#3 Building an Automated Pizza Preparation System

Transcript:

Adar: And today we have a very tasty episode, as I told Clayton, because our guest today is Clayton Wood, CEO of Picnic and Picnic automates making customizable pizzas. They have a very cool robotic system that does it. Hello, Clayton.

Clayton: Hey, good morning. How are you this morning?

Adar: Good morning, Clayton. So maybe you can tell us a little bit about Picnic. Was I right about describing your product?

Clayton: It’s definitely tasty. Yeah. We have guests in the office and the first thing they notice is the smell of pizzas in the air. And there’s usually plenty of free pizza around for anybody who’s hungry. So, we have a good time.

Adar: Sounds like a dreamy job, definitely.

Clayton: We’re having fun.

Adar: So tell me a little bit more about Picnic.

Clayton: So Picnic is a startup based in Seattle. We’re about six years old. The company was started by a group of automation engineers who were looking for a good use case for automation and they looked at food service. To clarify, I’m not the founder, but I joined about four years ago as the new permanent CEO. But when the company was started, they thought, well, let’s make food and then once you start thinking about food, you start thinking about pizza. And the initial idea was actually to make a fully automated food truck that had everything. You’d walk up to the side of it, push some buttons and the pizza would come out. And they thought, well, that’s pretty ambitious. Let’s start with how do you make a pizza? And that turned out to be pretty challenging, especially at that point. Several years ago, there were no other automated pizza systems to look at, there were no examples to follow. And we have some early videos that are pretty funny about how do you actually get sauce on a piece of dough and spread it around and get some cheese and pepperoni on it. So it’s been a journey, but we’re really happy today to be shipping to commercial customers and really on a growth track. So it’s been a fun journey.

Adar: That sounds like it. And when we prepared for the interview, you told me something like we made a working product and we didn’t realize that we were just beginning. So it looks like you had some massive challenges. How did they look like?

Clayton: One of the challenges of any startup is that you never have enough resources, you never have all the right expertise, all the right knowhow. So, you have enough to get started and in our case enough to get started with a bunch of talented engineers who are able to devise basically mostly an electromechanical system for how to handle simple food ingredients and manipulate them onto a pizza. But we had an ambition which is different than some other automation companies in the food space which was not just to make a pizza, not just to make any dish, but really to make any pizza. We wanted to make any pizza for any customer. And so we’ve always been faced with a variety of foods that you may want to put on a pizza. So when you think about, I always tell my team, you know, hardware is hard. That’s sort of a cliche. Automation is a hard space in hardware and food automation is a really hard type of automation because you’re taking a really highly variable input which is food. Even the same food is variable depending on the conditions and the age and other factors. And then you’re trying to make a very consistent output. So it’s super multidisciplinary, lots of integration of different kinds of skill sets. So, when we started we had electrical engineers and mechanical engineers and we were able to make a pizza. And when I joined we had different mechanisms that did the different things. One mechanism would put sauce on a pizza, one that would actually shred cheese from a block, sliced pepperoni from a stick. But they weren’t integrated into a system. It was just different mechanisms and we had about a month after I got there, we set up a customer appointment and we knew the customer would want to see a pizza. And so we had to do a quick integration and hustle and make a pizza. We did that and the customer was amazed. They were like oh my God, this is brilliant. I could think of five places we could use this. So we knew we were onto something. But at that point what we had was a tabletop prototype that could in fact make a pizza. But it wasn’t anything that anyone could use in a kitchen. It wasn’t anywhere near prime time. And when you’re making equipment for food application there’s a bunch of different considerations. So it’s got to be cleanable, it’s got to withstand the environment of a commercial kitchen. Ultimately, in the longer run it’s got to be manufacturable, it’s got to be cost effective, all those other things. And so the team that was really clever enough to figure out how to make/invent a pizza machine had to figure out how to do all these other tasks.

Adar: By the way, did you think about all of these challenges in advance? Or did you start by building something that worked first? Like, let’s just make something that makes a pizza and then deal with anything else, like manufacturability and cleaning and everything that involves this.

