In this episode of BuiltForward, I sit down with Andrew Krause, CEO of Northern Pacific Power, and Carl Lenox, Head of Product at Balto Energy, to explore an ambitious approach to building and scaling specialty contracting businesses through acquisitions, AI, and employee ownership. Balto and Northern Pacific Power are working together to build an AI-native platform of residential solar contractors, preserving established local brands and leadership while introducing new technology, operational capabilities, and opportunities for employees to share in the value they create. We discuss how their model works, why they believe traditional software adoption falls short of meaningful operational transformation, and how they're using AI to map workflows, identify bottlenecks, shorten project timelines, and improve working capital efficiency. We also explore the mechanics of their ESOP structure, how employee ownership could encourage AI adoption, and the challenges of scaling this approach across multiple acquisitions. It's a conversation about what it might take to build a different kind of construction company, and whether AI and employee ownership can reinforce each other to create more durable, profitable businesses.
00:00 — Introduction to BuiltForward and today's guests
03:00 — How Balto Energy and Northern Pacific Power work together
06:00 — Why residential solar contractors need to evolve
10:00 — Northern Pacific Power's origins and the creation of Balto Energy
14:00 — Combining a contractor roll-up with employee ownership
19:00 — The challenges of home electrification and energy modeling
24:00 — How ESOPs can help finance acquisitions
30:00 — The vision for acquiring and transforming established contractors
35:00 — Mapping 473 tasks and using AI to eliminate bottlenecks
39:00 — Acquisition strategy, financing, and market uncertainty
44:00 — What happens to contractors after an acquisition
50:00 — Balancing shared capabilities with local autonomy
52:00 — Connecting AI adoption, employee ownership, and financial performance
55:00 — Does AI threaten jobs even when employees own the business?
59:00 — Lessons from previous contractor roll-ups
62:00 — Financing the platform and future acquisitions
65:00 — The long-term vision for an AI-native construction business
67:00 — Why giving everyone access to AI tools isn't enough
71:00 — Final thoughts on transforming the solar industry
Andrew Krause is CEO of Northern Pacific Power, a California-based residential solar and battery contractor founded in 2009. He has helped lead the company's innovation efforts, employee ownership transition, and partnership with Balto Energy.
Carl Lenox is Head of Product at Balto Energy, where he works on applying AI, energy modeling, and operational technology to help residential contractors improve performance and expand their capabilities.
Jason Jacobs: [00:00:00] Welcome to BuiltForward, the podcast exploring where construction is heading over the next decade, and how AI and other emerging technologies will and won't transform the industry. I'm Jason Jacobs, a longtime startup founder, investor, and the host. Let's get into it
Today's guests are Andrew Krause, the CEO of Northern Pacific Power, and Carl Lenox, the head of product at Balto Energy, who are working together to build an AI-native platform of residential solar contractors with employee ownership at its core. Their approach brings together several themes I've been exploring on the show: acquiring established contractors, preserving their local brands and leadership, using AI to fundamentally improve operations, and creating opportunities for employees to share in the value that they help create.
We discuss their acquisition [00:01:00] strategy, how they're using AI to identify and eliminate operational bottlenecks, the mechanics of their ESOP structure, and why they believe employee ownership and AI adoption reinforce each other. It's an ambitious experiment at the intersection of technology, operations, and ownership, and I'm excited to see how it develops.
It's also a great discussion, and I hope you enjoy it. Carl, Andrew, welcome to the show, guys.
Carl Lenox: Yeah. Glad to be here
Jason Jacobs: I was waiting for you both to, to answer at once, which you did. Um, that's the tricky thing about doing the three. Like it's, uh, you know, it's nice because you get the, the additional perspectives, but then it's, it's a little bit of, uh, you know, every question's a jump ball.
So, um, we'll, we'll do our best, but if we talk over each other a bit, it's just kind of par for the course. Um, but, uh, really excited to have you guys on the show. Um, Carl, this is our first time, uh, meeting. Um, Andrew, we've, um, had a, I think a couple of [00:02:00] discussions, um, be- before this. May- maybe three. Um, and um, yeah.
What, I mean, what you guys are doing is, uh, interesting to me. Um, I, I wouldn't say that solar's a big focus area f- for me. Um, but what is a big focus area for me is, um, these, you know, small profitable services firms and, um, how they will and won't transform as more, you know, technology and capabilities come available, and also different ownership structures.
Um, and from my discussions with Andrew, it, it sounds like you guys are forward-thinking on the ownership and you're forward-thinking on the technology, and they reinforce each other. Um, and that, you know, makes for a really interesting relevant discussion for me, um, thus, uh, you know, why, why we're, um, having the show.
So I'm grateful for you guys coming on and, uh, let me know if I had any of that wrong as well.
Andrew Krause: I think you nailed it
Carl Lenox: Yeah
Jason Jacobs: Well then f- [00:03:00] f- for starters, so it's a bit confusing. There's, I mean, there's, there's Balto Energy and then there's Northern Pacific Power. Um, uh, I don't even know, you know, how to introduce you guys, if I should introduce you as Balto Energy or Northern Pacific Power. So, so maybe just kind of give an overview of, um, how to think about the entity/entities and, and how it or they came to be. Andrew, you wanna kick us
Andrew Krause: Yeah, I can take this one. Um, I, I don't think it's as confusing as it, you know, could be. Uh, I'm the CEO of Northern Pacific Power. Um, you could think of us as the platform company or opco in a roll-up, uh, which is sponsored by Balto Energy. So, um, the backstory is pretty interesting. We have been experiencing a bunch of challenges in innovating our industry going back five years now, and these challenges are [00:04:00] effectively retooling your standard small or medium size residential solar and battery contractor in California to meet the electrification moment.
And there's a bunch of technology, financing, and other challenges built into that. So, um, effectively, we have been part of Balto since day one, incubating them as a design partner, and Balto's assembled the team and the capital and the know-how to, um, solve these challenges, uh, faster and better than we could.
And we sort of take their, uh, product and implement it into our practices and give feedback. And together, we are looking to start acquiring, uh, small and medium contractors that look a lot like Northern. So yeah, I'm, I'm really just the CEO of the construction company, uh, and sort of the key interface with the Balto team, and I'll hand the mic to Carl, and he can explain a bit more about what they do.
Carl Lenox: Yeah. So this is, it's absolutely, um, uh, I think aligned with, uh, kind of a, a, a flavor of the, the Palantir style for a deployment model. [00:05:00] So, um, our CTO, Frank, is a ex, ex-Palantir as well as ex-Slack and AWS and other places. So he's a very, very, uh, AI native, uh, individual, um, who's brought a lot of that interesting DNA into our organization.
And so the way that we think about working with Northern is first of all working very closely together, right? This is not a arm's length software relationship. This is, we sort of get... We've gotten embedded with them. We've helped them adopt these tools. We've put frameworks around the tools that enable them to be effective, which we can get into.
Um, and by doing that, right, we're learning and bringing that back into the product, which is then extensible and scalable to other contractors as we roll them in. So that's the, that's kind of the AI roll-up motion. We don't really, you know, we don't think we earn the, the right to do this without being very, very close to the people doing the work, which is Northern
Andrew Krause: Hey, Carl, I think, um, maybe you could touch on the expanded opportunity for resi [00:06:00] solar contractors as well. So Jason, something that puts this in perspective is sort of like meeting this electrification moment via starting with the solar industry. And I think that, um, uh, that is part of the transformation that we're looking at because it's no secret that the, the solar industry's in distress right now and needing to adapt.
So maybe Carl touch on that for a second, um.
