Preparing to scale - Nate Walkingshaw,Scott Miller,Brady Forrest,Martin Kallstrom,Quinton O'Reilly

[00:00:06] Speaker A: Great to have so many of you here to join this panel discussion. Guys, thanks a million for taking the time to chat to us about scaling. So just, we may as well kick off straight away. Nate, since you're furthest away and I don't want to make you wait long for a question, like, let's just start off with a very general scenario.

[00:00:29] Speaker B: Sure.

[00:00:31] Speaker A: So I have a prototype, I've spent months putting it together, and I think, okay, right, this is the time I can finally start growing, turning this into a business, and think about making money and getting it out to the world. I have this prototype, what happens next?

[00:00:50] Speaker C: Yeah, I mean, there's lots of roads that you can end up taking here, candidly. But, you know, hopefully, you know, you've done a lot of your research, which is, you know, If you've talked to your user base— I mean, I could tell you a really candid story where I had invented a medical device. I had built a product for me, but not for we. And when I launched that product to market, not talking to any of my users, I ran

[00:01:15] Speaker C: my company into the ground the first year of launching a product. So I mean, I think some advice here would definitely make sure that you do all your customer validation and research even before you build your prototype, because that's going to shape the prototype. And then once you have your prototype built, validate it with users before you go to design for manufacturing.

[00:01:34] Speaker A: Is that a common problem that, you know, some entrepreneurs would create? You know, they'll see a problem and they'll say, oh, we need a product for that, and they'll develop it, but they'll think about their own situation instead of applying it to the wider audience. Is that like one of the big problems that happens, or something that you might come across often?

[00:01:55] Speaker C: Yeah, I mean, I think it's actually getting better and better. And I think just technology has, you know, especially in software development and technology, you can use a whole set of tools to get customer validation really quick. With hardware, that's a little bit different. If you have hardware and software integrated, you can get feedback relatively fast. But when it's just a hardware play, it's, I think, yeah, and people follow, you know, fall into those buckets

[00:02:21] Speaker C: from time to time. Mm-hmm.

[00:02:22] Speaker D: I mean, I think some of it is people— you've made a prototype, you only have one or two of them, and those prototypes are so precious to them, they're afraid to give them to customers. And they're worried they'll get broken, or they're worried that the customer might say something they don't like. And it's like this wound to them, and they need to get over it. It's something that entrepreneurs need to realize. They have to go to their customers and show off what they're doing.

[00:02:45] Speaker A: It's their baby, so they want to protect it from the world. If you criticize it, you're wrong, we're right, that's it. And Scott, actually, you're in a unique position in the sense that you've experienced life in terms of the scaling side from both your product with your time with iRobot and the Roomba, and your current role with Dragon Innovation, looking at startup— helping

[00:03:10] Speaker A: startups scale up their own products. Do you notice many parallels between your own experience and those startups that you help out at the moment?

[00:03:21] Speaker E: Yeah.

[00:03:21] Speaker B: Yeah, that's a great question, and it's, you know, such an exciting time to be in hardware. So when I started at iRobot back in the late '90s, there really were no other companies or small startups doing hardware. And for us, we had to do everything from scratch. So if we wanted to do something as simple as spin a motor, we had to not only write the code for that, but make the H-bridge and figure out what transistors to order, send it out, bring it back, and debug it. And that'd take about 3

[00:03:46] Speaker B: weeks just to get a motor spinning.

[00:03:48] Speaker D: Wow.

[00:03:48] Speaker B: But today, if you look at it using Arduino, my son and I can spin motors in like 10 minutes, and we don't need to know anything. So in some ways, it's a totally different experience. But in other ways, looking at the customers we work with at Dragon and sort of our own experience doing hardware at iRobot, there's a lot of parallels that— in our world, there's always cost, quality, and schedule. And living in consumer electronics, you have to manage all of those. Some people

[00:04:13] Speaker B: say you can pick 2. You really actually have to handle all 3 of them if you want to succeed.

[00:04:18] Speaker C: Mm-hmm.

