27 - Building business and tech with Jordan Appleson (Founder & CEO @ Hark)

[MUSIC PLAYING]

Hello, Jordan.

Thanks for coming on the podcast.

Thank you so much for having me. It's a lovely space I'm in.

Well, it's a bit echoey right now. So if you're listening to--

anyone listening to this and you hear the background noises or echo, we are getting the whole office revamped with a studio space and everything. And that means that we're out in the middle of the place. So yeah, apologies if there's echo. And Jordan, thanks--

hopefully, you don't have an echoey voice.

Well, we'll find out.

We're going to find out. And you came down by car from Leeds today?

Today? Yeah, yeah. We're based in the north. We're based in Leeds City center, been there since 2016.

Cool. And today, we're going to talk about your company Hark.

Yeah.

And I'm really excited to do this conversation because what you guys do is--

it's a bit like DSR, but it's more than that, right?

Yeah.

And I'll ask you about that in a second. But just before you talk about the company, who are you? What were you doing before this?

So my name is Jordan Appleson. I'm the founder and CEO of Hark. We're an energy analytics and industrial IoT company. And before Hark, I was a software engineer, been designing and building software since I was about 10, 12, all the way from games to PCB designs for lights and control systems.

And I kind of dropped out of school when I was about 17 and went to work in the industry as a software engineer, high-performance data applications, that kind of stuff, mainly making a lot of people money on things like ads and insurance, and that kind of stuff, so a little bit soul-destroying. But it got me into the industry.

And then a few years into that, I suppose, into that business it sold. The two gentlemen, they sold the business. I said I was leaving because I wanted to go and do what I'm doing today. I have a passion for technology and physical connectivity of things. And they kind of said, look, we'll write you a check to kind of fund your first year salary, and kind of away we go.

But what I've been doing for the last kind of several years is basically in my bedroom writing code, trying not to electrocute myself with things like energy meters or inverters or whatever.

And so you're a software engineer by background. But you're now running Hark, which is your company. And you've got a couple of other co-founders, right?

Yeah, absolutely. So when I was working at Branded3, which was the digital agency that Vin and Patrick hired me back when I was kind of 17, I met Jules and Andrew there, basically. And we ended up working together there on all sorts of really, really cool stuff. Then when I said, we had this idea for Hark, we'd always been talking about doing something together, they were like, yeah, let's do it.

So your previous employer that sold wrote you a check seven years ago?

Yeah.

You started Hark, three of you in various bedrooms writing code--

And building hardware.

--and building hardware?

Yeah.

And so the company, the company is a technology company that does what?

So we basically have a platform that allows companies, factories, retailers, to connect to any industrial asset, so it could be an energy meter, energy storage system, commercial building, [? VMS. ?]

We got wind farms on the platform, solar inverters.

And what we're able to do is monitor those in real time. And with this control capability, we can basically optimize those assets and send control signals in real time back down to them to do things. Typically, we're kind of improving efficiency of those assets, maximizing the output, or reducing waste.

And so compared to--

I'm not trying to put you in a box. But I'm trying to--

JORDAN APPLESON: Put me in a box.

--get grounded somewhere, right? So, how is this different from an aggregator?

JORDAN APPLESON: Yeah. So fundamentally, we are our technology company. We do not interact ourselves with the market, or anything like that. What we do is, with customers who are looking to control their assets--

for instance, Sainsbury's are one of our biggest customers. They wanted to optimize their buildings, their lights, their refrigeration systems, areas of their stores using their existing assets without having to replace them.

So they took our technology, which works with all of those 20-year-old assets, things like modulus [? Buckner ?]

[? Ethan ?]

IP, and QTT, all built into the platform. And they have--

Well, that was a word salad there. So these are communications protocols for old stuff in supermarkets?

JORDAN APPLESON: And new stuff. And not just supermarkets, it could be a factory.

It could be a Tesla battery, for instance, the power packs.