Carl Lenox: Yeah. I, part of the, uh, the motivation, uh, for forming Balto and certainly for me joining Balto, you know, coming from, I've been in this industry for, like, 25 years. I've done a bunch of different stuff, right? So, um, I've seen, I've kind of seen a, I've seen a lot of different flavors of this kind of idea. I think this is, for me, we're in a really interesting moment right now for a couple different reasons.
So the, the first reason is, is we're in regulated markets. We're, we're currently operating ma- pre- predominantly in California, and we're gonna be continuing to work in California for a while. It's, you know, it's a huge market Um, there's other states like California as well, which, you know, we'll build into.
But the, the, the, the [00:07:00] principal thing for me about in the California market is that consumers are exposed to very high utility rates. Uh, they have reliability issues, and they basically aren't getting the service that they want from the utility, but they don't have a viable alternative. There isn't really the notion of having a choice in your energy supplier in these regulated markets.
We think there's an opportunity to create that choice, and that choice rotates on the con- the confluence of solar and battery costs coming down, which they've, you know, have done historically and will continue to do. Electricity rates going up, which they have historically and will continue to do. Um, and the fact that you now-- it's now increasingly viable, uh, for, for homeowners to sort of start to switch off of gas and move their homes to an all-electric configuration.
The only way to do that economically in a place like California is actually to use solar as your generation source. If you use the utility as your source of energy as [00:08:00] you electrify, it's not going to be a very economically attractive proposition. It may be attractive for other reasons, but it's not economically attractive.
So we really view this as the unlock. Now, the issue here is that this is very complicated. It's a much more complex transaction to come into a homeowner and have-- start to have a conversation about how should we be thinking about setting up your home for this future when it's not just, "I'm gonna slap some glass on your roof and walk away," and then try and find the next guy, right?
It's like, we need to, we need to start to have a conversation that's gonna be an ongoing relationship over a long period of time. Um, that has to be-- You have to come to that with a place of trust and, and credibility and be able to, you know, know what you're talking about. You know, um, walk, walk the talk, so to speak.
Um, and this is why it's actually really-- w- we feel that the right sort of entry point to that is these solar, solar contractors that have been in their region for a long time. They have a significant customer base. They're trusted. They have, you know, high, uh, high [00:09:00] trust scores. Um, and they have cr- they have that credibility with...
And they have a long, long-standing relationships with their customers, um, that earns them the right to be able to go back to those customers and have these conversations, right? And so that, that's-- we believe there's this huge unlock there by working with contractors like Northern to enable this future of energy choice for consumers.
But there's a, there's, there's several, you know, there's several pieces to this where it's sort of this once in a generation chance to overturn the infrastructure within homes, right? Actually turn that over. Um, there's a generational shift happening in the contractors themselves, right? As people start to think about retiring and other things in the next sort of five to 10 years.
This is the su- the silver tsunami piece, which we'll touch on more. So there's the, the businesses themselves are in this moment of generational transition And there's also this moment of kind of market transition where these contractors are sort of thinking about moving beyond their tr- traditional niche into a much broader suite of products and [00:10:00] services to serve the whole home and the ho- home- the homeowners, the homeowners' outcomes that they wish to achieve.
Um, and so we're kind of-- We're, we're thinking about how, how we use, uh, AI- how we use AI to help contractors navigate that moment and serve homeowners in the, uh, outcomes they're looking to achieve.
Jason Jacobs: And, um, and j- just to understand the history, so, um, did Northern Pacific Power start independently and then find Balto? And did Balto start independently and find Northern Pacific Power? Or, um, how, uh, like, what, what is the origin story of each and of the combined, um,
Andrew Krause: Yeah. Like I can take that one. Um, Northern was founded in 2009, uh, by James Patterson as, um, a successor to another residential solar company that happened to enter our market and then leave. Um, [00:11:00] and I joined Northern in 2018 as sort of director of innovation. I was trying to bring innovation and growth thinking to this small business, and, and part of my thesis for joining was, um, getting really close to customers, just classic kind of human-centered design, having access to your customers, and understanding that customer choices are still made at the kitchen table, uh, and they're architected by practices in the industry.
We, uh, had five years of very active kind of innovation, uh, framework building, so we partnered with many companies to help bring their products to market. This sometimes is hardware, sometimes it's software, uh, sometimes it's even nonprofits. And in the process of doing that, we found this recurring pattern...
Actually, two recurring patterns. One is the succession planning transition. Uh, we founded our employee stock ownership plan in 2022 and have now run it for about four fiscal years. Uh, and we've become really a big believer in the alternative of employee ownership [00:12:00] versus private equity or other types of succession plans.
Um, and we can get into that. Uh, the other one was, you know, how can solar power the electrification movement here in California? 'Cause it's, uh, it's, it can be very lucrative financially for customers to move their home to electrification over time and to standardize that on low cost solar coming off the roof.
Um, here in California, that's, that's a great proposition. The opposite is also true. If you do it with the grid, as Carl mentioned, you will lose money, generally. Heating with heat pumps, uh, on PG&E is generally more expensive than burning, uh, fossil fuel in a furnace. Um, this had a, a set of problems in it that we were ill-equipped to solve, and we worked a couple years with partners to try to solve this, and nobody was taking it as seriously as we thought they should.
So we actually were helped to form the group that founded Balto Energy on day one. Uh, James Quasih and I were friends predating Balto. Um, he was [00:13:00] coming out of retirement, you know, most recently at Dandelion Energy and then at, at Tesla before that, and has, I think, really the ideal background for doing this.
He's built and scaled residential solar and battery programs and residential heat pump programs. So we thought, why not put the two together? And at that point, we offered Northern as sort of what you would call a test bed or a design partner or an ICP, um, and off we, off we went. Um, Balto got funding knowing that they would have an immediate access to a contractor to build their models around.
And, uh, since then, they've just added to the team, uh, both on the finance side and on the product side with Carl. Um, today, I would think of it as a lot of the ideas that we generate for Balto our first trial at Northern for, you know, approximately a year, and then we are looking to export them to others.
Um, right now some of the ideas like their third-party financing option is exported to [00:14:00] people who are, are, you know, direct arm's length customers. Um, and some of the internal ideas around the AI native frameworks that we're working on, we see applying those to companies that we would own and operate. Um, and we think that this is actually, uh, perhaps the best way to do it because you really have to have control of the outcomes.
Um, some of this stuff is deep, deep, deep change management, and we don't see kind of bolting it onto companies as, you know, a SaaS-like product, uh, being effective because you, you have to kind of get into the, um, financial outcomes of it to, to drive change. Um-
Jason Jacobs: So, so y- I mean, you mentioned that you implemented an employee stock ownership plan, um, but you also mentioned that, um, that Northern Pacific is the operating company and Balto is the holding company. Um, how, how does that dynamic work? And also, I mean, it strikes me that in a lot of situations like this, if there's a gap and there's a firm [00:15:00] to solve that gap, there might be a design partner relationship, but why cross over into ownership and, and also where does that sit relative to the ESOP that you mentioned?
Andrew Krause: Yeah. And I think, you know, using the, the common parlance, Northern is sort of the holding company and Balto's the sponsor. So Balto's investing money into Northern for the purpose of transforming Northern and also making these acquisitions. The way that that money is invested is sort of co-invested with the ESOP.
So our end state is, uh, two shareholder cap table where Balto is the majority owner. This is mostly a GAAP financial consolidation thing. It's not a control thing so much. It's just we have the controlling interest, so we can report it on our, uh, profit and loss and balance sheet. And then we see 49% ownership to employees through the ESOP.
