Creating Category-Defining Companies with John Bicket, Co-Founder of Samsara & Meraki

Want to join the conversation? Connect with us on Tribe.

Find “Founders in Arms” on Apple Podcasts, Spotify, and YouTube.

Transcript of our conversation with John Bicket:

Immad Akhund (00:01):

Hi everyone. Welcome to the Founders in Arms podcast with me, Immad Akhund, founder and CEO of Mercury.

Raj Suri (00:06):

And I'm Raj Suri, co-founder of Lima and Tribe. And today we have John Bicket, co-founder of Samsara and before Meraki, right?

John Bicket (00:15):

Uh, yeah, that's right.

Raj Suri (00:16):

That's right. Yeah. Great. Thanks for coming on. Um, Samsara is a really interesting company, and oh, you're a very interesting guy. You know, I've been following you for a long time. Another kind of MIT uh, engineer building great companies, um, especially in hardware, software. I mean, it's really hard to build companies. I mean, Immad and I talk about how difficult it is to build, you know, you know, these hardware, software companies, so many of them end up failing and, you know, um, is that something that you've always known you wanted to work in or that you knew that it might be challenging, but it's something that you really enjoy taking on?

John Bicket (00:49):

Yeah, it, it's, uh, totally interesting. So first, thanks, thanks for having me on. I'm, I'm a huge fan of the, the, the podcast, uh, and, and, and love to follow on, and I love hearing these stories about other founders, just like the journey and the details that, that they go through. Uh, it's, it's, uh, and it is interesting just, uh, uh, to go through, I think, yeah, the, the hardware and software is really interesting and, um, and in a case, I know everybody has the kind of saying that's been kind of almost cliche about hardware just being really hard, and it's completely true. Um, I, I think for, for us, uh, and, and, and for folks that, that, that don't have the context, um, you know, or the company that, uh, that I, I'm the CTO for now, uh, it's called Samsara. We, we serve the kind of industry, um, of, uh, physical operations.

(01:36):

So this is like, uh, folks that are either in transportation or delivering goods and stuff like that. And we make a, we make, uh, hardware platforms, but it's really kind of a vector for us to get software out there so that we can, like read values out, we can help folks. Uh, uh, we, we have hardware units that will plug into like a vehicle, if that makes sense, and read sensor stuff off of them. Those are telematics devices that folks can use to kind of understand where their, where their, where their fleets are out in the world, how they're operating with the fuel efficiency of them and things like that. And then we also have a bunch of safety applications. So we have an AI dash cam, uh, that we install, uh, you know, in vehicles and helps folks that have fleets that, uh, of, of hundred or thousands of vehicles run safety programs really efficiently and actually like, dramatically improve safety outcomes and reduce the number of accidents, uh, that, that they see out in the road. These are

Raj Suri (02:28):

Those truck companies, like, like, um, what kind of companies are these?

John Bicket (02:32):

Yeah. You know, and it's completely fascinating to see it's, uh, I, uh, you know, we say physical operations and like what does that actually mean? Mm-hmm. Uh, and it is, you know, a big chunk is, uh, uh, transportation. Uh, so it'll be trucking companies, uh, LTL company, you know, a, a bunch of those. It also could, uh, one of our largest verticals is construction. And so, you know, it might be the actual yellow wire and the stuff that you would see at a construction job site, job site. But, uh, almost all those companies also have, you know, large fleets of Ford. You know, it could be like Ford F1 50s that are, uh, uh, that folks are driving out to, to go to job sites. Uh, could be field services and things like that. Almost. If you just, if you go through like the Fortune 500, almost every one of those businesses has some major physical component where they're, you know, delivering good, or they have, uh, folks that are, uh, out in the field doing, doing, doing work. And that is really who we've ended up focusing on for, uh, for, uh, at Samsara and really kind of trying to find technology that we can go help make their businesses more efficient.