[00:04:18] Speaker B: So by being able to see around corners and understand the impact of the early decisions you make on the long-term health of your business, and get those things right up front, is absolutely critical. And we kind of stumbled around and somehow got lucky and figured it out at iRobot. Today, the stakes are so much higher. There's so much competition. There's so many companies, in a great way, doing hardware, that if you can figure that stuff out early, you can really set yourself up to succeed.

[00:04:43] Speaker B: And as awesome as hardware is, it's incredibly unforgiving. So a lot of people talk about failure being good, and I remember all the lessons I failed at, but ultimately the goal is to succeed. And getting that stuff right that you can early will definitely help lay the tracks to succeeding later.

[00:05:00] Speaker A: And especially now in comparison to 5, 10 years ago, there are a number of other routes that you can take to scale a product or to get developed in the first place, either join up with an organization like yourself, crowdfunding, investors. There are various other ways to go the same direction, basically. So is it the case that

[00:05:25] Speaker A: the stakes are higher, but at the same time, the barriers to entry, they've reduced immensely, as you mentioned there?

[00:05:33] Speaker B: Oh, absolutely. So when we think of the hardware revolution, which I'd say is maybe 5 years old, And as we all know, things have been incredibly democratized. We really break it into 2 phases. Phase 1 is going from the idea to the prototype, and that's gotten definitely much easier with 3D printing, Arduino, Raspberry Pi, GitHub, and all these great ways to share and learn. Where we see the next frontier and where we're really excited at Dragon

[00:05:58] Speaker B: and at Bolt is how do you go from a working prototype to high-volume production? Again, this is just one piece of the puzzle. There's retail, there's fundraising, there's logistics. Hardware is so interesting in that it's so multifaceted. Where we are really focused is making that journey much more easy— easier and transparent, and providing the tools to let great companies get from prototype to production in a seamless

[00:06:24] Speaker B: way.

[00:06:25] Speaker A: Brady, you're with Highway One, which also helps a number of startups at an accelerator. I can probably guess why you would choose different startups, why you would take them under your wing, but just to begin, how many applications would you get from different startups wanting to apply to Highway One?

[00:06:46] Speaker D: It's been going up. So we're on our 5th batch. We've had 58 companies go through, 150 founders. Navdy, heads-up display for cars, is one. Cue, that does at-home medical diagnostics. is another. But for our last batch, we had 350 applications for 12 spots. Right.

[00:07:04] Speaker A: So competition is super high. Yeah.

[00:07:08] Speaker D: When we're picking companies, you know, we look at the team first and foremost, because the product may change, market conditions may change, but it's the team that you have to count on. And I really view hardware as a means of deploying software. So the team has to have really come up with something novel, something that's going to build data, build value beyond the hardware that they're initially shipping. Like

[00:07:33] Speaker D: when I look at Glowforge, the 3D laser printer that just raised $30 million in crowdfunding in the last couple of weeks, I see a product that right now is just light-years ahead of the competition. Nobody else has a camera inside their laser printer. Nobody else is going to make it as easy to use. But over time, those features will be copied. And the price will go down on the hardware. And so what the Glowforge team is going to have to do is build up a services business and

[00:07:59] Speaker D: a content business at the same time. And that's exactly what MakerBot was doing. As it was making better 3D printers, it was also building out Thingiverse.

[00:08:08] Speaker A: Obviously, this is a machine and hardware stage, but is that something that's something for everyone to kind of jump in here? Is that something that startups might neglect to think about? That although they're so focused on the product, they don't think about everything else around it? So like sales, advertising, customer service, those things. Are those the common things that

[00:08:33] Speaker A: would trip up, or is that something you would see quite often?

[00:08:37] Speaker D: Oh, absolutely. The companies coming in are so focused on the object, they often forget about those other things. And that's the type of stuff that, like, Scott and I try to work with them on when they're at Highway One or at Bolt, is you have to think beyond the object. Because when I take a Sonos home, which is a, you know, the multi-room speaker system, I look at the speaker for the first week, but then it's the remote that I use every day.