These protocols are kind of ubiquitous and have been around since kind of the '80s. And fundamentally, there's not been a huge change in the way those protocols work over the years because these assets are typically designed to last a long time.

So we have, basically, turned our platform into something that plug and play with them, and then you can do stuff, whether it's from the cloud or what we call the edge, and the edge being close to the asset from a compute perspective.

So if you walk into any Sainsbury's store today, when the energy prices change from, for instance, a [? trade ?]

warning, the price event, whatever they want to call it, within 60 seconds of a warning, our system will orchestrate a power profile change. So it's not necessarily DSR in terms of the market. It's for them to kind of reduce their energy supply.

So it's like--

it is kind of DSR, but they're in control?

Yes, exactly. And they can then play in other systems to this right. So if you imagine, you have platforms that will do this for you. But if you're the end customer with these assets on site, you kind of have it--

it's kind of a black box. And if you want to do things as well, if you want to control your assets, change set points, or do things that are slightly outside of that energy space but related, then you can use our platform to do that.

And so let's get idea of scale. I'm going to come back to the technical stuff in a minute. So Hark, how big is the company?

We're about 20, 25 people. We've kind of organically grown. So we're not really backed by any VC, big VC money in that sense.

So you're customer funded from day one?

JORDAN APPLESON: Pretty much, yeah, yeah. And that's been a really interesting journey.

It's kind of unusual now. Some people call it bootstrap businesses or whatever. It's kind of unusual for companies to go that way.

JORDAN APPLESON: Yeah.

You must have been tempted to take growth capital, right?

JORDAN APPLESON: 100%. I think in the early days, you see everyone getting funded.

You see news about, this IoT startup or this energy analytics company getting funding. You're like, wow, those check sizes look massive. And then you think they're going to overtake you. But actually, the complexity of what this is and what we do, it was really tempting.

But actually, what we found out now is we've managed to build a significant--

I suppose a significant platform in terms of IP. But also, we've grown revenue. And we've proven to external parties that not only can we run the business, grow it, maintain customers on multi-year deals, we've also been able to kind of, I suppose, looking at the market, we can actually, maybe, take our pick when we want to do growth funding.

We're not like in that kind of runway mode of chasing the raise, I suppose, that you kind of hear sometimes, especially in the current market in economics.

Well, I was going to say because you guys are doing something very close to aggregators, but not. So with my old Kiwi [? Powerhouse-- ?]

there is Kiwi. There is [? OpenEnergy. ?]

There is Limejump. There is all these other companies. And terrific companies that had to do the market access thing and the technology thing.

And there was a huge variance in the capability on the technology side across those businesses. Some were fantastic, some not so. But what happened was there was growth funding left, right, and center for these businesses, and there was an aggregator boom in the 2010s, right?

JORDAN APPLESON: Yeah.

And that was just when you guys were around and you were doing your thing. And so you chose not to take part in that. You said we're not going to do the market access thing. We're just going to do the technology layer. And I'm sure you had a strategy behind it. You're still here, and it seems to be going well.

Well, when that was going on, I was still in school. So Hark didn't exist.

QUENTIN SCRIMSHIRE: OK, right.

So it's 2016 when we set the business up. And I look back on it now, and I'm kind of glad because I've read a lot about it. You see Limejump. You see Kiwi Power. You see Renko as well. And what they've managed to achieve and what they did back then, it made so much sense.

But if we'd have got into that kind of area at that point in time, I don't think we would be where we are today and definitely wouldn't have the knowledge as well that we have today around all these different assets because it's not just energy storage assets, not just solar. We've got building management systems on the platform. We've got that hooked into occupancy centers that do all sorts of control that are slightly outside of the energy piece, but also related right.

