00:09:18 · Kevin Hettrich, CJ Gustafson
[00:09:18] S0: I'll give you a big picture answer and a small picture answer. Big picture. Is it a big market? Is the product compelling? And will it be when... If you kinda give credit to your competitors as well? If that's true, you're like, okay, I'm heading in the right direction. There's a big market. My customers are validating just how compelling it is. That's a good long term thing. From the short term nature, it's like, can I break this up into bites? A long term promise isn't very valuable if you don't exist to get it. So can I break up the value creation into steps? And do I have sufficient capital to kinda get to each one? And making sure that you're doing that conservatively and that the NPV kinda continues to solve at every point. One other thing I would highlight is that there's a lot of degrees of freedom. We have effectively every end market you can think of, and then there's a diversity of products and product requirements in there. So that that portfolio nature of this does give us lots of degrees of freedom, and perhaps we may talk about some of those choices along QuantumScape's journey. [00:10:14] S1: Early on at QuantumScape, you looked at one of the manufacturing approaches, and you realized that even if the tech worked, the economics were nowhere close to what the auto industry could support. How did you figure that out, and and and what did it change for the company?
00:10:28 · Kevin Hettrich
[00:10:28] S0: When I joined, I was coming out of the kinda science and engineering side from Stanford as well as the business side, and I myself was like, I I think I can be helpful in this context, but I didn't know. It was my high... I've become a private equity consulting in my first kind of operating role. So I did have a chip on my shoulder and did want to try to make a positive impact on things out of the gate. And this, what you're referring to, is I think one of the first indications like, oh, okay. I've got a role here. Just made one other contextual point. Our CTO, when I was in the process of being hired, I was in the room, their CEO was like, hey, why are we hiring somebody who doesn't have a PhD in material science? Isn't that the work to do? And our CEO, Jagdeep Singh, is like, no, Tim, go hunter Tim. Like, trust me, there are other things to do in addition to the science. So I got there, and we were using a... An r and d tool, which was vacuum based. And just think think of it, it kinda makes things slowly wonderful for r for r and d because you quickly change out the materials and quickly get data points. The speed of learning is a really important kinda metric that we we we kinda track. Some of the hardware team is like, I don't think this is gonna scale. Then our CEO doing the right thing is like, prove it to me. Like, let's not take anything for granted. Prove it to me, and until you prove it to me, we're making the devices, they're promising now, why shouldn't I just kinda keep keep doing that? Let me take this on. So what I did is I kinda talked to a bunch of people, did a bunch of research, and then a few weeks later kinda brought people back in the room. When I says, okay, let's think about this. What are the different components of cost of goods sold? It's things like depreciation and materials and labor and consumables like electricity and process gases, those are my columns. Along the x axis, I put at the bottom, what's the best we've ever done? And then the second line was, what is the best anyone in any application on the planet has ever done? And then the line above that is what we probably need to be successful in automotive. And I just put it out, and I was like, let's count the orders of magnitude. One, two, three, four, five, six. And then the room just exploded.
00:12:28 · Kevin Hettrich, CJ Gustafson
They're like, blah blah blah blah blah blah blah blah And then after that, I was like, okay, fine. I'll give you one back. We are off of five orders of magnitude. Is it impossible that we can innovate our way there? But seems kinda hard. And then I left, and then our CEO, Jagdeep, pulled me aside. He's like, listen, Kevin, that was incredibly valuable to identify at this point. We need to dedicate kinda bandwidth to kinda making our special component far more effectively than conventional means. So that's an example where I wasn't a content expert anywhere, but I was able to kind of create a structure, gather the evidence into kind of a logical format where we could kinda make decisions, and then be able to present it and to get people on the same page. So that was kinda my my teeth cutting moment at QuantumScape in in the first few months. [00:13:12] S1: That's an amazing story because it just shows that you were able to use comps of... It's it's all you're calling somebody's baby ugly. Right? You're saying the... This is what the market will bear. This is what average looks like. This is what good looks like. This is what world class looks like, and this is where we are. Okay? Take the emotion out of it. What do you think about this?
00:13:12 · CJ Gustafson, Kevin Hettrich
[00:13:12] S1: That's an amazing story because it just shows that you were able to use comps of... It's it's all you're calling somebody's baby ugly. Right? You're saying the... This is what the market will bear. This is what average looks like. This is what good looks like. This is what world class looks like, and this is where we are. Okay? Take the emotion out of it. What do you think about this? [00:13:31] S0: Yeah. And let's choose the lowest risk path. Do we want to try to incrementally improve on this method and close that type of gap? Or alternatively, do we wanna look for a different way of doing it that has development risk but might significantly derisk kind of running that equipment later? And we we end up choosing that that second path. [00:13:49] S1: Kevin, something you mentioned is speed of learning. Can that be a metric?
00:13:54 · Kevin Hettrich
[00:13:54] S0: Absolutely. Cycle time of learning, we we track it. What is the speed with which you have an idea until you get data back and kinda go into the next piece? It's effectively a measure of cycle time, and that's very valuable for two reasons. One is that you want that to be compressed, and then maybe the why is that you can make kind of micro adjustments as you go. Our chairman of our board tells a story about the Apollo missions, and when JFK is like, hey. We're gonna fly to the moon. Very bold idea. Couldn't even conceive about probably how to do it, but what did NASA do? They they chopped it up into a dozen plus steps. It's like, Apollo one, let's work on the rocket. Apollo two, let's show that we can get a capsule up. Apollo three, let's put an animal in the capsule and make sure. A great scientist or a great engineer will take a daunting problem, chop it up into small pieces, and as quickly as possible, try to get data on that piece, incorporate the learning, and readjust. And that's also part of the answer to question you asked earlier of, gosh. You were there early. If you think about it from, like, hey. Is... If the if the endpoint is, like, mass commercialization of electric vehicles and data centers, like, lithium metal batteries, yeah, it is kind of a long point. But if you chop it into, like, hey, does a material exist? Can you get to a component? Can you get to a single layer? Can you initiate product development and customer engagement? Then get into late stage, then do a vehicle demo, and then start a pilot line. Those are kind of more more kind of tractable things where you can clearly be judged on am I am I winning or kind of losing? Is everyone willing to continue on the journey? Customers, investors, our own employees, etcetera.
00:15:31 · CJ Gustafson, Kevin Hettrich
Ad[00:15:31] S1: You are what your gross margin says you are. Sorry. My kids are tired of hearing me yell that at the kitchen table. Dad, I'm two. I don't know what cloud costs are. Well, I do. And cloud infrastructure is the second largest line on most tech P and Ls. Ad[00:15:45] S0: And