Clayton: It’s a little both. I think it’s easy to get confident in your team’s abilities when they show how talented they are as engineers and to think, well, if these guys can invent a pizza machine, they could figure out how to make it manufacturable. But it turns out there’s a lot of other skill sets involved, and there’s all these other disciplines. So when we started, our first system wasn’t refrigerated. It was just a piece of machinery on a tabletop. But if you want to make it where it holds food in a commercial kitchen to health codes, the food has to be kept at temperature so, you gotta build a refrigeration system and a different skill set, you know. Again, more mechanical engineering, different kinds of mechanical engineering, different skill set. And then when you’re trying to integrate that into a prototype, the first prototype, when we finish the first demo for the first customer, the next thing was, okay, well, now this is great. This works great. Now we just need one we can actually test in a kitchen and pilot it. So we said, okay, well, let’s just do that. At the time, it seemed like that shouldn’t be too hard. But again, one of the things we underestimated was even the skill set of going from, can you design a mechanism that produces an output to can you design an assembly that can be assembled, that can be built and built repeatedly, even on a prototype scale. So we assigned everyone jobs, and we ended up designing a prototype that was very, very difficult to build. And we were going to build four instances of it. And after two instances, the team just sort of put their hands up and said, we can’t build any more of this. This is too hard. We need to go back and redesign. So it’s easy to underestimate that step of how do you make something that’s actually buildable? Especially if you have a big, this isn’t a small device, this is a big thing. It’s about 7ft long, 5ft tall, you know, It’s a big thing. And just the fabrication methods, the material choices, how are you going to put it together? How will it actually work when it goes together? Lots and lots of considerations. And in a startup, you’re always dealing with a very small team. Limited budget, tight schedule, very very  challenging. 

Adar: So what would you do differently? Like, for the people who listen right now and they have these such considerations like, how would you treat that differently if you were to go back?

Clayton: You know in my career, I’m a mechanical engineer, and I started my career actually, I accidentally got into HVAC. Wasn’t my intent, but that’s where I ended up. That’s where a lot of careers end up. You do the job that you got, not the job that you thought you’re going to get. And I didn’t know anything about it. And my philosophy was one that I’ve frequently gone back to, which is to have a healthy respect for your own ignorance. Don’t assume that you know everything. Don’t be arrogant, don’t be afraid to ask for advice you know, really look at the problem in a really clear eyed way and recognize which skills do you have, which skills do you not have? And get help for any company, but especially for startups, there’s a lot of help available. Suppliers like to help, manufacturers like to help. There’s advisors who will work for a small amount of equity or even volunteer. But you’ve got to ask the questions, you’ve got to seek the advice, recognize all the dimensions of the problem you’re trying to solve and be honest with yourself about what you’re not actually equipped to do. And don’t try to do it all yourself. Get some help and figure out how to get it done. It’s all iterative. So, you’re trying to just get to the next milestone and then learn from that journey, and then you plan the next step of the journey. 

Adar: Such a great tip because especially in hardware, as you mentioned, there are so many different functions, no one knows everything and it’s better to address that and confess that you don’t know and ask for advice from the right people. Just together you can do amazing things like making a machine that makes pizzas. For me, it’s amazing.

Clayton: Exactly.

Adar: And what kind of skill set you mentioned there were a lot of disciplines like, what kind of skill sets you needed to work together and how did you make it work together? Do you have any tips for that?

Clayton: So our team, I think what usually happens is you end up either building the product that you can because of the skill set you have, or you really emphasize the main skill set you need. So we had a heavily mechanical engineering oriented team because most of what we’re doing is mechanical. We had to move food around, put it in a place, push it around, move the pizza down the line, a lot of mechanical motions, mechanisms. We had electrical engineering, but electrical engineering tended to be more straightforward; we’re powering motors and we had some simple controls. But, we had this phenomenon, there’s the old joke about “you don’t have to run faster than the bear, you just have to run faster than your buddy”. And so the software team didn’t have to be super software, they just had to be better than the hardware team. And that didn’t turn out to be too hard to do because the hardware team had all the hard tasks. So we didn’t heavily emphasize software and electrical in the early days. And then later on, I think as the mechanical systems became more mature and the bottleneck became software and electronics, we found that we really needed to beef up those skill sets. So, I think you have to kind of anticipate what are you going to need in the next phase? And it’s not always hiring. Like I say, you’re getting the skills or the capacity from whatever source you can. But recognize your needs will change from ideation to prototyping, from prototyping to piloting, from piloting to pre production, from pre production. You know, you need different skill sets, different disciplines will take the main stage in each part of the journey. And if you don’t anticipate that, you’ll sort of plow right in coming off of a big success at the pilot stage and think, okay, now we’re just going to turn this into production. And then you realize later than you should, that you don’t really have manufacturing engineering and you don’t have people who are going to do the documentation and the quality and all the various things that go into production.And I think, to me, that’s a headline. It’s easy to underestimate the level of work from pilot to production relative from ideation to pilot. That’s huge. And you’re creating something from nothing and it feels like an enormous accomplishment. But, there’s actually a lot more work to go from pilot to production than there is to go from ideation to pilot. 

Adar: Right, right. And always think about the next steps, the next things that you will need and anticipate that and know to bring the right people on board to help you overcome the new challenges that you’re going to face. So it doesn’t end in prototype, basically. That’s what you’re telling? 