Now, we take this in stages of getting there, um, and the reason this is effective is, is actually I think a kind of a [00:16:00] smart little compounding loop, which is today we help Balto, um, generate cash by distributing their prepaid lease, their third-party owned product. This gives them some cash that they can reinvest in the equity of Northern, and we can run that cash through the ESOP to buy at a pre-tax advantage.
So one of the, the killer things about ESOPs is, um, to the extent that the ESOP backs debt in an acquisition, you can write off both the principal and interest of that debt with the only cap being, um, 25% of the qualified, uh, payroll for the acquired company. So if you, if you acquire a company with a million-dollar payroll, you can write off $250,000 a year in principal and interest in the acquisition debt.
So there's a, there's a small virtuous compounding loop for this relationship that, um, we think is pretty nifty. Uh, we also think that, um, ESOPs generally over 30% are considered to be like, uh, pretty [00:17:00] dominant ESOP culture. So that's kinda however we got here, that's the threshold that the academics and the researchers and the companies of practice have put forward.
Um, we wanted to get it more to like a 50/50 position and just, you know, it ends up at 51/49 as an artifact of, of GAAP standards.
Jason Jacobs: Uh, and the, um, the initial aha that led to Balto getting created, t- y- I think you mentioned that there were some challenges around, um, uh, around the solar transition to play a more holistic role longer term like that, that thread. Um, what were those challenges and, and what was the, the initial thought around what the mandate of Balto would be?
Andrew Krause: I'm gonna set Carl up for this one 'cause we have a, a historian of the field on with us today. But in a nutshell, we were getting different customer questions. So we had customers that were [00:18:00] buying EVs and doing heat pumps and doing this sort of stuff that we weren't involved with. And, and generally, the contractors that were installing these items weren't necessarily thinking about how they were powered or how they might impact the bill.
Uh, it's very common for us to see people who recently put in a heat pump, and they go through their first heating cycle, and they get bill shock from PG&E because they've now taken cheap gas and made it really expensive electricity, and they call us to see if they can get cheaper electricity. So in trying to use industry standard tools to model that, we found that this industry i- is fragmented, not just in the installers, but also the tools and the finance and ecosystems and all the things you would need to kinda wrap the house in one orderly upgrade over time.
And, uh, with that, Carl, I'd, I'd love for you to expand on all the various things that you've seen in your career leading to this moment.
Carl Lenox: Yeah. So look, I, I think there's a, there's a, a series of interlocking challenges. Like all this stuff, [00:19:00] it sort of feels like, uh, you know, it's a combination lock where you gotta turn it to the left a few times and the right a few times, right? So, so part of this is, you know, this very long-standing sort of challenge that is not, you know, that's kind of been there through, through all the cycles, uh, in this industry, which is like fundamentally, you are selling someone a product or a service that goes...
that you sell them once, and then you sort of, you know, have, have, uh, some sort of liability associated with it for like 25 years, and you don't really have a resale opportunity to those customers generally. Um, that's changing, but to unlock that opportunity, you have to, again, you have to be able to find the right moments to re-engage people, 'cause this, this is not a constant thing.
They're not in an app all the time looking at their stuff. Nobody cares about that, right? Um, but there are moments where people hear about that, and we can understand and detect those and, and engage at the right point. Um, and we're, we're piloting that right now at Northern, right? We're sort of doing it proactively.
We can get into that. [00:20:00] But, but that's the... But in order to effectively re-engage people, you have to be able to come to them and say, "Look, like, here is a, here's a path for you," right? "Here's a trajectory that you could go on," or, "Here's a s- here's several different options that you could think about taking."
Um, and sort of lay those out clearly with some, with some r- some real numbers behind them. To do that, you need to be able to actually model a home. You have to be able to do very, you know, fairly sophisticated energy modeling of everyone's individual home, and it turns out that everyone's individual home is actually kind of a special snowflake Historically, in this industry, um, even, you know, uh, even today, if you're going to size solar and a battery at somebody's home, you're pulling a sort of pre-baked single profile for maybe an entire region off of a shelf and saying, "Yes, like that, that load profile represents a home."
Or maybe you're even just taking sort of monthly electricity usage [00:21:00] and saying like, "Yes, we think that this represents a home." We're in a paradigm now where, you know, your buy and sell rates vary hour to hour, right? You're putting batteries in. All these things, the load shape and h- what the home is using matters.
And very importantly, if you're gonna say, "Hey, I wanna turn that, that furnace into a heat pump," um, or your hot water heater into a heat pump hot water heater, or even I'm gonna add a pool or whatever it is you wanna do, right? Those things, uh, have very, uh, different outcomes depending on how the home is constructed and how people live in that home.
And so the, the first sort of area of investment for us is to get very, very good at modeling that, modeling these homes, taking actual weather, calibrating against actual utility bills, the home's geometry, the home's equipment, what we can detect and learn and infer about what's going on in that house.
And do this, and do this, you know, with a minimum, a minimum amount of questions to the homeowners. We're not, [00:22:00] you know, putting them through some kind of grilling. Um, and, uh, and do that remotely, so we're not trying to send people out in trucks to do blow door tests, right, which is the historical way of handling this stuff.
So this is where we've actually built a lot of, uh, a lot of capacity around the ability to say, "Hey, we can take an address. We can run it through this very sort of sophisticated, deterministic, but a- but agentically managed, uh, modeling process and come out and say, "Hey, here's, here's v- here's what your home is-- could look like in these possible futures, and here's how we would think about designing a system for you to meet those possible futures."
Whether that is a brand-new system, you don't have solar and a battery on your house today, or you do already have it, and we're thinking about what's the next thing, what's, you know, we're gonna add to that, right? Um, so that you don't have to go and add to it a third time, right? You know, like the, the, the worst outcome for everybody in this business is like, oh, you know, you sell, sell them a system, and it's sort of like, eh, didn't quite make it, and then you gotta come back.
Doing [00:23:00] that once is fine. That's good. It's a great opportunity. Doing it twice or three times, not so g- not so much, right? So that's, that's, that's kind of, uh, that was sort of one piece of it. Additionally, when we, you know, kind of got into it with, with Andrew and Northern and started to get into like, okay, how do we actually...
That's the sort of the sales, that's effectively the sales side of the, of the, of the, of the pipeline. Um, and then there's the operational side, right? And we say, "Okay, cool. We want, need to be able to deliver this. We need to be able to scale this business. We need to be able to help position the business overall for success."
There's actually a lot of opportunities when you kind of get, open the hood to help, uh, on the, on the back office side. So sort of think of it front stage, back stage, um, and financing as the, the triangle. Sorry, go ahead.
Jason Jacobs: And, and so when it, uh, when it comes to the, you know, acquiring additional shops beyond, uh, Northern Pacific, um, is, is Balto the one that's doing that [00:24:00] playbook or is Northern Pacific the one that's doing that playbook?
Andrew Krause: Yeah, I, I can take that. So Northern is the acquisition vehicle, again, the platform, right? In traditional M&A speak. There's a couple reasons for this, but notably it's because, uh, employee stock ownership plans in their charter can only buy the equities of the sponsor company. So the, the Northern ESOP can't buy the equity of an acquired company, nor can it buy the equity of Balto.
It can only buy Northern shares. Um, and so the, the key in ESOP roll-ups is that there's often a pretty material, uh, rollover equity component. So if we went to a small company, we would choose something around 40% or greater rollover equity, 'cause then they have in their hands the sellable security to the trust.
So then you can work with a trustee to show up and say, "Hey, I just acquired a company, 40% or greater of that is Northern shares. You're in the market for [00:25:00] Northern shares, right? Why don't you go ahead and use the money that you have, the pre-tax advantaged money, and/or your access to ESOP loans to buy out these owners?"