Immad Akhund (03:34):

So would you say most of the differentiation is the software, the hardware is like relatively like commodity units that you're kind of plugging in?

John Bicket (03:41):

Yeah, it, you know, it's interesting. I, I wouldn't <laugh>, I, I wouldn't go as far to say as their, their com, their commodity and, and really kind of, uh, but they do look capability wise. It's kind of a lot like a cell phone, you know, or telematics units. And you know, they have in, in our dash cams too, they have a G-P-U AI chip in, in, in there, and then they have a cellular, you know, unit. They look like that. Um, there are a lot of details that end up being very hard. And these are physically demanding environments, so we actually do spend a huge amount of time going through and trying to, you know, make sure that they will work all the time, if that makes sense. Uh, in, in these, these, these units. But, um, but some of, some of the components, you know, like within them, like are, are pretty similar. Like they're gonna have a bunch of sensors that are, are, are pretty common and, and aren't that special, but then some of the other units like actually, you know, need to be, uh, we end up spending a lot of time on, so we kinda have to pick and choose, if that makes sense.

Immad Akhund (04:36):

Yeah, it makes sense. Uh, one of the things that's really impressive about, uh, you and your co-founders, you kind of did this twice, right? Like, I think like getting a startup right, seems like such a hard thing. Like you've got timing, you've got idea, there's so many aspects to get right. And I'm really impressed by people who can do it twice. Right. What was your process for kind of coming up with an idea? Uh, how much do you kind of give to kind of like the idea being great versus like your execution being great and the timing, and like, how do you think about all of these kind of components coming together?

John Bicket (05:12):

Yeah, it's super, super tricky. Uh, uh, you know, and, uh, I think, uh, for, for a little bit of context, um, our, our first, uh, the first company that we ended up starting, uh, was called Meraki and, um, worked in the, what I call edge networking field. So that was like, uh, we built, uh, cloud controlled, uh, wifi access points. Uh, we ended up expanding kind of in concentric circles over time, and the switches and routers mm-hmm <affirmative>. Uh, it ended up getting acquired by Cisco in 2012, uh, after about seven, I think, uh, the company got started in 2006 or 2007. Uh, grew very quickly, uh, got acquired by Cisco, and then we stayed there to help and help, uh, scale it, make sure it really got integrated. And today, if you go to Cisco and buy a access point or a switch, you'll actually, you can, you know, almost by default, like the, the access points you're buying will probably be Meraki and even the switches and some of the routers have the software inside of them that we were working on.

(06:10):

Um, that work actually came out of the stuff that we were doing for our PhD project. We were working on mesh networks, and I was, uh, uh, doing my, uh, we were in a PhD program. Uh, I ended up getting my master's degree on like bitrate selection and wireless networks. So I was kind of like down in the details on, on that stuff. And, uh, we were really passionate about that. We built out a big actual wireless network in Cambridge. Uh, and there was kind of a interesting trade off, uh, trade going on where we would build out this physical actual network. Uh, we would give away internet access to grad students or undergrads that would be willing to put a big antenna on the roof. And then the trade was, you know, uh, between, you know, two and 6:00 AM we would run experiments on the network, so we would, you know, do stuff so that we could write our thesis on it.

(07:05):

And, um, the, the fascinating thing for us, I don't think we realized it at the time, but actually that was like, uh, it was a product, right? Like we, we weren't taking, we weren't selling a product for money, but we, there was an exchange there, and we had to like, go out and support customers and, and, and do a bunch of stuff. I think we were so excited about that. Um, and the technology that we were developing, we wanted to see it have a big impact, and that's why we ended up kind of starting the, the company. Uh, and it was, uh, you know, it's so gratifying actually to see people use your stuff, right? Yeah. Like, you build something, it didn't exist, and then that feedback loop was incredible, and I think we just started running with it. Um, that company was really interesting though, because like, it was really a technology.

