Show notes
California homeowners face a complex puzzle in decarbonizing their homes: electrification without rooftop solar could increase bills due to expensive electricity, while installing solar first risks oversizing or underutilizing the system. Balto Energy, a startup founded by James Quazi, uses AI to analyze utility bills and recommend the most cost-effective clean energy strategy. In this episode, we discuss Balto's tool, its potential to empower contractors, and what California's situation reveals about the future of clean energy policy nationwide.(PDF transcript)(Active transcript)Text transcript:David RobertsHello everyone. This is Volts for September 13, 2024, "A tool that enables solar first home electrification." I'm your host, David Roberts. Californians who want to decarbonize their homes face something of a conundrum. If they electrify their cars and appliances without getting rooftop solar, they could end up paying higher overall bills thanks to California's notoriously expensive electricity and cheap natural gas. If they install rooftop solar before electrifying their cars and appliances, they could either undersize the system for their eventual needs or oversize it and over-produce and export solar power to the California grid. Thanks to California's recent NEM 3.0 decision on rooftop solar compensation, utilities pay much less for that exported rooftop solar power than they used to.The most economical strategy for most homeowners is likely to be some mix of electrification, batteries, and rooftop solar. The more a California homeowner stores and consumes their own cheap rooftop solar power, the more value they get out of that solar and the lower their total bills. It is a complex calculation, though, that most homeowners are in no position to make. That's where the startup Balto Energy comes in. Founder James Quazi, a longtime energy modeler and entrepreneur, has built a tool that can use a home's utility bills to create a model of its consumption patterns, predict what they will be as appliances are electrified, and recommend the maximally economical approach.It's part of a larger effort to help contractors and solar companies navigate a post-net-metering world. I'm excited to talk to Quazi about why his tool is needed and how it works, how it will empower contractors, and what California's present says about the future of clean energy policy in the rest of the country.With no further ado, James Quazi, welcome to Volts. Thank you so much for coming.James QuaziThank you for having me. That was a great intro.David RobertsThanks. So, you know, I sort of went over it a little quickly in the intro there. But let's talk a little bit about this conundrum for Californians who are trying to decarbonize. So, just by background — I don't even know if everyone's been following the California rooftop solar wars, I kind of assume everybody has — but just by way of background, California recently basically issued a new policy on rooftop solar, and the long and short of it is that they're going to compensate homeowners much less. It used to be that basically you could get paid the retail rate for your excess solar, and now they're just going to pay much, much less than that.On the surface, this really damages the economical case for solar for homeowners, they'll get compensated much less. This has resulted in a huge blow to the solar industry in California. There are solar companies shutting down, jobs being lost, etcetera, etcetera. So, talk a little bit about the conundrum and how you think about solving it.James QuaziYeah, so about a year ago, the net energy metering policy in California changed from NEM 2.0 to NEM 3.0, now called Net Billing Tariff. The difference is, as you mentioned, that now customers get paid on a schedule. Each hour per year is a different rate. But generally, you can think of it as between like $0.05 and $0.08 for exported energy, while imported energy for me in San Diego is between $0.38 and $0.52 an hour. So it degrades the value proposition for residential solar for a homeowner. For contractors, it's also proven really difficult. So in the past, it was really easy to have rule of thumb sizing or heuristics, or if you took annual energy over the last twelve months and you designed a system that produced around that same amount of energy, it was generally going to be a good value proposition for the homeowner.But now, what you need to understand is, like, how much of that solar production is actually coincident with the load on the house, because the export of energy is devalued.David RobertsRight. So, the economics now have shifted to make it so that, I mean, maybe this was true already, but more true now that the ideal thing for Californians with rooftop solar to do is to consume as much of the generated power as conceivably possible.James QuaziThat's absolutely correct. So, if you can think of it as, and I'm sure your listeners are familiar with the terminology, like LCOE. So, the cost of solar, residential rooftop solar, is somewhere between, let's say, $0.10 and $0.12 a kilowatt-hour to produce, whereas the retail rate is much higher depending on the IOU that you're a part of. To