And that's how the liquidity is generated. Um, and so that's, uh, that's the reason that Northern is the acquiring company. Um, I'll pause there. Does that, does that resonate?
Jason Jacobs: Uh, I mean, this general area is, and not just specific to this case, but just this general area still gives me a bunch of brain damage because, uh, because for example, I heard or I read somewhere that, um, that there was an entrepreneur who, um, used an ESOP to make up the difference in available outside resources to complete an acquisition.
So it's like, okay, so, so an ESOP can actually be something that can like generate the funds to do an acquisition. But then it's like, okay, if we're gonna do other acquisitions, like where does the ESOP fit in? Where does equity fit in? Where does debt fit in? What happens to the ESOP ownership? Does [00:26:00] it get dilut- you know, do employees get diluted like they would if I went and raised a venture round and, and I, you know, and, and uh, you know, and, and employees in a startup would?
Like it's, it's still, it's just, it's a ... I'm finding it's a lot to get my, my brain around, but maybe I'm just not that smart
Andrew Krause: Uh, well, we don't think that's true, but, um, I think there's a, a bunch of Common myths on ESOPs that are unfortunate artifacts of maybe the early days of ESOPs and, um, a lot of rules have changed and banks have gotten comfortable. I mean, the, the big thing for us is it's very simple. Um, there's just a handful of residential solar companies in California that do more than $20 million a year in revenue, and they might do that at, you know, if they're really good, 8% to 10% EBITDA margins, right?
If they're not good, they do it at negative margins or if they're, if they have other strategies. Um, and this basically takes our market out of the private equity roll-up camp and into [00:27:00] looking for a buyer of any kind camp. And ESOPs themselves can't go down to very small levels. You generally need about 30 employees.
You probably need at least a million dollars in EBITDA earnings at the end of the year. Um, you can be an S corp or a C corp. That part doesn't matter too much. Um, and, and so we met these conditions and, uh, formed the ESOP. It does cost some money to form an ESOP, but once you're an ESOP, it can be administered very cash effectively.
It's not some big annual thing. And in fact, I think ESOP trustees and appraisers and other advisors are actually very helpful business partners. They're sort of like a non-voting board. Um, and, uh, the, the, the key with us, and I think this is ESOPs in general, is that expanding participant count is now more done through being acquired by an ESOP company than forming a new ESOP, just for, I think, rather obvious reasons.
If you're a small company, let's say you're $100,000 a year [00:28:00] to one million a year in, in bottom line earnings, uh, and you're thinking about your succession plan, the ESOP is probably your best bet. Um, one, if it's a C corp ESOP with some meaningful ESOP percent ownership, so 30% fully diluted, uh, you can actually defer capital gains on the sale of your business.
For a small owner, this is a big deal because your, your, your basis is probably very low. You probably put in 50 grand and spent 10 or 15 years, and you don't get, uh, any of that considered on the sale. You pay 40% to 45% effective tax here in California. So already net of taxes, ESOPs make a lot of sense when you can do the exchange.
Um, as I mentioned earlier, ESOPs can, uh, write off principal and interest up to a certain cap, and this is super helpful because it's basically a 30% advantaged sale. I can buy companies with pre-tax dollars, not my post-tax dollars. Um, in terms of how it works for employees, think of it [00:29:00] as employees have, uh, basically a profit-sharing interest in a trust.
That trust is entitled to its percentage ownership of the profits. Each year, a distribution is made into the trust And when employees gain shares, it's at today's value, which is determined annually. And when they retire out, let's say 20, 30, 40 years from now, those shares are, uh, uh, there's an obligation to repurchase those shares from them with cash at the future value.
So they get to ride the appreciation like that. Um, are ESOPs diluted along with other financing activities? Yes. They, they act just like any other shareholder. Um, in a minority ESOP situation, uh, you would just have to use kind of business judgment. We're not unduly diluting, uh, these members. In a majority ESOP situation, um, you're probably not seeing that as much.
Um, you're pro-- like in 100% employee-owned company, you're probably [00:30:00] seeing more traditional financing, not equity financing.
Jason Jacobs: Uh-huh. And so this is, this is... I mean, I know we're talking about the ownership structure here, but uh, what are you building?
Andrew Krause: uh, in terms of a roll-up
Jason Jacobs: No, the company. Like, what-- if someone asks, like, what, like, why do you exist? Like, what are you, what are you trying to do? Um,
Andrew Krause: Oh, yeah, I think
Jason Jacobs: how do you answer?
Andrew Krause: I think that's simple. I mean, we're here to purchase the best residential solar companies, improve them with the products that Balto is creating, give them access to the tools they need to grow, and help the owners find a dignified exit while giving the employees a soft landing. I mean, what you're seeing right now, Jason, is a lot of small residential shops are distressed.
Their number one move is cutting headcount. You have people with lots of training and lots of time in the saddle, uh, just being removed for cost purposes 'cause there's not enough sales to run the company. We think that with our structure and with an ESOP, we can both avoid that and we can help people [00:31:00] accelerate adoption of new tools like notably AI to improve their company and have better profile.
I mean, the, the dirty little secret of solar companies right now is most aren't sellable. There's not a buyer. They're too small. They're too disorganized
Jason Jacobs: And, and so if each one is a stress and you're rolling them up, um, what is the theory of change in terms of what will make the, you know, one plus one plus one plus one equal 10 on, on the other side?
Andrew Krause: Carl, you wanna handle the investment thesis?
Carl Lenox: Yeah. Yeah. So you're kind of getting at, like, what's the, you know, what's, what's our reason for existence? I think Andrew answered sort of a, a, a portion of it. Um, that's in a way, like the, the tactic of w- like how we would transform these, uh, these target long-tail companies, right? Why are we targeting this particular type of comp- companies like Northern?
The reason is that they are, [00:32:00] they are reputation-driven businesses, right? So what we're buy- what we, what we're rolling into the company is talent and brand equity and install base that is, that is, that is open to a, you know, follow-on business in many ways, right? Um, as opposed to other approaches to this market that have been done historically, where the, the customer base is somewhat, uh, disillusioned with the decision that they made, right?
That's, that's a thing in the solar industry, unfortunately. Um, the companies that, that we work with flip that script, right? And so we're, we-- because we-- it's very important to have this long-term relationship over time and helping customers transform their homes, it's very-- that, that, that sort of credibility and reputation is super important.
So that's why we're targeting [00:33:00] these companies, right? But they're in a, they're, uh... As, as Andrew alluded to, many of them are in a bit of a, a sort of a growth trap. Um, and the reason for that is, you know, and we haven't talked about this, but kind of alluded to it, um, a lot of these companies, the owners actually have personal guarantees behind the debt that they incur, and there's a lot of working capital risk.
And that working capital risk is not, you know, to them actually personal. Um, we can help with that in a number of different ways. But the, the fundamental thing to keep in mind is that th- there's, there's a bit of a dilemma there, where if you see an opportunity coming and, and, you know, expanding, expanding market share or expanding your, your offerings or whatever the case may be, you may not be able to or want to actually grow Because that is-- You're taking, you're taking a risk that that opportunity doesn't materialize, and then that risk is, is your [00:34:00] mortgage, right?
So, uh, so we-- Th-that's part of the-- Part of what we're trying to do is to trans- help these companies transform, give them sort of a, uh, uh, a financial basis that allows them to pursue those opportunities, give them the tools to pursue those opportunities with better operating leverage. They're not... You know, they can basically increase their top-line revenue without increasing their headcount. That's one of the things that AI is doing for us, right? And ultimately, again, like the big, the big... And ul- and ultimately, to serve the homeowner, right? To go after this opportunity to provide a, uh, a, a, a more, uh, a, a- an alternative supplier for their energy needs, right? That, that, yeah.