We did our first battery in 2018. It was one of the Tesla powerpacks. It was like a 270 kVA in a 340 kilowatt-hour battery. And that was quite interesting because that was at the time when even still people struggled to connect and really orchestrate control of these assets for--

Even now, right, controlling batteries is still one of the hardest things to do. And you mentioned Renko. Renko has got--

I don't know how many people--

50, 60, 70 people. And it takes that many people to get it right, which is insane.

It's because of the complexity and the amount of moving parts. If you think about what goes into controlling any asset or connected to a building, you need to understand the protocols, understand the asset, you need the gateway device, you need the software, you need the connectivity piece, which is 4G--

how do you get network there? And they need all the back end systems to monitor all of that, do the stuff, bean count. You need so much expertise all the way from electrical.

Plus, you need to be ahead of the game in a market side, which is--

that can be half your stuff could be on the market side.

And that's why we don't do that. We don't deal with the market. We are a pure technology company.

So you mentioned Sainsbury's. That's a great customer. Congrats.

And we can all relate to that.

Sometimes we walk into Sainsbury's, and you've got empty freezers that are still running. So, oh, what's going on here? I think it's going to become a lot more important with--

especially power prices now in this winter, everybody's going to be thinking about it.

Yeah, even just the automated lighting control solution that's in place, that's running--

our software runs--

it saves them seven digits year, fully automatic. No one touches it.

Just the supermarket lights.

JORDAN APPLESON: Just a bit on the lighting, yeah.

So what's the deal there? Surely, the shops are open, lights are on. Shops close, lights are off, no?

JORDAN APPLESON: Yeah. Well, the thing about--

well--

A bit facetious, really.

JORDAN APPLESON: Well, no, typically, that's how it was. But if you think about it, there are different, I suppose, types of layout for store. But also some of those stores are 24 hours. The lights don't need to be on full for 24 hours.

There's also [? look ?]

[? sensors ?]

and how bright is outside, and whether the cheese counter should be on 3:00 PM on a Sunday because they don't sell as much cheese.

It should absolutely should be on. The cheese counter should be open 24/7.

JORDAN APPLESON: I love cheese. But whoever--

You like them.

JORDAN APPLESON: --is planning that at the retailers--

I mean, I don't deal with the planning. But the other side of it is, believe it or not, a lot of people had manual overrides and stores to turn things on and leave them there. And there was no visibility, no way of automatically controlling and sending that signals back down.

For us, what we demonstrated originally was existing system, zone-level control, all the way from lux levels to percentage of how each individual zone in the store. And when someone hit a button to say override that to 100% and left it on for two months, 100%, our system within three seconds of detecting that would switch it back.

So if I walk into Sainsbury's tomorrow--

probably not Sainsbury's man anymore, I must say. They are too expensive, got an ALDI at the end of the road.

JORDAN APPLESON: I make no comment about where I shop.

If I did walk into Sainsbury's because ALDI was closed--

so I'd walk in, and the site would be controlled by Hark technologies.

So there is a gateway on site. And it's not just connected to the lighting. It's connected to anywhere from 20 to 30 individual energy submeters.

We stream minute-by-minute data of instantaneous power of all the assets. We also have minute-by-minute information around the lighting systems, all the zones. And there are other assets as well. If you go into the depots, or the refrigeration systems, ammonia panels, fire alarms, all our software, one gateway, connected to all of this.

This is awesome. And what I like about this is it--

coming back to the aggregator thing, we all got hooked on--

have you got a 1 or 2 megawatt generator? Or have you got--

basically, are you big enough to get you into a bucket for store is the old game?

Can we get you and some others up to 10 mega--

well, it became 1--

but 10 [? meg. ?]

So we all got addicted to sort of like if you're not half a megawatt or bigger, you're too small for us. We're already going after the low-hanging fruit, if you like. And what's nice is that you guys have come along and said, well, we'll take the small stuff too.

And when you actually--

when I look at the platform and I can see instantaneous right now at 1 minute, what each type of asset class across the whole of the UK is? It's tens of megawatts. Now obviously, you can't just shift all of that. But I'm hoping with the way the energy transition is going that we will be able to eventually utilize these more in, maybe, the market somehow.