(07:49):

It was, you know, a little bit of a, the hammer looking for the nail in it. And, um, I would just describe it as a lot of the technology we were working on for mesh networking, which is these, uh, you know, we would put out these wireless devices and they all would kind of send a packet out and like, it would go hop by hop, uh, to reach its its destination, which is a little bit different architecture than normally you would do. Uh, the spot that all of that technology ended up being very useful was when wifi just went through an explosion of use because of the iPhone. Mm-hmm. That came out. And all of a sudden wifi had previously been something that was like, needed in a conference room, and all of a sudden businesses were like, well, crap, actually, everybody's asking for, you know, to be mobile, I need it in every square foot of my office facility.

(08:40):

So they went from like having to deploy one to like hundreds of wifi access points. And, and that the, at that point, the same technology we're working on to make the mesh networks like be automatically configurable and talk to each other automatically, stuff like that ends up being very valuable for managing a large network, a large fleet, you know, of network devices. Um, and so that was what kind of the bridge was, if that makes sense to the kind of commercial side. And that ended up, uh, for myself, <laugh>, I was a networking engineer getting to build networking products, right? Like, it, it was like, you know, I was building products for myself. And, and same, I think kind of the same thing for my co-founder. Like we were both in the PhD programs working on it. It was amazing, you know, just had the most fun time and you're kind of designing stuff for yourself, right?

(09:25):

So it's just like very fulfilling. That was, uh, totally and we ended up starting with wifi. And then I'll be honest, like we just started expanding in concentric circles, and that ended up kind of being a theme. Like we went from wifi, uh, wifi if, if, if you look one way to look at, it's actually very similar to a switch. It just happens to be wireless. So there's a lot, it's a lot more complicated. So we were like, well, we already built 80% of switch stack. We should just build a switch. And, and, and that's how we kinda expand it from wifi to switching, because every wifi access point, if you walk into an office building like this, it's plugged into something.

Immad Akhund (10:00):

Yeah.

John Bicket (10:01):

You follow the cable back and it's actually a switch on the other side, right. So we were like, I think people will, will buy them if we, if, if we build it. 'cause we're already selling 'em wifi, and you know, we walked up the stack at that point.

Raj Suri (10:12):

It's so rare though, right? Like for like, uh, a PhD or like, um, you know, an academic endeavor to be turned into like a successful company. Uh, I think even Paul Graham, I had like a recent tweet, he's like, he doesn't invest in any like, spinouts from academic research, uh, because they're always hammers looking for nails. They're, they're never problem focused. There's always technology focused, right?

John Bicket (10:32):

Oh, yeah. It's, it's total totally true. And I think honestly, like, it's a little bit glossing to, to say, to say like, you know, making that jump, it was incredibly difficult. And one of the hardest things, right? 'cause like, we were out, we thought that mesh, this mesh technology was gonna be the main selling point of the company. That wasn't the case at all. Actually. The, the users, their needs were much more basic. And they were like, look, how do I, you know, it was kinda like you get out in the field and people were like, look, I, I need, how do I configure the network name? Like, you know, the SSID and I, I was like, they were like, I need to do that on like a hundred devices. How would that even work? Y you know, and you're like, oh, yeah. Uh, I didn't think about that.

Raj Suri (11:10):

Yeah. That wasn't part of

John Bicket (11:11):

PhD.

Raj Suri (11:12):

Yeah.

John Bicket (11:13):

Okay. Makes sense. Makes sense. So like some, some of the, the, and, and it was like we kind of had to pivot and the, and the problems like that, and it was very difficult, right? Uh, but I think there's something magical about companies and, and one of the things I'm, I'm, I'm most excited about is like, actually, if you get that right, these, these companies are like a small, like nuclear reactor, right? Like, once you get that reaction going, it's like sustaining and, and it picks up, right? And all of a sudden, like, you know, you solve those problems for customers, you, you know, you're getting revenue back in the company, you can reinvest it in solving, you know, and, and so it kind of starts to, you know, have its own own trajectory, which is amazing. And, and it's like, it's hard to describe like, when that gets hooked up and it's so much work, right?

