with it starting today. But I just wanted to point out that I will show you in a minute that these numbers were what we reported before, and they were close to being right. I'll explain why they weren't. Q1 was $72 million in revenue and a $12 million loss. If you ask why the loss, it's here: operating expense went up dramatically because we were getting ready for Q3. Now we're going to get ready for Q3, but we're going to do it in a shorter period of time. If I look at the 10-K audit, this is what is called prior quarterly results. It's what I told you last year in meetings like this one. We were profitable in every quarter, minimally profitable in every quarter. Our non-GAAP operating income added up to $10.9 million on $308 million in revenue distributed like this throughout the year. After adjustments, post 10-K audit, the total operating income for the year dropped to $7.33 million. Given the changes in the quarters, that was a pretty good result. The revenue for the company dropped to $300 million, and I'll explain that in a minute. There's one error that caused that. Looking at the quarters, this is the first quarter of the new year after the acquisition. In the 10-K audit, we uncovered a bunch of stuff and had write-offs, which took our profit from $2.94 million, what I would call cash profit, down to a loss. There was a little bleed-over into Q2, and then Q3 became more profitable. In this case, this is non-GAAP profit where we put in actual cash gross margin. The GAAP numbers have a different gross margin, which is lower than actual cash collected due to rules about acquisitions that don't allow reporting more than your average gross margin for acquired business. In this case, the actual cash gross margin was 80 percent, and that means we made more money according to GAAP than I reported. Old, new, and one point: this is the new source of truth. So when I talk about record profit in the future, it will be because we're above $4.85 million, not above $3.5 million. When I talk about record revenue, it will be because we're above $91 million. This is history. I wanted to show the comparison to show you that we played it straight for the entire year. If you ask which is a more believable scenario from a businessperson's point of view, the answer is this one. This includes a lot of reclassification: put this in that quarter, put that in that quarter, clean up what's on your books. We actually took the record quarter we ever had, and it got bigger. So it is what it is. This is the new source of truth and this is our base. The good thing is that for this amount of revenue and this amount of profit, I now have fully audited quarterly results, or I will in a few days when we submit the restated report. We go forward with a clean set of books and better accounting capability than we had. I want to talk about the audit for a minute. The standard auditing method is to sample line items from our books. You have to sample because there are too many line items to look at every one. Then the auditors ask us to supply independent third-party documentation that validates the books. For a given order, for example revenue, they want the home improvement contract, the work orders that show we sent people to the house, the drawings for the system that show we designed it, the invoices for the panels we bought, and other things — everything we bought to work on the house. They want the work logs showing what crew went to what house and when. They want the customer invoices we billed them, proof of payment, bank accounts showing money went into the bank as cash flow evidence, and proof of activation that the system we built is running. That typically involves getting a utility bill and showing a change in the bill based on the solar. It is an arduous task, the 10-K audit. There's only one audit a year that matters, the 10-K, which covers the entire year. The quarters are unaudited until the end, and that's when you have the final statement. How big is all this? There are nine steps in our solar installation process that lead to revenue. Our auditors required proof with hard third-party evidence for each of the nine steps on each of our 11,500 jobs in 2025. Multiply those numbers and you get over 100,000 line items. That means you have to sample. In the sampling process, they go in, grab some, and make you go through all of this stuff to prove it and document it. The sampling this year led to 390 formal requests. Our auditors said, we need this or we need that. I'm not making an excuse here. Our accounting isn't where it ought to be. I'll tell you in a minute: the head accounting person is now T. J. Rodgers. He's not used to this kind of accounting and we'll be better in the future, like right now being the future. Our prior quarter reports that I showed you compared to the 10-K, the truth for full year revenue, were $308 million versus a 10-K audited number of $300 million. When we went back to find out what happened, the extra $8 million in revenue in the prior quarterly reports came from double-booking at a legacy company, Blue Raven, from a defunct computing system, Albatross. Somebody way back, probably Q4 of the prior year, booked jobs twice. They came into our books, and we didn't start selling the things we acquired from SunPower until midyear, and we didn't really start looking through what we inherited until the end of the year for the 