To do that, the technology is there, regulation and how that actually is orchestrated. Because technically speaking, if you wanted to shift 10% of that within 60 seconds, it's doable.

What about the--

so you guys save money, right, for other companies? Are you also measuring the carbon saving? Are you doing anything on that?

Yeah. So it's like the biggest requested feature of our energy product. So we have three core products, three core areas we work in--

the connectivity side, the analytics side in real time and modeling assets, and then our energy tools product, which is the bit that does all the half-hourly analysis, time-of-day profile analysis.

It does integrations with billing providers as well if you're rebilling stuff, which is quite interesting. And that is the most requested feature around carbon emissions. But a lot of people are looking at Scope 3. And Scope 3 isn't just what you do. It's about your whole supply chain.

So what's Scope 3.

JORDAN APPLESON: So Scope 3 is basically anything in your supply chain that essentially is contributing to your mission. So that could be your suppliers, how much they use, how much energy they use, the--

Where does the word "Scope 3" come from? Who decided this? Is this government?

Yeah. So it's like the--

it's basically the standards around carbon emissions. You know I wish Angus, my data scientist, was here who works on all this. But there's things like ISO 50001 certifications and that.

But those are defined by government bodies. And they're used in regulations to, then, track and that's what we're seeing.

So companies in fact--

I'm not really about to say because again we need like a gong for every time we say it.

JORDAN APPLESON: Well, we can say it like "faster."

Faster, yeah.

JORDAN APPLESON: OK.

They're under pressure to do this Scope 3 thing, and that is to consider all these additional carbon emissions.

The thing about Hark is, like, we're not like a sustainability, like, emissions-tracking company necessarily. But even getting Scope 1 and 2 from an energy perspective--

and that's not just electricity rates, gas, water, everything else, that's difficult enough right. And then you have to do that in Scope 3.

So we are seeing factories--

we work with a massive chemical plant. They're doing all of their factories, connecting all their distribution boards, gas, water, electricity. Every minute that it comes to our platform. Their sustainability and net zero team are using that to figure out, how do we get rid of gas?

How much is it costing us? How much in terms of emissions as well?

And they will take that data. And they can hit a button. Because previously, it would take months to figure out, how do we get this data? The utilities, their data feeds are not great.

QUENTIN SCRIMSHIRE: Well, I was going to ask you about that. So if you guys are--

to say you've got a big industrial plant that's making a little lizard things like this. Actually, that's a--

JORDAN APPLESON: Well, I mean, that would be--

--Komodo dragon, actually.

JORDAN APPLESON: That is a [INAUDIBLE]..

It's not a lizard. I'm going to get crucified for this.

JORDAN APPLESON: You should call it "Modo" dragon.

That's what it's supposed to be. It's supposed to be a "Modo" dragon. I've got it wrong.

Anyway, so we've got a factory that makes these things, and it uses tons of energy because it's really high end. And so--

Believable for these.

QUENTIN SCRIMSHIRE: Yeah.

But that factory has signed a contract with, let's say, EDF or [? ENGIE ?]

or whatever to supply them power.

And EDF or [? ENGIE ?]

or whoever the supplier is has had to buy that power in advance. And then you guys come along against a curve. They've been promised a curve by the industrial process, where we said Monday to Friday, we're going to make these Komodo dragons.

And then Saturday and Sunday, we're going to stop. And overnight, we're not going to do anything. And then you guys--

so they've got this shape of power that they've bought. And then you guys come along and do a deal with the Komodo dragon factory.

And you say, oh, we can save you money here. You can save your money here.

And suddenly the shape, that EDF has been promised on this, like, five-year deal. You guys messed around with it. And they're exposed to all this imbalance costs. So explain that, Jordan.

I'm sorry. I'm sorry. Look, this is why we don't interact with the markets.