(12:02):

To like actually figure out like the, you know, um, but it's incredibly gratifying. So, uh, I, I think that that, that was one of the interesting things I think after we connected that like, yeah, we, you know, it was really difficult to come to those realizations and actually just be honest with yourself about like, oh, I need, I need to make something that people, people really want and are willing to pay money for, if that makes sense. That's the ultimate check, right? Mm-hmm <affirmative>. On stuff. But I think, uh, it's one nice thing, and, and like, I would've never, never, if you talked to me when I was in school, if you would've said like, Hey, are you gonna, are you gonna be really interested long term? And, and, and selling, you know, enterprise, enterprise software solutions. I, you know, I, I didn't, I just didn't understand or understand the nuance. Now I'm like, it's amazing. You can go talk to people. They'll tell you exactly like what you should do, what problems they're trying to solve. Like having that kind of clarity is incredible. So I, I think like, it, it's hard to appreciate like some of those stuff, like outside of the moment and stuff like that. But kind of, kind of interesting if you can get past some of the challenges.

Immad Akhund (13:08):

I remember the first time we installed a Meraki system at our office, uh, this was a long time ago. Uh, it was like magic. Like it was the first time you could get wifi everywhere. Before it was always like, oh, you had to like, get close to the router, and it would, you would get around there and then you'd go to the restroom and the wifi would stop working. So, uh, I mean, it was really awesome. We

Raj Suri (13:28):

Deployed Meraki at all of our restaurants, you know, when we were installing press.

Immad Akhund (13:31):

Oh, you

Raj Suri (13:31):

Did? That's that's amazing. Meraki supported all of our tablets, which we, you know, people use at Applebee's and Chili's

Immad Akhund (13:37):

Yeah.

Raj Suri (13:37):

To order.

Immad Akhund (13:38):

And it's a really nice feeling when, When you talk to people that have used your product. Like, I, I completely get that. Like, I, you know, I go to like a startup event and someone's using Mercury, and I'm like, oh, that's so cool. And it like, makes you feel great about it. Uh, was there a similar process for Samsara? Was there a kinda, uh, like a research thing that you were like, Hey, like this should be productized? Or was there like a very different kind of discovery process?

John Bicket (14:01):

Yeah, it's, it's really interesting. And, and I'm curious to hear from, you know, you guys also like, how, how you kind of like, have thought about this and, and thought about different, different opportunities and stuff. Um, you know, to kind of give you a little bit of context. So, uh, we ended up, um, uh, partnering and selling the company to Cisco, and in about 2012 mm-hmm

Raj Suri (14:20):

<affirmative>.

John Bicket (14:21):

Um, you know, we integrated in a company and spent, you know, a, a a a while there making sure it would be successful. 'cause it is actually like a really big kind of challenge organizationally for them to do, you know, to take in an acquisition and for it to be successful. And, uh, it, it was an amazing home for the, for the, for the Meraki team. Um, we kind of continued, you know, to do, to product development and things like that. But, uh, I'll be honest, like we, so we, we worked on those couple products. We had two or three others that we had been working with, uh, with, with, with the teams to kind of expand and everything. Uh, someone that was installing a network at a facility would need. So we, uh, you know, uh, the, the teams kind of worked on a couple new products, like we were working on cameras and, and a few other things, uh, there too, uh, aside from routers, which is one of the, uh, another one of the interesting product lines that they had.

(15:10):

Um, and after about a year, you know, a year and a half, uh, I think we did a lot of work to, you know, enable the rest of the organization to be able to carry that forward. And that has been really cool to see. I'm, I'm so glad we, we, we worked with that and we, we also went through and kind of identified the next, um, helped, you know, partner with the rest of the organization to say like, Hey, these are the folks that can really help lead the organization to the next stages. And like, uh, you know, someone ended up taking over my, my job as CTO and they did a, they did a great job. And similar for, for the, the other execs. Um, that was really gratifying to see as a result after about, you know, um, 18 months, it was kind of like, so what do you, what do you do next?