Jason Jacobs: So, so what, what is it about being part of Northern Pacific post-transaction that will give them better odds of running a healthy, enduring company than, than when they're, um, independent?
Andrew Krause: Yeah. Um, so we've, we've actually [00:35:00] solved this piece or, or have done the first proof on it The value stream of Northern is like the very first thing we mapped. Um, uh, long, you know, long story short, Balto, uh, was part of the second cohort of the Palantir Builders Fellowship, which happened back in March.
And as part of that, we adopted the Palantir Enterprise AI, uh, platform and started building the ontology of a small solar business. We mapped our value stream end to end, and looking at this map, you go, "My God, this thing is complex," right? To take a contract through to completion, median install time is 154 days.
There are 473 specific tasks that go into producing that project, and they're often picked up and put down, and then there's wait, and then they're picked up and put down, and then there's more wait. So when you map the value stream, you can kind of see using existing data that we had, just our CRM data and our ops data, uh, where the bottlenecks were.
And so all of a sudden, [00:36:00] we could start to reason about what would happen if we removed bot- this bottleneck or that bottleneck. And we started pairing the, the owner of that bottleneck with skills that we built. So think of a agent and a owner co-piloting these skills. Um, and we both, uh, attacked bottlenecks with, uh, these, but we also created, um, baseline metrics.
And in those metrics, we saw potential to reduce cycle time by about half. So taking 154-day cash conversion cycle down to about 60 days, and we do that just by, um, in part automation, but also a bunch of other things that, um, agents have proven to be good at when paired with humans. Um, and then we took the net working capital, uh, and unlocked it.
So if you imagine how much net working capital is tied up in these long cycle projects, and at any given time, we might be doing 100 plus concurrent solar projects with a team of three project managers. Um, speed equals working capital. And so you have this kind of one-time jump [00:37:00] in working capital for the same number of projects.
On the margin side of things, we took an 8% margin profile and built a roadmap to 20% margin profile. I mean, that's pretty significant. Um, it's pretty rare that a small contractor in our field's doing more than 10%. And what goes into that, um, you know, we can work through, but this is sort of the stuff that you would, uh, traditionally associate with like a shared back office.
And in this case, we've just built the AI part of it. Um, I talk to a lot of small contractors in our industry, and everybody's adopting AI, and some people are doing really great with it, and some people are just, you know, talking to Claude on a episodic basis. What I think is missing is the difference between using AI tools from the frontier labs and building an AI native company, of which there are many components, some of which aren't AI, including the change management piece and other things.
So our goal, and this is like unproven right now, Jason, but this is our goal is to acquire [00:38:00] the best local legacy brands in key territories and help accelerate that process for them. In the process of doing that, we make their company more valuable, and we give them a built-in buyer. And this to us is like the virtuous cycle
Jason Jacobs: So, so, um, uh, what, what criteria makes it-- Well, f-first of all, just from practical standpoint, um, how many acquisitions have you done? How many p- do you plan to do, and over what timeline? Just to kind of set the stage for, for, um, you know, for the substantive discussion about, you know, like the playbook once you incorporate acquisitions.
It's like, well, like, are, are... You know. I th- yeah, so I, I think how many you've done, uh, just kind of f-
Andrew Krause: Yeah. So our-- Just going back a couple points, I mean, our model is phased, and in the first phase, there are partnerships where we're working with partners to distribute the tax equity product, which raises the funds to do this. So right now, [00:39:00] Balto's in that phase and has completed the initial acquisition of Northern Pacific.
We have not tucked in any companies yet. We are working with the companies that are targets for that, but they're still in phase one. So there's a lagging dynamic to this. That is designed because it's the most cash-efficient way to do it.
Jason Jacobs: And the, and the, and the, and the, um, and re- and remi- and j- and just repeat again to make sure, so I make sure I understand, um, what is the relationship in phase one? What are you providing for them?
Andrew Krause: So the industry, uh, after the, the big beautiful bill last year was sort of abruptly transitioned from traditional cash buyer to now prepaid lease, which is the nearest thing to cash. It's not exactly the same. There are some traps to them. There are good prepaid leases, and there are bad prepaid leases.
Um, but what a prepaid lease does is allow a third-party owner, like Balto, to, uh, give the customer effectively a 30% discount, which is equivalent to the, the ITC that they n- can no longer access. And in doing [00:40:00] that, Balto can turn around and transfer that tax equity to generate cash to an institutional buyer.
Balto has done this, and we have, uh, we're repeating this cycle. The more prepaid leases they place, the more cash they have to run acquisitions. So the model is find good target companies, which I'll let Carl expand on, partner with them to drive prepaid lease uptake, and then use those proceeds to reinvest in that same company to improve and buy them over time.
This is a, like, a positive feedback loop kind of mindset. We have a handful of companies that we think would be, uh, good long-term partners right now. The real question we have in our industry is timing and pricing risk. The industry is extremely volatile right now. So while we think it's good to partner with these companies and start generating cash, we're, uh, outside of Northern, we're moving with extreme caution on, um, on just basically ecosystem risk.
And I [00:41:00] expect this to resolve sometime in the middle of next year. In terms of, uh, how big could this get? I think we are thinking between 30 and 50 companies in California, which would drive a total value of about a billion dollars, um, when you, when you add up the sum of the parts.
Jason Jacobs: And, um, Carl, can you maybe just touch on the c- the criteria that makes a good acquisition target? I know, I know it's not a big focus area given the volatility, but once that settles... Or like, what, you know, what needs to settle, and then at that time, what, what will you be looking for, or is it too early to say?
Carl Lenox: Yeah. No, I think we, we have a pretty, like, reasonably clear criteria for what the kind of companies that we wanna be working with. So as, as Andrew alluded to, we have, we have this prepaid PPA. The, the basic idea here, right, is that there are a number of these out on the market. They're generally oriented towards, um, providing customers who would otherwise be paying cash with this discount.
But then there's also some other, I'd say, sort [00:42:00] of terms and conditions attached, such as you give up control of your battery system for these, uh, financiers to, uh, try to create VPPs and things like this. Um, d- most customers actually maybe don't-- aren't fully cognizant of, like, the implications of that when they go and sign these contracts and are often surprised.
And so a lot of the feedback that we... And/or, uh, the way the, the- these companies kind of approach, um, the projects financially create sort of cash flow risks for the contractors. So these are both sort of failure modes that we either see or anticipate that we're working around. We've designed our product, and this is-- it's, it's very deliberately a wedge, uh, to get-- to start to solve a real problem the contr- the kind-- that contractors that we are interested in have, uh, which is to be able to continue to offer something that's as close as possible to a crash-- a cash transaction [00:43:00] while maintaining eligibility for the investment tax credit, and therefore this discount.
So, uh, that's a great way to sort of start a conversation with contractors that meet, that meet our criteria, which are that they are at a certain sort of scale, you know, they're, they're well-established. Um, they have a great reputation. Um, they predominantly have been selling cash because that's been m- I would-- so there's, there's very strong correlation between being focused on cash historically and being a high-quality contractor in our, in our space, as opposed to selling other people's, um, lease paper. Um, and, you know, have a, have a, have an owner typically who's, uh, very engaged, very interested in intellectually-- in-in-interested in, uh, transforming their business, intellectually curious, not looking to sell out and go to the beach on day one, but looking to stay engaged and maybe change, you know, have a more intellectually, uh, uh, stimulating, [00:44:00] uh, job for the next few years, right?
Um, so those are the sorts of things that we're, that we're looking for. We're looking for people who are actually very interested in, in, in transforming their business and eventually transitioning it to their employees, but not, not tomorrow. Um, and are, you know, really building a durable, a durable company.