10-K audit, which I don't consider noise. I consider it to be something that gives credibility to the company. That's why we're complete, working hard on it, and I'll show you what we've done. On the income side, operating income — and I always use operating income rather than EBITDA, which I don't like — our prior statement was $10.9 million. The new number was $7.3 million. The difference was primarily due to pre-acquisition balance sheet assets, as I told you, and using actual gross margin instead of a calculated gross margin. These are the quarters that came from these quarters. This is the yearly total, and it's really the only part of the 10-K. These adjustments exist because of the requirement to do restated 10-Qs. These differences are clear. It went from profit to loss in a quarter, and that's when the auditors said we had to restate, and we're doing that. We've already done it. We already have agreement on the numbers. There's a filing coming up this week or early next week. So revenue did well. The extra revenue came from a double-booking in an old system. The operating income had a difference, but really the quarters being so different triggered the requirement to restate Q1 2025 through Q3 2025. I stand for financial integrity. I get extremely upset when I don't see numbers that are perfectly right and believable. I've always been that way. This is the first time in my 40-year career I've ever had a restatement. Then I had a horrible thought. When you lived in semiconductor nirvana, did you really never have a restatement? Or did your finance people, who let you work on Moore's Law and transistors, take care of finances and maybe have a restatement here or there you didn't know about? When you're bragging about never having a restatement in your career, is it really true? So I asked AI, "Did Cypress Semiconductor ever restate a quarter?" I did it four times. I've shown two of the four. I changed the question because the answers change, and I wanted to get a good look at it. Based on available search results, there is no direct indication that Cypress Semiconductor ever formally restated a quarterly report. The information shows that during its time as an independent publicly traded company, it warned of misses and often warned of upcoming quarterly shortfalls due to changing market conditions. They lowered guidance and lowered earnings and revenue targets, such as in September 2004. They adjusted results and reported GAAP versus non-GAAP results in 2016 and 2017 to account for acquisition-related costs. However, the search results do not contain reports of accounting errors, fraud, or formal financial restatements. I read this one second because it gives me a little kudos. It says, "The company, particularly under longtime CEO T. J. Rodgers, was known for its strong no-nonsense approach to financial reporting." What has changed? I will not tolerate imperfect finances, period. No question, no debate, no meetings. So we've changed. We have received and accepted the resignation of our CFO. I am not blaming this on our CFO. It's my fault and T. J. Rodgers' fault. That's simple. I run the company, and if it's not perfect, it's my fault. Our CFO is leaving by mutual agreement, and we've agreed not to sling any mud in either direction. I have been appointed by the Board of Directors to SunPower's Principal Financial Officer. That's what you get called if you aren't really an accountant but you run the finances. For approximately a month we're in the process of closing a new CFO, and I'll be the Principal Financial Officer for that month. I can guarantee you I go to two meetings a day on finances. Although I'm not an accountant, I can read stuff and understand what's right and wrong based on all my experience. The Board has appointed Bernard Gutmann, who was CFO for eight years at the $42 billion chip company ON Semiconductor, to the Board and to serve on our Audit Committee. We've made changes on our Board to bolster it. I want to introduce Bernard now. He signed up last week and was at his first Board meeting last week. Let me tell a story about Bernard. Are you showing his picture now? You should be able to see Bernard. I met Bernard at Enovix, which is my free TV studio so I don't have to do something at SunPower. Bernard is on the Enovix Board, the battery company. I sit right across from him in Board meetings. He's extraordinarily meticulous. Unlike me, he's got almost perfect handwriting, and he walks out with three pages of single-line item notes every time. We've had zero problems at Enovix. I now have two validations, my old company and Enovix, that the ship can be run right and you shouldn't have things like restatements. It's not okay. I know Bernard. How did I meet him? My SPAC invested in Enovix. We took them public. On my SPAC Board was a guy named Manny Hernandez, who was my CFO. He created CFO heaven for me while I was running a chip company, and I got to work on Moore's Law while he had no restatements for 30 years. He wanted to retire to spend time with grandkids. We said, "You can't do it." He said, "I've got a guy who's as good as me. I trained him." So I met Bernard, and he was absolutely right. These are the two best CFOs I've ever met. With that little anecdote, I'd like to introduce Bernard and have him tell you a bit about himself and what he saw at the first Board meeting, if he's up for it. Bernard?