the extent that you can consume cheap on-site electricity, you are hugely benefited as a customer.David RobertsRight. So then the question becomes, well, there's a bunch of different ways of approaching this question, but from this sort of like, if I'm trying to sell solar, right, I need a little bit of a new pitch, right? Because before, with full retail compensation, it's kind of a no-brainer, you could make a lot of money, but now you can make a lot less money. So this changes the value proposition for solar. So, explain exactly how the sort of calculation shifts.James QuaziSure, I would actually reframe it a little bit in terms of, like, I believe so in the previous net metering paradigm. We often saw simple solar paybacks in the five to seven years. I believe that those paybacks are still available to homeowners, but it's just a different set of products and services than simply rooftop solar on the roof. So, I think our goal is to help retool the solar industries, to help look at a house as a whole, maybe converting a lot of the energy on site that we previously ignored, whether that's natural gas or gasoline, and then power that all with cheap onsite renewables, and that will drive the value proposition for that homeowner.David RobertsRight. It's still worthwhile getting solar, even maybe still a comparable payoff period, but a different approach. And basically, it's going to be a little bit more of a complicated approach. Right? Like, it's one thing just to stick solar on the roof. Like, how much energy do I use, let's stick that much solar on the roof. Pretty easy. Once you bring in the whole home, just the combinatorial, you know what I mean? Just the calculations get a lot more complicated.James QuaziFor sure. So, like, I think in two respects. One, it's more complicated for the contractor to feel confident in the system that they're proposing and the financial outcome for the homeowner. And then two, from a homeowner's perspective, it's more complicated to understand and digest and comprehend a suite of services that might include solar and a battery and a heat pump and an EV, than it is simply like panels on a roof. Our goal at Balto Energy is to sort of do the modeling and ingest the complexity and then deliver it in a way that's consumable for both a contractor and a homeowner.David RobertsRight. So, talk briefly about what your tool does. What is the outcome supposed to be? What is it trying to accomplish?James QuaziYeah, so our perspective on it is that oftentimes in the past, if you asked for a solar quote, you would get maybe one option, two options, or three options, max. Really, like, if I take my own house as an example, so I live in San Diego, I can fit up to 30 panels on the roof, which is constrained by roof geometry, area shading, what have you. So let's call it maybe 20 to 30 different flavors of solar systems that I could possibly engage in. If I layer on batteries, I could have 1, 2, 3, 4 batteries. And then EVs, one or two EVs, and heat pump or not heat pump, water heater or not water heater.And our first step in the process is to ingest an address and then interval bill data. So, we need hourly electric reads and daily gas.David RobertsAnd that, just to be clear, this is the sort of raw information that's going into the model?James QuaziYeah, that's correct.David RobertsIt's utility bills. And this, these are available from the utility. There's no, it's not difficult to get this information.James QuaziSomewhat loaded question. It should be available. I just finished listening to your podcast on "Free the energy data." I have —David RobertsThat's why I ask. I'm wondering how straightforward it is to get the raw data that you need.James QuaziI would say that having been in this industry for 20 years, it's much easier now than it has ever been before. That being the case, there are still hurdles. There's a lot of missing intervals. There's patchwork to be done. There are services that provide synthetic intervals. It's not as clean of a dataset as I would ideally like, but it's generally like the authorization, and there are a couple of third-party companies now that do it and are making it easier.David RobertsIs it notably easier in California than it is in other states? Different in California than in other states? Or is this just a utility by utility thing across the country?James QuaziOur focus is in California right now. So, I have the most depth and experience there for this problem. Even within California and the IOUs, it is utility by utility.David RobertsSo are you restricted geographically where you can sell your product based on the utilities, whether you can get these to utility information or not?James QuaziOur position is that to accurately model a home's energy use and consumption profiles, you need two things. One is you need a physics-based model of the building, and then you need to be able to calibrate that with what is actually happening in the home. I've done a lot of energy modeling, auditing, that sort of thing. I think the one definitive thing that I've learned is that the best site observed data is actually bills. It will help you ferret out how people use their