Um, and that's, that's essentially like we're, we're, we're really for quality, right? Is like at, at the highest sort of like macro level.
Jason Jacobs: And what are the things that, that, um, that remain independent, and what are the things that, um, uh, you know, that like, uh, you know, for example, the brand. Um, you know, are there centralized functions that, um, you know, that, that they then use that they don't, you know, that they don't have to have in-house anymore?
Like, like what, what does life on the other side of the transaction look like?
Andrew Krause: Yeah. I've, I've thought a lot about this, and we actually have a team of four at Northern Pacific that we've labeled [00:45:00] forward deployed engineers in the classic Palantirian sense of the word. Um, the way we see this working is Identify the brands that Carl mentioned. Um, I would also add to his list a large legacy fleet of homeowners going back to, you know, even NEM one time.
So we're, we're building tooling, improving this at Northern, that we can go back to our customer base and expand on our interactions with them. It's not one and done, but it's multiple upgrades, referrals, et cetera, um, in this kind of declined market. Um, we would install some forward deployed engineers in the company for the transition period.
They would work with the CEO who keeps the brand, keeps the employees, but maybe, uh, starts the process of transitioning to basically co-pilot mode. The very first thing we do is we use AI context capturing tools, think of Granola, uh, video capture, other things, to map out the value stream of that business, and we attach whatever [00:46:00] data they have to help us enrich that value stream.
Um, we had a lot of this at Northern. We had, you know, uh, timestamped, logged database data for progression through a sequence of milestones. Not everybody has that, but, um, we're, we're trying to build the best picture of the value stream as possible. Then we're working with the employees to understand their upside in the employee ownership piece, what that is worth today, and to build a financial plan for them to help us transform the company.
Note basically choosing where we intervene first. Um, and we go through these repeated cycles of finding the bottleneck, valuing the solve for that bottleneck, solving the bottleneck, and then evaluating the outcome of that. At Northern, for instance, we had a process where the sales that were made would be handed off to operations, and this process ran through one highly critical person on our team, and it was responsible for 30% of the entire downstream pace of all [00:47:00] projects.
We cut that process time in half and improved this person's, uh, life basically, 'cause they were doing this tedious checklist over and over again, uh, in order to take the sale and turn it into an operational, uh, production. Um, and that bottleneck today has like made this person's life better and it's also, um, increased the flow of products through.
And we're doing this now systematically through all bottlenecks. W- we imagine at the end of that period, the company would be more or less AI native and that the culture could be transformed to understand the incentives of helping to co-pilot this. Um, we don't have any proof outside of Northern that this is transferable.
That's what we intend to do next. And, um, uh, I think there's like lots of, lots of sub-issues in there if you wanted to get into them. But, um, I, I have a high degree of confidence that we can do this. Our companies are mostly similar. We mostly do the same thing the same way most of the time.
Carl Lenox: Yeah, let me add, add a couple of things to that because I think, like there's a... [00:48:00] Yeah, just to add to it, right? I think one, one area that is important to sort of, uh, understand about the model we go in and we start having these conversations, you know, with, with companies like Northern, um, part of it is getting, kind of getting the right sort of talk, kind of like frameworks, um, capabilities in place that sit behind these agentic tools, right?
So it's actually a tooling challenge. It's sort of a data challenge. Like let's get data, make sure we have good data, make sure we have the right data governance in place. These are not things that sort of happen on their own, right? And you can't operate agents effectively if you have bad data feeding them fundamentally, right?
You'll make mistakes. They'll make mistakes. And the third thing is you start to identify the people in the organization who may be in one role today, and they're the, they're, they're the ones who are sort of the most effective in, uh, in actually gaining outcomes from those tools. You start to think about, okay, cool.
We've made-- We've, we've just like kind of, [00:49:00] um, worked you out of a job, right? To some extent. But those are the, those are the exact folks that you want to then go and help make that move in the next company that you bring in, right? They have, they have the context, they have the experience, and now they have the AI experience to be able to credibly move into another company and help them guide them through that process.
So this sort of becomes sort of like a, an, a, you know, an AI breeder reactor, right? Um, maybe it's only 10% of people in a company that are, that have that profile, but those are the 10% of people that, you know, you kind of get the system set up and then move them on to the next, and that, that's sort of this forward deployed motion.
Um, the, the other, um- piece of this is you kind of asked, like, what does it look like for these companies? I think what this is, this is an ongoing negotiation discussion, right? Um, it's-- our view of it is it should, should be very federated, right? We wanna centralize the things that are valuable to be centralized to the folks that we're [00:50:00] working with.
This is not a tops-down thing. This is a, we come in, we have a conversation, and we say, "Hey, these are some areas where we've been able to deliver value to f- to Northern and other folks as they come in. Is that something that you want to be a part of? Is that this is part of your business that you do want to kind of have the back office handled," right?
But by and large, I would say the reason why we're acquiring these businesses is because they're good at what they do. And we don't wanna lose that. We don't wanna lose that leadership. We don't wanna lose those employees. We wanna make them better. We wanna federate that, um, that domain expertise and, and th-th-those ways of working, right?
So it's a balance, right? You wanna bring in, you wanna bring in, like, new ways of working, new ways of doing things, centralized tools, um, and, uh, you know, where you have economies of scale, leverage that, but without, you know, without losing the, the DNA that makes these companies ultimately that's made these companies been successful for a [00:51:00] long, you know, for a long period of time.
Jason Jacobs: Uh, and, and so, um, w- when you look out over the next 12 or 18 months, what are the key priorities for the firm?
Andrew Krause: I think we have respective key priorities. I'll take Northern's. Um, I think the biggest thing for us is we have our value stream map, and we're solving a bunch of the independent, you know, bottlenecks. And when you solve one bottleneck, the new bottleneck emerges somewhere on the stream. The question of how to show how this adds up to monthly, quarterly, or annual financial performance remains, like, a hard to answer question.
We have ideas around showing the impact of these interventions to long-term financial performance, but keep in mind that our, our capital cycles are 150 plus days. I mean, you do something today, and that's on just one project, and then you have to see it matriculate over [00:52:00] lots of projects for a long period of time.
So we just can't know yet what degree this experiment is working on at the, at the systems level. I would love to solve that. We have very good engineers thinking about solving that. We have a strong observability culture that helps us try to solve this, but y- you just have to pay for it in time. Um, the other thing that I think is something that's on my wish list is how to connect the financial value creation from employee ownership to daily actions.
And it's, it's like theoretically possible to have almost a daily moving stock ticker for the business because it's, it's valued in a kind of repeatable, predictable way. Uh, but connecting people's daily actions to wealth creation, I think is key. And the thesis behind that is, I think AI will do substantially worse in a firm if there's not an employee ownership component to it.
I think AI without upside for the frontline workers, uh, is, is [00:53:00] more like a threat to their job. Whereas what we're seeing is people leaning in and being super excited about using AI to make more wealth, and I'd like to actually connect those dots for them. Um, and I think employee ownership without AI maybe struggles to generate, uh, value to help create wealth.
So, you know, in, in some forms of employee ownership where businesses are not growing, they're not actually, um, generating wealth for employees. They're typically paying off the ESOP debt that was used to buy the shares. So you kinda need both, is our view. Um, and that, that-- those are really the top two big points at Northern.
And Carl, I'll let you handle the, the Balto, uh, talking points.