Thank you, T.J., for those kind words. I'm very excited to join the SunPower Board. As T.J. mentioned, I had the opportunity last Friday to observe the Board meeting, and it was quite exciting. The finance issues T.J. described are a short-term challenge, but I'm up for it and that makes it even more interesting. I believe we can put the right processes and controls in place to prevent this from recurring. From a business perspective, and again I will be careful not to give the lawyers another reason to act, what I saw was very encouraging. With the series of recent acquisitions, the pipeline looks strong. Dan mentioned more than 4,000 bookings, which bodes well for the future and suggests we can move beyond breakeven toward the roughly $1 billion opportunity T.J. described. That is very exciting business-wise. A bit about my background: I am an industrial engineer by training, but I have spent nearly 40 years primarily in finance. I worked at Motorola and ON Semiconductor in various roles, starting as a financial analyst in a semiconductor factory in Guadalajara, Mexico, and eventually becoming CFO for the final eight years of my career in a heavy manufacturing environment, handling debt financings, audits, and operations. My background equips me to help T.J. and the Board with this upcoming challenge. I am ready for it. Thank you, T.J.
I will give away one little secret: when Bernard accepted, he said, "It's a really interesting company, but the pay sucks." So Bernard is like me. He's doing something that's interesting to him. Okay. What has changed? The SunPower team responsible for implementing Sarbanes-Oxley, the accounting procedures, if we had that in place right now, we wouldn't be talking about this topic. Now reports, we've changed the line of command through the Quality Vice President, Surinder Bedi, directly to the Chairman of our Audit Committee. So the people who have been sucked into the hubbub of the audit, working on SOX, are going back to work on SOX only. The Chairman of our Audit Committee is Ron Pasek. He's the only other former CFO on the Board. And he and I were overwhelmed when the audit came in with so many adjustments required. Both of us were quite surprised. I always brag about having eight former CEOs on the Board. We have an extraordinarily good Board. Right now, I wish I had four CEOs and four CFOs, but we've made a big step forward here. All SunPower responses to audit questions, so the 390, are now formal documents as opposed to telephone calls between us and the auditors, formal documents that are pre-reviewed by the quality department. They have a spec for when you respond to an auditor, what that response needs to contain. If you don't follow it, they reject your response and it doesn't go to the auditor. They understand they can't be slow, so they typically deliver to our auditors and provide our own internal audit with the requested audited document within two hours of the request. We never would have made it through on the audit in time if we hadn't turned on this process. And that's the way we're going to work forever. We bolstered our finance team. We needed to do more with people from operations and quality. I'm talking about ten-ish from either or both groups because the finance guys, and the reason I'm not casting aspersion on any of them, were getting questions that were beyond their scope. I mean, a very common result in solar is the guy owes you money. You call him up, he doesn't answer. You call him up, he doesn't answer. You call him up, he doesn't answer. Then you find out he doesn't live there anymore. Then his financing expires, so you can't get the last payment. And then you have to get permission from the new owners to get in the house. If there's something wrong, you've got to work on it. And all you have to do is have something like 500 to 1,000 jobs like that pile up, which they easily can over the period of a couple of years, and that is the quality poison that I see as more responsible for the malaise in the solar industry than anything else. This way, we're putting together a team to respond to those questions and preempt them in the future, creating processes such that nothing happens that isn't preaudited in our own company. Okay. Then I gave you this one and this one and this one. So we've done our cuts. They'll be two-thirds effective this quarter. This quarter will still be weak, but better than last quarter, and the losses will be contained a lot better. I validated again the model we used; it's on our websites and it shows we are going to make this jump. It's real. One of the reasons you've been hearing about Sunder for a while, I'll just make this one point. Sunder is a sales company. Sunder manages 1,500 reps with 100 or so internal people. Their product is a signed contract, one of those contracts with all the parts that I said earlier. They, therefore, have sold their product and it's gone. There is no pipeline inside of