[00:09:02] Speaker B: Hmm.

[00:09:03] Speaker D: If I take the new Narrative home, you know, I'm going to love having it on my jacket, but then after a while, it's really going to— what's really going to matter is the software for dealing with the thousands of photos I take every day. And that's what we were just chatting about backstage.

[00:09:17] Speaker A: And speaking of Narrative, nice segue by the way, it started off as a Kickstarter project, the original one, and you originally set out to raise $50,000 and raised more than half a million, which is a very impressive outcome. But when I was looking through the Kickstarter page immediately before this talk, What the thing I was struck by was just

[00:09:43] Speaker A: how much detail there was behind where it's been manufactured, who is behind it, what the product does, you know, the pros and cons. And there are 2 reasons behind that. Obviously, the first one would be transparency, because that's what Kickstarter is based on. But I imagine the second one is because you have— when you're developing a product like that on your own, you have to have that foundation in place. Otherwise, it's going to trip you up sooner or later.

[00:10:08] Speaker E: Was that the case?

[00:10:09] Speaker C: Yeah.

[00:10:09] Speaker E: Yeah, exactly. And we really wanted to be transparent with all the work we did before the campaign. And I think it was both setting up all that in the supply chain, but also something that many companies or projects launching on Kickstarter and Indiegogo forget is the huge preparations that

[00:10:34] Speaker E: need to go into the campaign before you launch. It's very easy to say, yeah, we're doing PR, we know all of that, but to actually do it is very, very— takes a lot of work and a lot of time. And from crowdfunding in general, you get sort of market information, you get marketing, and you get the funding, of course.

[00:11:00] Speaker A: Did you get much help with the project or with the Kickstarter campaign, or was it just the The narrative itself, the startup?

[00:11:08] Speaker E: Yeah, so the way we did it was actually that we brought in seed financing before the Kickstarter campaign. So Kickstarter was a go-to-market channel for us, our strategy. And so we had a lot of help with setting up the campaign, and we learned a lot from a network of both suppliers, as you mentioned, but also people doing PR, helping us with that.

[00:11:33] Speaker E: And so on. What we weren't at all prepared for was the overwhelming customer support load that immediately comes with a crowdfunding success. We had to spend 24 hours a day just answering messages on Kickstarter and outside.

[00:11:58] Speaker A: And that's probably the benefit of Kickstarter, is it does act like a preorder service in a sense, that you can gauge interest and you can gauge how big this project or how this product is going to grow. But at the same time, there is— if you have overwhelming success like you mentioned there, there's obviously going to be some drawbacks in the case of having to deal— if you get, say, let's say 100,000 Narrative cameras out when

[00:12:23] Speaker A: you wanted 10,000, that's fantastic. But then all of a sudden you have 100,000 people to tend to because unfortunately with all kinds of products, there's never going to be a 100% Something is going to go wrong one way or another, and you have to deal with that.

[00:12:38] Speaker E: So we get a lot of requests to— I do Skype calls with some companies that are going to do hardware on Kickstarter or afterwards. And there are sort of 3 different problems that they're dealing with. Either they are doing preparations and don't know how to set it up, or they are in a campaign that didn't take off, as good as I wanted. And basically, when you're there, it's too

[00:13:03] Speaker E: late to change anything. But then the third one, which I had a call like that just a week ago, how do you handle customer support when it becomes overwhelming for the team? Because communication is really key to keep the value in the community and keep a good spirit.

[00:13:21] Speaker A: So I guess it brings up the question, like, how early should you think about these issues when you're developing your product? Should you be thinking about it when you're putting the prototype together, or should you wait on it for a little bit to have the foundation in place first before you think about it? Is there a happy medium that you can strike at all?

[00:13:43] Speaker C: I think some of the things that he's really striking on here is you're going to have to find a point in time, and when you're a startup, you're accelerating so fast, but you really have to pump the brakes. And really define a strategy about how you're going to deliver it. And honestly, it comes back with who you're aligned around. Like, if it's whatever user base it is, where you think you're going to have the most impact, you really need to align a lot of strategic initiatives and get

[00:14:08] Speaker C: everyone kind of aligned with that thinking.