Raj Suri (15:51):

Mm-hmm <affirmative>.

John Bicket (15:51):

Right? Like, you know, we finished, we kind of ran out of all our, all of our initial ideas for edge networking, right? We had built like three or four different product lines. Each one of those you

Raj Suri (16:00):

Papers, right? Like you were writing academic paper, you were going back to that.

John Bicket (16:03):

It, it, it's funny you look that up. So this, um, I, I, I would, I would say, at least for me, and I'll speak for myself, and I, I don't necessarily think I could speak for Sanjay. It was a major, kind of existential moment and I had a lot of trouble with it, if that makes sense. And the reason we actually wrote a paper we're like, maybe, you know, maybe we'll go back to grad school, <laugh> <laugh> and work on that. And, and I'll be honest,

Immad Akhund (16:28):

Was it an existential moment because you had so much of your identity kind of wrapped up with Meraki?

John Bicket (16:33):

Yeah, completely. And so, so kind of what happened was like, you know, I, transitioned my, my role over to the new person. 'cause I wanted, like in case anything, I didn't think anything would happen, right? But I was like, if, if he doesn't like it or the person like, you know, doesn't end up being right fit, like, you know, I'll still be around and have some time to come back and just help figure out what to do. They did awesome, right? Like I handed it off and that was great. And so I was like, well, I guess what we can do is we can go back and write a paper and see what that was like, and I'll be, uh, so I explored that. Uh, I'll, I'll be honest, in the process, you remember that writing papers in grad school is not necessarily the funnest kind of process, if that makes sense, <laugh>.

(17:16):

So, you know, I was, I was like, oh, this sounded sounded great in theory. Uh, you know, oh man, I don't think I really want to go back and do this full time be, you know. And I think it was partially after experiencing, you know, like you guys said, when you show up to a customer site and you talk to someone and, and like, I mean, just like you, you guys to hear about saying you actually like, use Meraki. Like, that's an incredible experience. Um, and it feels different when you're writing academic papers because you still feel like it, it is like, you know, it's a, it's just a different type of satisfaction. And, and I think I was like, oh, it's not, not for me. I mean, I'm, I'm curious you guys, I know you guys do a bunch of investing stuff too, right? Like, um, I tried doing a couple angel investments

(17:56):

And also decided that, like, it just, I, I love it. I love supporting founders and stuff like that, but it just also, for me, it felt a little bit, a little bit removed, uh, and, and didn't, I wasn't getting the same like satisfaction. And then like, also it's, it's a, it's a lot of work. Uh, like, you know, I, it's not like magically is like easy to do. It's a different type of work, I think. Um, and so kind of just went down the line, went down the line and was in a little bit of like, you know, oh, should I do some non-profit work or something? Or, or just was trying to figure out. And I was like, I kind of feel like I just, you know, the more I thought about it more such a, you know, we, we kind of talked, we were like, I, I feel like I just spent the last like 10 to 15 years training to do these really hard things and I still still got some gas in the tank here. Like, I'm like, ready to go. Right? And it, it was really hard also because I, I don't know about you guys. We had recruited everyone we could possibly recruit

Raj Suri (19:00):

Mm-hmm <affirmative>.

John Bicket (19:00):

To Meraki, if that makes sense. Uh,

Raj Suri (19:02):

So they're all tied up. Yeah.

John Bicket (19:03):

They were all tied up and, and, and they're, they have great spots and stuff like that. And so, you know, I think part of it was we were like, you know, going to, and, and the other thing was at the time, uh, my, my wife was finishing up her PhD thesis, so, you know, can I come home in the evenings? Had handed all my stuff off trying to figure out like what to do. I'm like, Hey, great. Let's, you know, go, go out to dinner. And I, she, she was a little bit like, dude, I gotta finish my thesis. Like <laugh>, yeah,

Immad Akhund (19:29):

I'm busy.