Sunder, is really the point. So when you buy them, you buy a machine that creates orders, and you don't start collecting orders in your own pipeline until after they're signed up. And that's why it's taken a while to fill up the pipeline. The same is true for New Homes, where we've got a lot of orders for New Homes, but that pipeline was dumped actually before we took over the SunPower assets. That pipeline was already dumped too because the builders, the corporations, they moved on and they were gone, and we had to refill that pipeline. And then I pointed out, and you can write these numbers down and do incremental calculations on them when we get profitable and positive cash flow. Dan talked about the business. Now I'm ready for questions. I apologize for the mix-up. I'm sitting here thinking, "how could that possibly happen?" And the answer is we mailed a few of my slides, not the final slides I worked on last night, emailed here, and those slides were used to bring up the projection system and everything. They weren't intended to be the report, but that's what I ended up showing you. Apologies for that. Questions?
Thank you. Our first question today comes from Derek Soderberg at Cantor Fitzgerald.
分析師問答
T.J., I appreciate all the detail you provided on the business here. I want to start with the record bookings number. Specifically, what's the average revenue per job in the current mix? And what's the assumed conversion timeline from booking to recognized revenue? And then I've got a follow-up.
Derek, the average selling price right now is about $32,000 per installation. Just as an aside, that's going up as more and more of the installations have battery attach. Battery attach is big, as you know, in California, where it's almost 100% battery attach, and we're getting very strong in California and about 45% in Texas. So ASP is $32,000 and climbing through the year. Your next question involved, I think the cycle time associated, the time between an FTC or a hard order and revenue. Was that your question?
Yes.
The median right now on that is right around two months, two-plus months. It ranges anywhere from a low of about 35 days to a high of about 115 days, depending upon the complexity of the roof install. So we use as a general rule of thumb about 90 days. So if you see our bookings pop in Q2, just track about one quarter forward, and you should see the grand bulk of that revenue.
And then the corollary of that is in the fourth quarter, when things start to slow down, you've got a bunch of bookings, and that 90 days' worth of bookings jumps from Q3 to Q4. Then you hit January, and we're still promising to come out of the January, February, March malaise, and we can see it.
Got it. That's helpful. And then, T.J., we're seeing some other solar companies over the past six months or so, specifically a big residential installer filing Chapter 11. Can you talk about that dynamic a bit, what you're seeing out there, and are you guys benefiting yet from survivorship?
Benefiting from survivorship. Okay. So yes, we're seeing bankruptcies. The big surprise, and it's public, was Freedom Forever. They're bigger than us, so therefore they should be more robust than us in hard times. We've benefited from getting some of their salespeople, not a lot. We've already acquired three other sales forces to bring us up to over 1,500. I don't want to say the other area where we acquired because I don't want the other guys to know about that. We've gotten people, and we're not hiring right now. So when we hire five or six people in a key area, that means five or six people elsewhere, typically in an administrative function, go away because we've got a 700-person limit in the company. What's bad is you'd like to say we are benefiting from the malaise in the industry right now. What's bad is we had to lay off some people, and it's not good — it screws up morale, gives you a reset. Also, your sales force are 1099s, and what that means is they run independent companies that you don't control, they don't work for you, and they can disappear whenever they want, and they often disappear anonymously, and you find out later when nobody is answering the phone why. So it creates unrest in the sales force. We're working on that right now because a lot of our salespeople are new. They're coming in thinking they escaped, hearing SunPower bragging about records, and now they're watching things unfold. There were some minor things on the Internet, minor. SunPower stopped buying leads. Yes, because the group that used them was cut way back since that group wasn't effective as our main sales force. But SunPower stopping buying leads is interpreted as imminent bankruptcy, and it feeds the frenzy. The solar industry is full of rumors, almost always unfounded. It's rare to hear the truth on the street in the solar industry. So I hate that worse than I like the benefit of being able to pick and choose good people that we can lock in.