[00:14:10] Speaker A: Mm-hmm.

[00:14:11] Speaker C: You know, just kind of hearing some of the customer service pains that you guys saw, I mean, I don't know what type of preparation was involved early on in that, but for myself, I knew that it was kind of ready, fire, aim. I didn't— I wasn't thinking about all these different attributes that were going to happen to me. They just happened to me after I had already made all the decisions. And just a word of caution, just to make sure you sit down with your team and

[00:14:37] Speaker C: you plan for what you think the outcome's going to be. It's going to change, but that's part of it.

[00:14:42] Speaker D: Something that we've learned from the film industry is when they go to Kickstarter, They raise a little money for the trailer, then they raise a little money for distribution, then they raise a little money for the festival, and they slowly build up their community. So what our hardware companies have started doing is small, constrained Kickstarters. We call them Kickstarter betas. So Fishbid is making kind of a nest for aquariums, and it's a pretty high-end product, so they produced 30. And

[00:15:08] Speaker D: it was really hard for them, they built it by hand, but they delivered it to their customers on time. Now they're gathering data, And when they go out in the next couple of months and do like a real Kickstarter, they'll have customer validation, they'll have data so the product will be better, and they know that when they actually ship the hardware, it's stuff that customers actually want.

[00:15:28] Speaker E: Mm-hmm.

[00:15:28] Speaker D: And so I think we're going to see more of those. However, founders have to get over their ego of having like the big blowout Kickstarter and realize that they need to ship incrementally. and treat hardware more like software and iterate before assuming everyone's gonna love what they do.

[00:15:49] Speaker C: I think that's good.

[00:15:50] Speaker E: Just to add to that, you have the perception when you go into a campaign that you will have a dialogue with your customers on Kickstarter and you will be able to change the product if there's something people are not happy about. As it turns out, there's something each and everyone is not happy about, and you get points on All over, and it's very hard to, once the product is launched like that, to actually change

[00:16:15] Speaker E: any aspect of it because everybody has something they're happy about and something they're not happy about. So you actually need to do the validation before you do the big start, big Kickstarter launch. I think that's a very good idea.

[00:16:30] Speaker B: Yeah, and preparation is absolutely critical. We find that it's such a new paradigm that you're selling before you've built it. So it's possible today, as we were talking about, to build a beautiful working prototype and then assume that that can scale to high-volume manufacturing. And the problem is, if you don't understand really how much money you need from a deep dive in the cost of goods sold or the tooling, which most people don't account for, the giant steel tools that will ultimately make your product, or the schedule, so from

[00:16:55] Speaker B: where you are now, when are you actually going to be able to deliver the things, that it's possible to get the backers all excited about the thing you've built, And then if you're a year late or you're losing money with every product you build, it's really hard to recover from that.

[00:17:10] Speaker A: And as you mentioned there as well, like the funding needed to create these products and manufacture them at scale is something that you don't initially think about because you probably think the amount of money it takes to develop a prototype and you multiply that by like 500 and that's your overall production cost. But really you have manufacturing costs to take in. Delivery as well. Are

[00:17:36] Speaker A: there many cases where startups say, or some startups might underestimate the costs of production? Because that can balloon up very, very quickly without warning.

[00:17:48] Speaker C: I was going to say, to his point about founders checking their egos at the door, the thing with a prototype and then going into DFM, design for manufacturing, is you actually have to be completely prepared for your product to change quite a bit. Like, when you're building a product at scale, it does not look at anything sometimes like your prototype, but it's the same outcome. But you really do have to be prepared for what the product is going to become, I think, after

[00:18:13] Speaker C: you go through, you know, DFM.

[00:18:15] Speaker D: And I mean, if you're a startup, almost by definition you're creating something new. And so you might have to invent some new manufacturing process, and that could end up being the big challenge that you face. And so manufacturing can go long. You could cut the wrong tools. You really just— you don't know what it's going to take.