home, what their preferences are, and is actually the ground truth data.So, our position as a company is because we want to be able to confidently project — like, let's say if I converted a gas furnace to a heat pump, and I want to know on an hourly basis, what is the energy input to that system. To do that accurately, I believe that you need interval data.David RobertsSo you are in some sense beholden to utilities here or dependent on utilities to be forthcoming?James QuaziYeah, I think, unfortunately. And then to "Free the energy data" podcast. Yes, this is true, and it is being in some ways held hostage, and that's not great for the industry. I would say that our success rate right now is like, it's significant enough that we see this as somewhat of a hurdle, but not a deal breaker.David RobertsRight, right. So, I mean, getting utility bills seems straightforward enough. You just ask the homeowner and they give them to you. But when you say a physics-based model of the house, you have to go do that in person. Can you construct that from publicly available data?James QuaziYes, you can. We've done this in several iterations in the past. So, the background engine that does this is an NREL product called EnergyPlus. And it has, let's say, a full set of data requirements, which you can imagine has a lot of physical attributes of the specific house. And what we do ourselves and through partners, is comb, let's say, permit record databases and MLS listings. And we can get close enough with that set of information to build the first model. And then it's really, in comparing that model to the billing data, what's actually happening on an hourly basis, that allows us to calibrate it.David RobertsInteresting. So, you don't have to do a site visit to do any of this, really. You could theoretically do all of this modeling remotely?James QuaziYep. Everything like roof geometry, shading, building modeling, tariff engines, all the things that are sort of the processes to get to an output, can be done remotely.David RobertsAll right. And so, you put all this information into the model, and then what is the model supposed to do? And here's a question I had also: Am I the homeowner, interacting with this model in any way, or is the model a tool for contractors?James QuaziOur plan, at the very start, we're working with a set of contractors, and we're in Napa and Sonoma to start, most notably Northern Pacific. Our plan is to deliver a tool to a solar contractor that they can use to propose a wide range of solutions that a homeowner might want. I think that this will become a customer-facing tool or exploratory tool in the future, but we are definitely starting with solar contractors.David RobertsInteresting. Yeah, because one of the questions I had about this is just that I'm sure I'm not telling you anything as someone who's worked in energy for a long time, but just like, people are pretty lazy, and the way people make decisions about appliances and stuff like that is generally to ignore it until it breaks and then go to Home Depot. So, like this comprehensive, long-term, holistic planning, I'm just like, wondering, like, how many homeowners are really that committed?James QuaziSo, let me give you an idea of, like, what the output of the tool is, then where I see this going. So, you know, back to my house, 30 panels, batteries, EV's, all the things. What we want to do is expand the solution set for all possible outcomes for that house. So, if I permutate those things, it ends up being a set of maybe like a couple hundred to a thousand different individual pathways. It could be 28 panels —David RobertsAnd these are like mixes of the number of panels, the number of batteries, what kind of appliances, that kind of thing.James QuaziThat's absolutely correct. And then what we've created is sort of a decision-making framework that allows you to search that space for the thing that's right for you. At first, contextualized in one of three goal seeks. So the first one being a very standard solar approach, which is "Deliver me the best financial outcome." The second one, which we're seeing a sort of increasing adoption around, is like, "Yeah, I want a great financial outcome, but I also want to power this set of critical loads or my entire house through an outage of this duration. And I'm not cost-sensitive around that."So, like, if I need to add a battery or two batteries and it provides that service, that's fine. And then the third one is a sort of immersion. Ten years ago, when I was in the solar industry, it was like there was a time when we thought we had to deliver day one savings to get adoption. And it turned out there was a segment of the population, mostly retirees or people that were about to retire, who, let's say, had a $150 utility bill. And they're like, "You know, saving money isn't as important to me as, you know, I experienced the grid cost is volatile, but always volatile in the upward direction.And if you are going to put on the system and it has a 20-year lifespan, can you lock in this $150 for 20 years? And I don't experience any increase in cost…
Full show notes at the publisher 
Our TOPPODCAST Picks 
Stay Connected