But picking and choosing has gotten us some top talent. That's why I think you're going to see extremely strong Q2 bookings.
Our next question today comes from Gus Richard from Northland Capital Markets.
I'm just curious on the bookings in the quarter. I'm assuming those are all installs. How many of those were converted from Sunder sales?
Make sure I understand your question, Gus. Are you asking how many of those bookings...
So the first question is, I guess the first part is, those bookings are installs, correct?
Correct.
Yes.
And then of those installs, some of them, I'm assuming, came from Sunder sales. And I was just curious how many of the Sunder sales got converted into installs.
Well, the answer is these are the ones being installed. Now, Sunder — it's difficult to say; no, it's not. Actually, I prepared a slide. I'll show you the slide. Probably regret it later. There it is. Okay. So this is our 1099 headcount, number of salesmen. This is old SunPower. These guys sold loans, not TPOs, to people in the Midwest who wanted a five-year loan to put on solar. To me, it's pretty simple. The TPO pitch is actually more attractive to an individual. But this group, essentially three-quarters of them, have gone away. Then we picked up Sunder, this one you're talking about, and we still have 713 of the 900 people we had there. We picked up Ambia, that's another 300, and we picked up a company we haven't talked about publicly because we just hired them, and that's Purelight. This is another company that got in trouble, and they've got an excellent sales force. So right now, we've got a rejuvenated sales force that specializes in third-party ownership sales, and lucky we did. So how many are Sunder? About half. Direct, all Sunder. But Eric Nielsen, the head of Sunder, president of Sunder, and now our VP of Marketing and Sales, runs all these groups. They've been mixed together for 90 days. I only had this graph created so I could look at what we acquired and what it looks like. So there's your answer: half Sunder, but all Sunder, because the guy that ran Sunder runs sales for all of us, except for New Homes, which has a different sell to corporate customers and is a very small sales group that deals with that.
And then obviously, in the news is the war in Iran. New England, for example, uses LNG to produce energy, and it's better for the guys who sell LNG to sell in Europe. And their utility prices have been going up as they are in a lot of places. So how much has the change in the energy landscape, if you will, starting to incentivize consumers?
That's the biggest driving force. You asked all the questions I put in the appendix to save time. This is solar energy additions to the grid, so we're now talking about utility-scale solar. There's no oil of any kind up here. We have only natural gas that's being added today. First it says here's solar, and it says solar didn't matter enough even to be a blip on the graph until 2011. If you look at the growth of solar, it's been spectacular. Here's a bad year, here's another bad year that lasted for three years before we recovered. So solar is not immune to dislocations. Battery is the second one. And batteries, if you really think about it, there are batteries in the grid where they take some power source and store it in the battery. There are also batteries on one million houses in the United States, and they're the best kind of battery because what they do is they don't add power to the grid. They reduce the power that house requires. They store daytime sunlight energy and then let the customer use it at night to avoid the high-priced natural gas kilowatt-hour fees. So right now, if you wanted to talk about whether this market is good, the market's great. If you wanted to talk about what it means when the price of utilities goes up, that's great because our prices are going down, not up. There may be a glitch due to something in the supply chain, but our prices are going down every year and have been. This whole rise in solar is because we've become truly economically competitive. I'm not someone who goes around talking green this, green that. I go around talking about how much you pay me per month and that your bill will go down by more than that per month for the rest of your life. What do you think? Okay. By the way, I won't discuss it, but this is Q1 '26 revenue: first plan, a positive event we wanted, our second plan, our third plan, and then actuals. I look at this every day. I got nervous right about there when this second plan got created; we had drifted off just a few percent. We started reacting right then, and if we hadn't done that, we wouldn't be in the shape we are in right now to react to this crisis. Let me leave it there. If someone asks another question, I'd love to show you that graph, but I'd rather take questions. Go ahead.