John Bicket (19:30):

I'm

Immad Akhund (19:30):

Got time on your hands.

John Bicket (19:32):

So I think as a result, like Sanja and I just started hanging out. 'cause we were like, you know, and, and I'll be honest, I played like every video game that I hadn't done it for like five years or something. So like, I'm, you know, going through, you know, he's calling me in the morning. Like, and so finally we just started meeting up for brunch and like talking through things. I think we, we actually went back and we had seen these customers at Meraki at the very tail end that weren't like an office building. And we would, would go visit them just because it was, we thought it was cool. And I grew up in Oklahoma, so I grew up around a lot of farming and my mm-hmm <affirmative>. Grandparents were ranchers. I would go out with them and like, you know, spend a couple weeks with them usually during the summers, like bailing hay and working with farm, farm equipment and like what I, what real physical operations.

(20:18):

And, um, we went and visited a bunch of like wastewater treatment facilities, some of the transportation companies. Uh, we, we were trying to get manufacturing facilities 'cause they were using wifi to put like barcode scanners and stuff in. And so we started getting back into expose those customers, just go talk to 'em and see what they were doing. And it became very clear, this was like, you know, 20 started to be probably 20 14, 20 15, and he couldn't quite put our fingers on it yet. But all these folks you walk in and, and, um, they had a map up in, up in their office of where all their folks were that were out in the field for that day and pins. And then they would pull out their phones at some point during the, like, you know, conversation of getting the tour of their facilities. And they all had Facebook and, and they were using Uber and Lyft for stuff. And it felt, I don't know what the word is, is disco congruent or something where it was like,

Raj Suri (21:19):

Yeah,

John Bicket (21:20):

They're, they're using this technology in their, uh, in their personal lives and the stuff they were kind of like, it hadn't made its way into their actual operations. And it was like that, it seems like that's gonna change a lot. Yeah. And then I think, well,

Immad Akhund (21:37):

The way you describe it seems obvious, but you spotted that very early on. Like, people were kind of talking about iot, but no one was like really building kind of these enterprise systems. Um, I had a similar kind existential kind of like, what do I do next after I sold my previous company, uh, and I tried doing the investing thing. And it is, it's hard because like, you feel as an entrepreneur, like you want to get involved and tell them to fix their things, but like, that's exactly the wrong thing to do. And if you're not doing that, it feels kind of, it feels relatively empty compared to like starting a company, uh, where you are like

(22:11):

Responsible for everything. But I also tell people, and I have a bunch of friends who were successful entrepreneurs, and I feel like they kind of, like, there's a dangerous path where you just kind of become like, I don't know, they seem really bored <laugh>, right? They're like, they're traveling around, but they don't really have a purpose. Uh, and I always think like, like these are, like, these people could build great companies and I feel like you almost have like an obligation to build a company. I dunno, maybe obligations too much, but like, it's actually rare to be able to build a great company, right? Like, it's, it's quite a hard skill set, uh, and hard one, as you said, it's, it seems like such a waste to not like, use that to like better the world.

John Bicket (22:48):

Yeah. Yeah. I, I know it's such an interesting thing and it, it is like, it's, it, it's one of those things that it did, you know, and I had to do a lot of kind of reflections on, like, I, I don't know if you went through this too, I'd be, I'm very curious to hear your kinda your take on it. Like, I had to really think about like, what, you know, what, what, what do you want to do? Right? Like, you know, uh, and, and how do you wanna spend your time? I think the thing I, I, I did realize is like, there's, there's a big, there's so many different types of like, you know, may maybe for me there's a big difference between happiness and satisfaction, right? Like, doing very difficult things and like having to work through it, through the process is actually just incredibly gratifying.