I know I'm losing like as an actual kind of domestic customer. I see my bills going up. But I think fundamentally, we know in the industry that something has to change.

It isn't sustainable, fundamentally. Yes, the market dynamics, they work the way they do.

And it was something--

I remember in 2018 when we were doing the Triad avoidance stuff. And I was thinking to myself, who's going to pay for this? And now we know who.

But it's not--

I mean, just to be clear. Disclaimer, we are not the reason that people's bills are going up, just to be clear if anyone's listening to this and thinking Ofgem are going to increase the--

They'll be protesting outside Hark HQ in Leeds.

JORDAN APPLESON: We're all remote anyway.

[INAUDIBLE].

JORDAN APPLESON: So good.

Super glued to the front of your thing.

But we know that then has to be a shift, right?

We're not the only ones doing it, and for good reason.

These businesses are being charged a huge amount of money for electricity, gas. And they do not have the cap. And it's fine when you think about big company. But we work with--

JORDAN APPLESON: --small companies. Yeah, yeah. Yeah, exactly, the price cap.

OK, fine, you take the big players.

Let's Gong Sainsbury's. Let's talk about the--

I think that we've got a sawmill who have seen their electricity prices go from 300,000 to something like a million. And that's a family-owned business.

What are they going to do? They're not thinking, well, if we shift this, if we reduced energy and we were in contracts or whatever--

I know the contracts are slightly different for smaller businesses, what they're going to do?

I'm quite anti-pricecap, though, I've got to say. I think the price cap--

so we're going off the topic here.

Cool.

So the price cap covers--

it's there to protect vulnerable people, i.e. People who can't afford their electricity bills and gas bills, which, I think, as a society, we should try and protect those people.

JORDAN APPLESON: Yeah, of course.

But the government rushing to step in and do this price cap methodology, which basically pushes all of it onto suppliers and essentially puts it onto--

yeah, it sort of democratizes the cost, but also suppliers have got the headache of doing it. I just think it's crazy that the government is now stepping in and setting prices of a commodity that is set by a global energy market. But yeah, I can see why businesses aren't included. But I mean, imagine the headache if they were.

Oh, yeah. So I'm 100% not advocating a price cap for businesses. I think it--

yeah, I mean, I think that would be the end of the world as we know it.

QUENTIN SCRIMSHIRE: I think we should--

firstly, I think we shouldn't have a price cap. But what we should be doing is stepping in to support businesses and people who are vulnerable to pay their bills. I think we should do that, rather than trying to put this weird thing in place.

Yeah.

I don't disagree. I think fundamentally that the reason we do what we do--

and it isn't--

just to be clear, it's not just energy that we save. We optimize assets reduce downtime. So there's lots of other things that we do with companies. And that's why we're purely a technology play.

The price cap piece is interesting because you kind of shield it from it. Like as a consumer or as a domestic consumer of electricity and gas, you have been shielded from it for so long, and then it's a massive jump. And I don't even know if it's a zero-sum game. But we are where we are because of global issues.

And something's got to give. And this is where we are today.

And unfortunately, it's a cycle as well because if you think about some of the--

the Komodo factory is UK based, where prices are going up. As prices go up even further, people look to other countries that, maybe, have a slightly different way of--

from an economic perspective--

Or a different carbon-accounting regime.

JORDAN APPLESON: Exactly. Exactly.

Business go under because people choose to go for the cheaper option. So it's a vicious cycle. I don't have the answer to it. I wish I did.

Well, you do have some of the answer--

JORDAN APPLESON: Some of it.

--which is like, don't run your fridges when they're empty, or don't run your fridges overnight when you--

Exactly. Or some really cool stuff.

QUENTIN SCRIMSHIRE: That's fridges.

Well, it might be fridge.

QUENTIN SCRIMSHIRE: The latest stuff, right?

Well, in the industrial depots, you're basically overchilled during the cheaper periods, the rack periods.