We have a couple questions coming in from the web. The first one is, with the increased bookings that you've discussed, what is SunPower doing to ramp up installs to meet this incremental demand?
That's a great question. Well, we were in the process of hiring 86 people for our install organization to handle all that business. Then I came in one day, and I'm the hotshot from Silicon Valley, and I said, "Wait a minute, don't hire 86 people. Lay off 115 people." So the market has whipped us around, and the 4-day workweek I discussed was designed to allow the company flexibility. The reason my graph for revenue shows $130 million in Q3, and I've only guided to $96 million because that's cash flow positive, and that's sort of a minimum step we have to take, is that we still have to do the ramp. And actually, I was driving over here today, I was thinking about next time I'm going to call Spencer Jensen. He runs our ops. And I'm going to tell Spencer he needs to take his new employee training time from his current four weeks, where I pay salary for four weeks and don't get anything, to like one week. We do that in our sales division, and we need to get faster. So we need to be able to react faster because I'm not going to buy it upfront and spend money now on that increase that's coming later.
We have a question, T.J., to you. Last year in July, you spoke about potentially looking to wind down and exit as our CEO in about a year, which is coming up. And so the question is, would Dr. Rodgers like to revise that time line and reinsure investors of his continued attention and leadership within SunPower?
Well, one thing I kind of like is that I was retired; I was on six Boards, so I wasn't exactly doing nothing, but I was retired for six months. Now I'm enjoying being back in the full war mode. So that's one. Two, I would never leave a mess behind and have them say "Rodgers screwed it up and then took off." That won't happen. So the investors aren't going to win. We're going to win, and I'm going to be there as long as it takes.
We have looks like one final question regarding battery attachments. What effect do they have on the overall profit margin of your sales?
Batteries are more profitable than solar. The best job is called a grid-tied battery. That's where you don't even back up the house. You'd think, "Well, why would you buy a battery and not back up your house?" The answer is you buy a battery to collect cheap, free energy at noon and then dump it into your system at night if you live in San Diego and they want $0.40 a kilowatt-hour for it. And a grid-tied battery is one thing hanging on your garage wall and then one hookup. You can do two of them a day. So batteries are sort of an afterburner for us because to that $32,000, you add another $10,000 for a battery. It's $42,000.
Thank you very much. That concludes our call for today. Dr. Rodgers, do you have any final closing remarks?
Well, yes. I'm embarrassed by that. It is the last event that basically is tied to the string of misfortune we've had surrounding the 10-K. I frankly would like to thank our auditors; you realize they bulked up from 10 to 17 people. I created the memo machine and started machine-gunning them with answers. They bulked up from 10 to 17 people and stayed with us until we tied it up. Now I'm going to have three perfect quarters restated by next week, and I've got the year of the 10-K done. And going forward, I now know, as you've seen today, the details of how that happened, and it's an interesting problem to manage. I've started to realize the reason there are so few install companies that are public is, being in solar and the vagaries of having your stuff spread all over the United States, as opposed to in a nice controlled factory, being in solar, the accounting for a public company are not incompatible, but it's difficult. One of the things I'm going to do is make our accounting a weapon that's cheap, efficient, and accurate, so we can focus on the other things. I didn't talk about our new products. I didn't talk about our new bifacial panel. We just put it in the boardroom to show the Board last week so that our people can focus on that, not on the arrow there.
Thank you very much. That concludes our call. You may now disconnect.