Carl Lenox: Yeah, I'd just, I'd just add to what Andrew said. I think, like, a key point, right? When you're, when you're looking to bring these tools into, into a firm, what you're really asking employees to do is to own outcomes. You're sort of saying, "Hey, here's a bunch of, you know, minions you can go manage to deliver outcomes." Um, [00:54:00] you don't get employees that are focused on outcomes without an ownership mentality, and I don't think you actually get an ownership mentality without ownership, right? Um, otherwise it feels extractive for the f- for the people who are participating in it, and they, you know, they may just peace out, right?
They're not... They, it's gonna be, it's gonna be a much hard... The change management becomes much harder, um, if they feel like, "Well, what I'm really doing is giving up all of my hard-won knowledge and working myself out of a job." Right? So that's, that, that's just to add to that, that part of it.
Jason Jacobs: yeah, I, I mean, I guess the question that comes to mind for me is that if it truly was a threat to a job, then how does giving ownership not make it a threat to the job? Because at a certain point, if you really do automate yourself out of a job, then is the company just gonna keep you on the payroll indefinitely?
It's like, well, no, you're an owner. It's like, all right, well, what is that worth relative to, you know, to a salary every year in perpetuity? And, [00:55:00] and, and if that math doesn't equal out, it doesn't matter if I'm an owner or not, like a threat to my job is a threat to my job, right?
Andrew Krause: Yeah. So I love this question. Um, not if you practice continuous improvement. So what happens right now is employees are often sub-optimized in what they can do versus what they do do. They're doing things like data janitorial work or data entry or things, you know, that are, are literally taking 30, 40, 50% of their time when they could be off in a higher and better use.
So we kind of ascribe to the, uh, blue ocean thinking concept of value cost trade-off and constantly trying to break that. I think what we find is, uh, like a maturity model for AI. Yes, we are going to help take you out of some of the lower value roles that you're currently in. Um, for instance, I ran a study earlier this year that showed that, um, people who do permits in our office were at about 20% historical utilization, and the project [00:56:00] manager is below 40%.
That doesn't mean that they can't do other valuable things. It means that we need to help backstop the work that's not producing the value and move them to the edges of where the value actually is. And what we find in that process is people have incredibly good ideas for improving the business. This is classic, you know, high engagement culture thinking.
Um, they didn't have an outlet for manifesting that or to get rewarded for that. And so, you know, we have our emerging stories of people who are doing ordering differently. They're no longer ordering parts way in advance where we're sitting on inventory and working capital. They're now ordering using AI to do smarter ordering just in time for a project without, uh, supply chain risk, right?
So that's kind of the transformation that I think we're after, Jason, and you're right. Y- you can't just slap AI on what we currently do and call it a day. Um, that would be, you know, the rational thing to do there would be to let everybody go. You can repurpose people who have [00:57:00] hard-won knowledge into higher and better functions, and that is like the required non-AI part of the transformation.
We have strong signal on this, but honestly, it's-- I think it will take a couple more years to really see where that gets us.
Carl Lenox: Yeah, and let me add, I mean, the other side of that, right? Like, there's two sides to this coin. One side is, okay, cool, like you can make your existing work-- Like, uh, just to be super clear, like our thesis is you make the existing employee-based workforce more effective. More effective to what end? Well, in order, in order to actually get, uh, get the results of that leverage, you also have to grow the business, right?
So the idea here is very simply we're, we're growing top line without, without growing, without growing opex, and that's, that's how you get the EBITDA wedge, right? So not-- That, it-- By making people more effective, you're able to move more projects through the, through, through the organization, uh, annually, right?
And so your revenue goes up. Um-
Jason Jacobs: the, the, the-- Oh, finish your thought, Karl. [00:58:00] I didn't
Carl Lenox: no, that, that, that's it. Simple as that.
Jason Jacobs: So, so they s- they say that, um, that history doesn't repeat, but it rhymes. Um, when you think about this moment that we're in and what you guys are trying to build, are there, you know, kind of corollary moments and corollary firms and corollary industries that you think about historically, or do you feel like you're pioneering in every dimension?
Andrew Krause: We do a lot of first principles thinking for sure. Um, I think we're students of the AI native roll-up concept in professional services. Uh, get to see a lot of success in white glove, white collar type businesses. I think this is more and more coming to blue collar trades. I think it's gonna be fundamentally different, um, just how trades work and the fact that we're in the field and not at the desk, and there's a lot of complexity.
Um, I would maybe contrast this most either against the private equity roll-up, particularly of HVAC shops [00:59:00] or former failed solar attempts to roll up. And, um, I think we're studying the failure modes of these, and there's not a clear answer on, um, what it will take to succeed. But, you know, the-- that's kind of the opportunity for us, right?
Like we have a great product in a distressed industry. This product has a bright future, but needs, you know, the, the companies need to be transformed to meet that moment. There's good examples from not in our category, and there's bad examples from our category. So how do we kind of proceed forward with that?
I mean, that, that's part of the fun of it. Um, Carl, do you know of any comps that you would throw out there, or is that kind of it?
Carl Lenox: I-- Well, I would say, you know, the, the general thesis of-- that I laid out in the beginning of, hey, you're in a-- you're a homeowner in a regulated utility market, you don't have a choice. This was actually the original thesis, uh, uh, well, people didn't articulate it that way, of, you know, the original, [01:00:00] you know, 30-- so- solar to third party ownership, like solar only, right?
Th-thinking the Sunrun and SolarCities of the world, right? It's like, "Hey, you sign this contract, you get 10% off your electric bill, everything is happy." That sort of approach, and, and I, and I, I wanna say this is important because that approach worked really well under this traditional net metering construct, which is a very, very simple value proposition for the homeowner, so that the sales, the sales motion was also very, very simple, and you could use very minimally trained people to go out and sell this value proposition. Okay? Um, this has also built up a lot of, I don't know, cultural debt, let's just say. Um, there's sort of a lot of path dependence in the way these companies have evolved over time, um, that led me to say, "Look, this has actually gotten a lot more complicated." The approach to that has been to sort of like bolt on a bunch of like different things to this business model, but it's sort of discordant. Um, there's [01:01:00] a lot of misaligned incentives across the whole value chain in the way that this has actually played out, given the complexity. If you were to do this fresh, clean sheet, you'd probably do it differently, right? To be able to provide that same fundamental value proposition. And so for me, when I sort of looked at that, it's like, look, there's actually a really interesting counter-positioning opportunity here. So the historical precedent is, hey, we wanna deliver promise, right? That I think a lot of people in, in the residential solar space have been aligned to for a long time. Um, but there's an opportunity to kinda come in and, and take a fresh look at it with the tools we have today and build up from this base of, um, really high quality operators to create that bigger company that can deliver those outcomes at scale and with quality.[01:02:00]
Jason Jacobs: So, so what do the capital requirements look like, and what type of capital do you anticipate, and how do you think about the, um, the return profile for this type of, uh, I would say company or companies? I, I still don't, I still don't know how, how quite to, uh, to get the wording right.
Andrew Krause: I mean, just to restate, um, is a sponsor, Northern is a platform. Classic technique. On the Balto side, you know, they've raised some corporate equity. The returns would be in line with what you would expect a venture return to be. Um, Balto also uses, uh, credit to carry tax equity until it's monetized, has a facility with US Bank to do that, so that's traditional.
Uh, and then has, uh, investment tax equity buyers, has an institutional partner to do that. Um, that partner has been great. They're a OEM in our ecosystem. [01:03:00] Um, they really believe in the mission, and at some point, Balto will need to scale beyond that to a second partner. Uh, we could imagine that being a data center, uh, pr-provider or somebody who's got some experience here, a Google, a Microsoft, a AWS.
On the Northern side, you know, we need capital to acquire companies. We think of that stack as, um, acquisition debt that we raise, so that kinda runs through the ESOP. That comes from Balto as its investment in us. Uh, and then we see some opportunity for, um, seller financing and rollover equity, which would be very common in these smaller, uh, deals.