(23:25):

And there's, uh, there's so much to be said for that. And like, man, you know, uh, I, I wouldn't describe myself as like an uber capitalist or anything, but like, the way that things are set up where you can have a bigger impact through a, through a company and it's bigger than yourself, like by definition you have to do it with a bunch of other people. Like, you can't just do, you know, it's very hard to do stuff like by your, you know, you can't just do it by yourself. So it's just like such an interesting process and by its nature, like makes you a lot more connected to like other, other people. Uh, and, uh, totally fascinating. It's like sometimes it's easy to miss that like day to day, right? Like, and, and all the struggles that you're, you know, are just required to like, get through. But, um, but yeah. No, I'm, I'm, I'm, I'm curious like what, like, how, how, how did you guys kind of come to this... Was it just like this was a realization over, over time for you? Or like,

Immad Akhund (24:19):

Uh, yeah, I've never, I'm a bit weird. Like, I don't mind doing hard things. Like it's kind of my default. I have a hard time doing easy things. Like I have a hard time just sitting at a beach. Like, I just find that like so dull, uh, uh, whereas like, you know, I don't mind doing like a grind or like, you know, I think those kind of things, as you say, are much more satisfactory, uh, like, finding meaning and satisfaction is way more important than like, short term happiness. And, and I just find, I've always kind of felt that, uh, yeah. Raj, I'd be interested after, you know, you had this kinda, I guess, difficult exit from Presto. Like how did you kind of decide what to do next?

Raj Suri (24:58):

Yeah, I, I mean, I, I think, um, I love building companies and it's kind of what I've always done. Uh, I was also a PhD, you know, student at MIT and, um, you know, uh, you know, I, I spent a lot of time with other, you know, really smart PhD students, like, who were doing, working on great, uh, research. And I did explore the, this idea of like spending out research, for example, right? And like that, um, there were a lot of great people doing that at that time. There was a lot of cool energy companies, and I was, my, my thesis was in chemical engineering, so there was some cool energy companies coming out of MIT. But, you know, uh, I think, I think when I look back on it, like almost all those companies struggled, right? Almost all those companies failed. Um, and, uh, um, you know, and I've always been thinking like, why is that there's all these great people at MIT and they really struggled to, to come up with like a good, uh, company right after like this ac you know, this academic journey. Um, and, you know, it sounds like what you did that was unique was like you actually did this like pilot test almost at m mit. You were actually testing it with real people, right? Right there. And maybe the technology also enabled you to do that, which is much harder to do that with like a battery company, for example, right? Necessarily.

John Bicket (26:04):

Oh,

Raj Suri (26:04):

Yeah, yeah,

John Bicket (26:05):

Yeah. Yeah. And I think it's really hard on these things. I feel like, you know, in some sense it's like, you know, the balance of like trying to get lucky with the initial starting conditions or like even what the research was if, if that makes sense. I mean, I even, they think like, considering anything in the hardware is so difficult. Uh, the, the fascinating thing I know that like Sanjit and I, and I spent a lot of time was, was, uh, you know, I mean, not, not dissimilar for, I, I think that stuff you, you guys do here is, is actually, and back then it was, you know, there weren't podcasts with founders that you could like go kind of hear the stories. We, we just went through systematically and actually looked up every successful academic entrepreneur and then like, really, really studied them. And like, there were, there was a bunch of folks at, uh, especially at Stanford that seems to be like the, the, they classically, I mean, there's so many stories about startups coming out, especially from the computer science department. It is a little bit, the culture is a little bit different than MIT

(27:01):

Mm-hmm <affirmative>. Uh, if, if, if that makes sense. And, uh, and, uh, you know, I, I know even, um, Sanjit ended up, uh, I think one of his advisors was, uh, Nick McEwen, uh, you know, and, and there's a, there's a cohort of professors, you know, from stand, I mean, you fall 'em every, every few years on, you know, they'll, they'll do, do some research and then, uh, and then that that'll end up giving them a bunch of ideas. So we, we, we did study those folks like very carefully, and I think it was helpful for us because we were in the networking space. So the run up to the dot com bubble and then after was networking was like the thing that, that, you know, kind of bubble was, was based on mm-hmm <affirmative>. And that was such a interesting, you know, case study to go through and see for both good times and bad, right?