QUENTIN SCRIMSHIRE: The rack periods.

Red, amber, green--

so do us.

And it sounds like you're ragging your depot.

They're not. But they might overchill. And they'll set the set points a lot lower, so during the more expensive hours, they can turn everything--

JORDAN APPLESON: Off.

So depots that are chilling foods are overchilling food and exact that upfront.

JORDAN APPLESON: They can do.

They can do--

JORDAN APPLESON: It depends.

--to avoid the red rates on the [INAUDIBLE]

[? charges. ?]

Very smart. That is the sort of thing they should be doing.

I guess that's what the red, amber, green was

JORDAN APPLESON: That's why--

exactly.

--to encourage you not to take power at that time.

And now we're kind of going into this area of--

in logistics depots, where everyone wants to put in, like, electric charge points and also like the electrification of the fleets. Oh, my goodness. You can imagine what's going on there.

So question for you, as Jordan of Hark, my concern is--

our business, Modo Energy, is--

we are so tempted to do so many different things. We've got this core product that does this thing. And then we could do anything. We could do long-term forecasting.

We could do some international exposure. We could do EV charging stuff.

It's so tempting to do so much. But we've somehow managed to get this far by being laser-focused on one thing.

JORDAN APPLESON: Exactly, yeah.

My question to you is, are you not even more tempted to do that stuff because there are so many places you can go to? So you talked about about uptime. You talked about carbon removal. You talked about electricity reduction.

You talked about different assets that you can control or monitor. You can do dashboards. You can do integrations. You can do--

JORDAN APPLESON: Every single day, I wake up and I want to do more. But we're a small team, and we have to--

you have to ask yourself, what do you want to be the in the world out? If Hark's going to be the best in the world at one thing, what is that? And for us, it's basically the asset connectivity and giving our customers the ability to control any asset, plug and play.

We have energy tools. We have analytical capability because that's born out of necessity.

You need those point solutions. But that's typically what companies and people are coming to us today because if you can't see it, how are you going to fix it?

And what about B2C? You guys talk about doing B2C, the idea you buy a washing machine. It becomes the QR code. You scan--

I buy washing machine. I scan the QR code. It takes me to Hark. I put my little details in.

And it's already got a bit. And you guys control it from the word go.

Don't get me wrong. Again, I wake up every day and think that's something. But to your point, you've got to be laser-focused on that one thing. And I suppose the consumer market is very different.

It's a very different ballgame because if you think about it, you would go up against someone like Amazon, who have--

if you've seen they just bought the robot Hoover company or whatever.

They've got Alexa. They've got smart controls. You've got Philips. You've got all of these other companies that are doing in that smart home space.

And yeah, while we have the technology, technically do it, how do you get into that market? How do you make it a success? And I am probably the worst consumer when it comes to tech because I'll tweet people, like, it's not working. And I don't want to be on the receiving end of me as a consumer, basically.

Yeah, so I did say that. I don't know. I've just got--

I really like B2C businesses in general. But I just know that I would never, ever be able to be involved in one because so much of it is marketing, so much of it is customer experience.

You have to get some I know influencers on Instagram. It's all well beyond my pay grade. I don't understand it. But I am very impressed by people who managed to get technology in people's homes.

The amount of funding required--

and don't get me wrong.

The money is out there. But the ability to do that--

I mean, it's a long game. The market's quite crowded. But it is important because if you think about domestic and homes, and how much energy is wasted, or what you can do to optimize, there's a huge opportunity.

QUENTIN SCRIMSHIRE: There's a massive opportunity.

But Hark are not going anywhere near that. I can tell you.

QUENTIN SCRIMSHIRE: No, no, no. So you guys are B2B.

Let's talk a bit more about your customers. So we talked about Sainsbury's, Gong. What other customer types do you have?

So I'll give you example. We work with a chemical plant or chemical plants factories. We've got restaurants as well that are all looking to do things.