Clayton: That’s right. I mean, part of the game in startups is you’re trying to survive, that’s rule one. Yeah, survive to fight another day, you’re going to make mistakes. And what you don’t want to do is waste time or waste resources making mistakes that you could avoid. And so if you want to go as smoothly as possible, gotta spend a lot of time anticipating what’s the next thing I’m going to need and what is the team capacity need to look like and I would draw the distinction between team capacity and skill set. It’s easy to go from, here’s my needs, here’s the skills I need, here’s the people I need to hire, and then this is my fixed team. And then when that changes, I need a different team. Well, it’s not practical to change out team members at every phase, nor the team won’t like it either. So you need to figure out who’s your core team, which expertise can you get from the outside to come in and make a contribution. And either do that as an outsider or a consultant or temporary, or in the fullness of time, you can bring those people on board full time, but you can’t keep changing your team out all the time, but your skills, the capacity that your team needs is going to change all the time. 

Adar: Right. You also mentioned Clayton, about something about the manufacturers helping you to succeed. What are your thoughts about that? Do you have any tips for the listeners all about working with manufacturers to get to production, to get to better results?

Clayton: I think this is one of those areas where it really, and this is one of those awkward points in a startup journey where the timing is challenging, where it’s easy to postpone adding manufacturing supply chain expertise to your team because you’re not yet building anything. But there’s a long lead time in getting ready to manufacture. And at least having someone on your team, whether it’s an advisor or a consultant or a full time part of your team who’s been there, done that, has the relationships, knows what it takes, who can help the team understand what needs to be done. I think, in my engineering career I’ve seen you know, there’s these classic friction points, and one of them is getting engineers to sell or getting engineers to do documentation. Those are two things that are never easy. And you can’t manufacture without documentation, and you can build something without documentation. You can build some really amazing products without documenting things. You can’t manufacture that documentation. And successful manufacturing is all about communication. You’ve got to communicate every last detail to the manufacturer about how to build the product so that they can actually build it repeatedly and consistently and with high quality. And everything that’s left to doubt is an opportunity for a defect. Something that gets built wrong, something gets left off, something it was assumed and not actually realized. And so, I think when you have that senior level, experienced manufacturing leader who knows what those things are, that person can orchestrate and recognize where the weak points are. We need to bring in someone to do documentation. Team needs to stop designing and start writing things down and start amending and annotating drawings and start preparing a set of documents that a manufacturer can actually build. Now, in some cases, depending on your product, your manufacturer may be able to help you with some of those things and there’s manufacturers out there who work with startups and have groups that are heavily supportive. And if you can find that kind of relationship, that’s super valuable because a manufacturer who sees a very promising product will be willing to extend themselves and really help you bridge that gap, that new product introduction phase. And because they see lots of business downstream that they’re going to get from manufacturing large quantities of this product, they’ll help you out and help you understand what you need to do. But you got to reach out and form those relationships. And if you’re just an engineering team that’s never built anything before.That’s a pretty tall order. So again, it goes back to find somebody who has some of those relationships. You can help pave that path. 

Adar: Exactly. And this is part of the tip of anticipating the next step, basically. If you can find someone on time that has the relationships or form them yourself, but don’t get to that point where you now have to manufacture products and you don’t have the resources or the relationships to do that. So just take it into account. 

Clayton: Exactly.

Adar: Do you have examples for things that you need to communicate to suppliers, things that you experience that are important to keep in mind?

Clayton: Yeah. So, another one of those, all these lessons learned, come from, they say good judgment comes from experience and experience comes from bad judgment. So, I’ve got a wealth of experience that comes from past bad judgments. One of the things in our product is that we have to certify to certain standards. In our case, we want to be in a commercial kitchen. So there’s an actual sanitation foundation, it is all about cleanliness and food safety then as a machine that’s going to go in a commercial setting, UL or ATL certification. So, again, anticipating that certification requirement is really important. We had one instance where we had a motor specified that was perfectly suitable and available and met all the specifications. But when we went through the certification process, it wasn’t a certified motor. And so then we had to hurry and go find a substitute part that was actually certified, which actually came from a more challenging source. Got hung up with some of the recent supply chain issues, and that one motor ended up becoming a sticking point in our supply chain process where otherwise we had everything nailed. And a lot of it was about not anticipating that you can’t put an uncertified motor into a system that needs to be certified and expect it to pass. So those are the kind of things where I think, someone who can anticipate the future, anticipate those next steps, can help you avoid the previous stuff that slows you down when you get into that current phase.

Adar: Such great tips. Clayton. Clayton, thank you so much for being with me today. That was very, very insightful. 

Clayton: Great. Well, it’s great to be here. Thanks so much for the opportunity.

Adar: Thank you.