And in terms of numbers, I think it scales basically proportionally to our ambition. So right now, we are, uh, lucky to be in a position where we've checked every box on the capital stack. But as we get out there and start to be more acquisitive, let's say after the first two to three acquisitions, we'll need to continue to raise at these levels.
And, and one thing that, um, is really exciting to me is that there's a [01:04:00] major interest in capital flowing into employee ownerships, uh, right now. Uh, and, uh, we are part of a employee ownership capital lab accelerator, which is kind of interesting. The goal of that effort is to raise one point five billion to help bring employee ownership as part of the succession plan for Silver Tsunami.
We think we can do that for residential solar
Jason Jacobs: Uh, so in 10 years, if you have knocked the cover off the ball and, um, you know, over-executed in every way, um, uh, what, what, what does, um, what does the combined entity look like at that time and, um, and what have you achieved?
Andrew Krause: Well, um, we have as a floor the goal of a billion dollars in equity value, and this could grow in a very large market. So I mean, let's say we've really hit a home, or it's a $10 billion company. There's been lots of money generated for employee [01:05:00] owners. We have transformed the unit economics of your traditional solar, uh, business, you know, from that 8%, eat what you kill kind of business to that 20% ongoing revenue type of business.
Um, we've transformed some large number of homes. I mean, there's 120 million-ish single-family homes that this could be, uh, uh, brought to in time. In California, uh, we think is the right starting point. We can help people reduce their exposure to volatile utilities. And in the process, you know, it, it starts with the solar and battery trade because this is where the cheap power comes from, but it can pretty easily scale adjacently to, we think, the HVAC and plumbing trades as well.
And so that's how I would see us, um, uh, you know, in, in 10 years, I think sort of sky's the limit. Um, one, one thing about our posture is we are sort of building lots of structure and, um, kind of, uh, not trying to accelerate time artificially with [01:06:00] capital. We're building a lot at the pace of learning and the pace of doing in a new space.
So we're trying to build a 100-year company here, uh, with very deep kind of incorruptible values to it. And so that, that's super important to us.
Jason Jacobs: Awesome. Well, um, well gosh, thanks guys for entertaining all my, uh, all my newbie questions. Um, uh, is there anything I didn't ask that you wish I did, or any parting words for listeners? And I'm asking that to, uh, to each of you guys separately Carl, why don't you, why don't you go,
Andrew Krause: let's Kara first
Carl Lenox: Yeah. Um, so I think like one, one area we didn't, we didn't really, uh, get into, which I think is a, a good provocative question is, uh, why, you know, if you're a-- Actually, I think there was a commenter on one of your LinkedIn threads who sort of poses like, say if you're, if you're, uh, uh, you know, running your, your business independently and, you know, why not just give everybody Claude and, you know, have them go at it and will you, you know, why, why would you not [01:07:00] achieve the same outcomes?
Um, so I don't know. That could be an interesting question to talk about. Um, so yeah, I think it, I think it comes back to like... You know, there's actually kind of an interesting article that I, that I read recently by, by a guy named Marty Cagan, who's a leader at the Silicon Valley Product Group, and he's like a long-standing sort of product, uh, management thinker.
I think October of, uh, uh, I'm not sure what the date it was, but anyway, he did a rel-relatively recent ar-article where he sort of talked about, um, what product management looks like in an increasingly AI world, and was sort of-- he's sort of been, been, been thinking out loud, thinking in public around this, right? And one of the, the, his, his insights, which I think is very, is very true, is that most people just in general, um, don't think about prob- solving their problems in the same way as you would think about if you were coming in from sort of a, a product mindset perspective. Meaning that I can pick up an, an AI tool off the shelf, and I can use that to [01:08:00] solve the problem that I have today or maybe the, this very specific thing that I have to deal with.
But most people aren't, even if they go through the effort of doing that, um, don't have the experience, um, or the, even the time sometimes to sort of, to think, what, what is the more general version of this problem, not just for me, but for the people around me in my company and maybe in people in companies adjacent to mine, right?
And what are the broader implications of the, of, of implementing this tool? And this kind of gets back to the data governance and kind of the guardrails and, uh, everything else. Um, and are, are there, are there, is there a set of adjacent, you know, jobs to be done that I can sort of expand this thing out into? Um, so those things are probably not gonna happen if you just sort of give capabilities to people and then say, "Let her rip," right? So this is where I think, you know, kind of coming in and taking a step back and sort of saying, "Look, let's, [01:09:00] let's understand your business. Let's help you get a lens on your business and think through," in kind of a train the trainer's mentality of how we can apply these tools most effectively.
Um, so it's not just helping you and this one little thing that you're doing, not little thing, but this one thing that you're doing. Um, but also that we're gonna bring those learnings back in. And the way that I look at that when I'm working with, with folks is it's really cool, right? There's like this sort of...
You, you kind of come into the company and it's sort of chaos mode. Everybody's doing this stuff, and some people are really leaning into it. Um, and those folks, like what are, what are they doing, right? They're creating the, they're creating... It's like, "Ooh, I created this tool. I used Quadric in this artifact." It's like that is an amazing expression of intent. I see where you're going with that. You've built a prototype. You've built proof of concept, and okay, let's play with that. Let's see where that intent, you know, if we were to sort of like build out, push out the boundaries of that, uh, and think about, well, how that's gonna generalize, and think about how we could sort of take that and sort of use that and [01:10:00] template it and use it and roll it out more broadly, what might that look like, right?
And so that, that's where I think it's, it's not so easy as just saying, "Hey, like throw out the tools and let her rip." I think that's gonna create, you know, kind of a... It could create kind of a dumpster fire. Um, it, it also, it also is not, not gonna be nearly as effective, um, as, as, as it could be if it, if that, if you're doing that transformative work inside of a framework that is deliberate around how transformation happens in a firm.
Jason Jacobs: Yeah, I think that's a great point. And, um, what about you, Andrew? Is there anything I didn't ask that you wish I did, or any parting words for listeners?
Andrew Krause: I am super passionate about the people who have built companies in this industry for five, 10, 15 years of their life and are probably struggling right now to figure out what comes next. We know what worked for us in the past isn't gonna be the thing. There's a major, you know, [01:11:00] set of psychological shifts that I myself have gone through and I see other people going through.
I think being receptive to conversations around how to do better as a group and, and, um, being open to some of the things that Carl's mentioned and just basically holding ourselves to a higher standard, uh, and accountability, which r- which requires some vulnerability to, to do this stuff. Uh, but I think that, um, people in the solar industry are, are well prepared to do it and I, I really would love to see our industry meet the moment and thrive.
We're gonna be a small part of that, but, uh, just for everybody listening in the industry, I think getting back to basics on why we started doing this in the first place and what it might take to, to transform ourselves, uh, and our customers is, is kind of the mood over here. And, um, yeah, I appreciate you, Jason, on, on, on this podcast and being part of your journey and look forward to checking in with you as we get more and more, uh, results [01:12:00] over here, good, bad, or otherwise.
So
Jason Jacobs: Well g- well guys, thank you both so much for coming on the show. I f- now feel like I at least have a, a meaningfully better understanding of, uh, of what you're up to with, um, with Balto Energy and with, um, Northern Pacific Power. And, uh, yeah, I'll be fascinated to see how it plays out, and, um, and rooting for you guys, so thanks again.
Andrew Krause: Take care, Jason. Bye
Jason Jacobs: That's it for this episode of BuiltForward. I hope you enjoyed it. If you found it useful, share it with someone building in the industry and follow the show so you don't miss what's next. Thanks for listening, and see you next week