I suppose the chemical plants are my favorite because I get to see really high-voltage stuff. For them, what we're doing is we're going into that--

they've got substations on site--

2 square miles of plant--

and they're looking to try and optimize energy, utilities, figure out how they can optimize their processes.

So with those customers, we're going in there, installing really interesting equipment. It's quite cool to see one site use 40 megawatts of power in live on the platform when you're thinking this is one place. And you see that--

What is that? Is it steelworks or--

JORDAN APPLESON: Unfortunately, I'm not allowed to say due to what they do. But it's--

Oh, yes.

JORDAN APPLESON: Yeah. But it's--

So anyone who listens to this, we're going to have to kill them--

JORDAN APPLESON: Well, if I--

--or you.

JORDAN APPLESON: I think I would get murdered.

You're getting murdered.

JORDAN APPLESON: I'm joking.

Someone's going to pay the price. We won't ask any more.

It's all good. But those kind of things that are really interesting because you actually--

it goes back to all those moving parts and how they're doing it because it's one thing for supermarket. But for factories, where security and cybersecurity is so important, how do you deploy that?

How do you take what is quite invasive technology--

if you think about what you can do and what data you can get out of it. And security is the point where it can't be hacked.

So, what is it? Is it a box--

JORDAN APPLESON: Yeah, so we--

A box you plug in somewhere?

So HARK Connect runs on any industrial gateway, anything on Linux. And typically, we use industrialiazed gateways from--

it could be Advantech. It could be Dell, we're completely hardware agnostic right now.

QUENTIN SCRIMSHIRE: A gateway being--

It's basically--

QUENTIN SCRIMSHIRE: --a computer-looking box.

It's a computer-looking box. We made the decision. I said we made hardware in the early days. We did do that.

It's bloody expensive. And 3D printing--

I remember in the early days, we were 3D printing enclosures for this Raspberry Pi-based compute module thing that we had built and we were shipping to customers.

And it was expensive. It kept me up at night, even though they all still work.

Some have been in five or six years. They all still work because of the way we designed them. But the amount of money you need in tooling to do all of that--

so we just said, OK, let's just make it work with anything.

So you want to run our stuff on a Raspberry Pi? Fine. You run it on Raspberry Pi. We don't advocate that.

We typically use industrial gateways from providers of industrial compute for--

we've got a big property company.

They've installed all our kit on a server in their server room that's connected to all our assets. We don't mind. And why?

Because we can, then, democratize. It's not hardware that people need to worry about. You take our software. You put it on whatever gateway you want. As long as it's got the right I/O, we can connect to those assets.

QUENTIN SCRIMSHIRE: Cool. And so want to come to forward-looking stuff--

so you have been going seven years, right?

About six now.

QUENTIN SCRIMSHIRE: Six years.

Five full time. Five full time. I don't count the first year because it was just me and Andrew in a room trying to figure out what we were doing with our lives.

We should probably discount our first year.

JORDAN APPLESON: Yeah, exactly.

Exactly the same as that. So yeah, we've gone through the customers and what the problem is. What about the future? Where is this whole thing headed?

Is it smaller assets? Is it more complicated assets? Is it just more of them?

Yeah. No, it's a great question. So fundamentally, there is a massive problem with edge control. So now we see you've got solar.

You've got EV. You've got storage. Then you've got your actual load of your building on site. How do you optimize that because you can't do it from the cloud?

How are you going to make--

QUENTIN SCRIMSHIRE: What can't we do from the cloud?

Well, it's not that you can't do it from the cloud in the sense that you can't compute that stuff in the cloud. It's more that you need local control and very fast changes to the way things are like what assets are doing. It could be set points. It could be turning off EV charge points based on what load or what power limits you want to put in place.

It could be based on wind or solar or the sun and how--

does a cloud go over? What does that mean?

So there's a big challenge now of how do you orchestrate all of that to optimize power.