[00:18:33] Speaker B: Mm-hmm.

[00:18:33] Speaker D: And VCs do not like to invest in inventory. They don't like to put their money down for working capital. And so they like to put in money to create the technology and to create the brand, but not at the bottom end of the smiley curve. And so that's where you want to work with partners like PCH or Flex, or perhaps get working capital from Your crowdfunding campaign.

[00:18:56] Speaker A: I know this is going to be a how long is a piece of string question, but when you have your prototype together, how much time should you give yourself to get all these things in place? Realistically, would it be just a few months? Could it even be a year, for example? Obviously, different startups will have different expectations and different needs, But at the same time, from

[00:19:22] Speaker A: what I gather, they can't rush into this because if they do, they're just setting themselves up for serious problems further down the line.

[00:19:30] Speaker E: Yeah.

[00:19:31] Speaker B: Go ahead.

[00:19:32] Speaker A: Give yourself, Martin.

[00:19:33] Speaker E: You didn't really define now what period you were asking for, but that doesn't matter because in a hardware startup, if you're asking yourself how long is this going to take, the answer is always a year to a year and a half. Almost whatever you're talking about, because if you don't know already, that's the time period you're looking at.

[00:19:53] Speaker B: Yeah, I think that's very consistent. Like, we find to pick a great factory is about 6 weeks, and you'd think it'd be a lot quicker than that. You could just go on Alibaba and find it, but whatever you find there is probably not going to be the best partner long-term, more often than not. So getting that relationship on solid ground is, you know, at least 6 weeks, and then it's often about 9 months from when you hand it over to the factory until it starts rolling off the line. And then from there, you've got to build up enough inventory potentially

[00:20:18] Speaker B: to fill up a can and then put the can on the water, and then it keeps going from there. So rule of thumb, you know, probably a year is a good, good timeframe.

[00:20:26] Speaker D: And the thing is, is the prototype is not the documentation. Like, nobody can build what you built by just handing over the prototype. You have to work out the documentation, how the product is going to be made, you know, the CAD drawings. There's a lot more rigor that goes into it than a lot of people realize.

[00:20:44] Speaker A: We're going to be wrapping up pretty soon, but just to finish up very quickly, what would be the main piece of advice you would give to a startup who are thinking of scaling or growing their products? Nate, shall we start with yourself first?

[00:21:01] Speaker C: Man. There's a lot of pieces of advice I'd probably want to hand out. But I think— I don't think we as entrepreneurs or founders, you don't spend enough time— and I'll come back to this— but really communicating a clear vision for your product. And really understanding— at Pluralsight,

[00:21:27] Speaker C: we actually boiled our entire business strategy down to one page.

[00:21:30] Speaker D: Wow.

[00:21:30] Speaker C: And that seems like it's impossible to do, but I just can't tell you how clear that made our vision and the implementation of that vision for the rest of the employees about how we're going to go execute our strategy. All the way down to everyone that's in the middle of doing daily work, it just has really helped drive a lot of clarity. So, I mean,

[00:21:56] Speaker C: The advice would definitely be try and shape a really clear vision down to one page at the end of the day.

[00:22:03] Speaker A: Cool.

[00:22:04] Speaker B: Realizing we're short on time, the one thing I'd impart is think a lot about quality. There's cost, quality, and schedule. Cost is obvious. Schedule is obvious. Quality is the unsung hero. If you don't get that right, you'll bankrupt your company.

[00:22:16] Speaker D: And sticking with the pithy statements, validate before crowdfunding. If it hasn't been in a customer's hands and they aren't happy, you should not go to Kickstarter.

[00:22:24] Speaker E: For me, it's build a really simple supply chain. You can easily get layers upon layers hidden for you, but have direct access to the factory because problems are going to happen. You need to solve them.

[00:22:35] Speaker A: And we'll wrap it up there. Nate, Scott, Brady, Martin, thanks a million for your time, and thank you for your time as well.

[00:22:39] Speaker D: Thank you.