How do you optimize all your generation on site? So where Hark is going to go is bringing more of that to the edge. We do a lot of edge control already because we have to. But, how do you make all of that intelligent stuff happen on the edge?

QUENTIN SCRIMSHIRE: What does "on the edge" mean?

"On the edge" literally means it's close to those assets as possible. So that small box is the edge.

QUENTIN SCRIMSHIRE: It's doing the thing.

It's edge computing.

QUENTIN SCRIMSHIRE: So this is nuts because if you think about--

so if you think about PlayStation, when I was growing up, I had a Nintendo 64 mega drive or whatever. And then this new one would come out every year. And we're now on PlayStation 5.

And every year, it gets better and better and better. And then, out of nowhere five years ago, Google said, we're going to launch this Google Stadia thing.

And everyone said, no, it's not going to work in the cloud. The latency is too long, blah, blah, blah. And here you go. You've got people now streaming games over the internet.

So that whole world--

I've got off a bit of a tangent now--

it's all going towards cloud in a way that people thought it was impossible. So why can't the business power control world also do the same?

It's not that it can't. It absolutely can use the cloud. The cloud is a necessary part of what we do. But if you think about the sample rates that you need--

so fundamentally, we work with a company that sample something like 16,000 times a second on like 16 channels of sensors that deal with power. And they generate something like 10 Meg to 100 Meg of data per second.

And if you think about where these assets are and what kind of infrastructure is in place, you need a pretty beefy connection. And that infrastructure isn't there, fundamentally. So fine, that's one problem.

The other challenge of that is fundamentally, if you're sampling 50 times a second 100 times a second, you need to make really quick, fast changes to those set points or whatever. You don't want around trip going 8 milliseconds up to the cloud and then back to then work out what your set point control is. I mean, that might work for certain scenarios.

Yeah. But they said that about rumble packs on controller. That's right?

JORDAN APPLESON: Yeah, yeah. Yeah, yeah. Don't get me wrong. I definitely think there will be areas of control that will use the cloud, but I mean--

QUENTIN SCRIMSHIRE: It's not there yet.

--I think if you've got 4G or you've got latency problems--

I mean, what happens if the internet goes down? And what happens if, for instance--

I mean, this is a challenge. What happens if you are, in the market side of things, executing some kind of trade or something expected to happen with dispatch, and it doesn't happen?

OK, fine, batteries. But think about buildings and the way they're expected to run against maybe co-located storage or co-located generation. What happens if it all goes down even for 10 minutes?

Yeah. We've got some stuff to work through. I'll ask you about batteries now just before we finish. So you guys controlled the battery before. You do the Tesla system.

QUENTIN SCRIMSHIRE: Yeah.

Are you not tempted to get into the optimisation space?

I would love to. Like, again, I've got so many ideas. The thing about the optimisation space for us is, I think there are companies that do really, really well. And they've been at it a long time.

And I think if you're just doing batteries and you're just doing market and batteries, I think that market is--

I don't think it's like fully saturated. I think there's a lot of optimization still left to do.

I'm looking ahead and seeing--

looking at, OK, well, what if you do batteries with solar, with wind? Like I said, I think that's the area we're going to focus on because I think it's a much harder problem because you've got more--

well, you've got more inputs. You've got to factor in algorithms that take so much stuff into play.

Doing that and computing that takes time. Doing that in real time takes a lot of compute. It takes a lot of--

it's been solved yet. And I think that's where I like to kind of spend my time.

So you guys come into co-location soon behind the meter?

We will see. We will see. I think it will be with a partner, 100%. It'll be with a partner. We'll just be providing the technology that makes that happen.

Awesome. So that's it. We've come to the end of the time.

I just want to say thank you very much for coming on. We're going to put Hark's detail, your details, and also the company's details on the show and show notes below, so you can find it. And everyone listening, please--

I've been told we have to say this every time now--

please, do subscribe and like the link. And we will see you again soon. Thank you very much.