(27:45):

Yeah. Like what, you know, what, how do, how do technology, uh, uh, kind of development work, uh, uh, in, in, in both kind of environments and how do people respond to it? But we, we ended up studying quite a bit, some of those, some of those folks that had, had done well and like, similar thing, like they're really incredible. Like they've actually, you know, they've had several, several companies that have, have been successful. And, and like, like you said, it's, it's just, it's, it's kind of interesting to go back and look and really try to think hard about like, what were the elements that were common between their, their, their startups, if that makes sense. And then like, what was different and stuff like that.

Raj Suri (28:18):

And, and, and you know, it strikes me right now, you know, I was thinking about it like, how many great, you know, companies have been started in the last 20 years, let's say, or 30 years in hardware, software, uh, industry, which we all agree is like, the hardest thing you could do is like work, work with this combination, uh, that have been successful. And I, I can only think of really like three, three categories of like, entrepreneurs, maybe I'm wrong, so let's, let's play this game here. It's like, it's Elon's companies, right? Like SpaceX and Tesla, right? Um, and uh, and then it's like Palmer Lucky, uh, with and Orel and, and uh, you know, and, and first he had the VR company as well, right? So he had kind of a similar journey in some ways. He had a first kind of big exit and then, you know, started Andel and there's you guys. Yeah. I mean, I can't think of anyone

John Bicket (29:02):

Else. I think it's one of these things like, it, it does end up being fascinating. I think if you expand the aperture a little bit and start going through, they're, they're do end up being a, a lot of, a lot of other companies, uh, that kind of end up fitting the bill that just aren't on, aren't consumer names, if that makes sense. Mm-hmm <affirmative>. And then some of those founders just aren't, aren't, are, aren't, arent, uh, you know, uh, talk, talking as much. I mean, I think, um, there are some companies, like Palo Alto Networks is a really interesting, like if you look at the financial side and like how they've done and how they've performed over time, it's like, actually, you know,

Immad Akhund (29:33):

It's insane.

John Bicket (29:34):

Been been pretty incredible.

Immad Akhund (29:36):

That's right.

Raj Suri (29:36):

Yeah. Yeah. Yeah. Better. Great journey. Yeah.

Immad Akhund (29:38):

There's also a few other kind of connected device companies, uh, like the Ring Doorbell, the Nest camera, uh, so consumer connected device companies. Good point. Yeah.

John Bicket (29:47):

Yeah. Yeah. So I, I think, I think it, it is fascinating. It is, I think the hardest part is getting started, uh, for, for these things. 'cause it's like, uh, and, and we've even found this, we, we just, uh, you know, we, we have been launching, uh, you know, a few more, few more products. And the, the interesting thing when I think through it, um, we, we, we just ended up launching a, um, an asset tracker that's Bluetooth based that I mm-hmm. You know, I'm super excited about. Um, and I mean, it's, it's tiny and folks are putting in on all kinds of stuff and physical operators, which just kinda makes sense. And there's stuff that, that, that previously, like, they couldn't track. So we, we've seen, uh, folks that have basically been sticking it on equipment that might get lost or stolen, if that makes sense.

(30:27):

And then being able to track it, we are able to use our, like, large network of devices. 'cause we estimate we're, we're on somewhere between, and, and I'm not exactly sure the, the numbers, but like of commercial vehicles in the us, you know, somewhere between three and 10% of them. So it means if you're watching trucks go by, there's a probably a Samsara device. And so, uh, the, these Bluetooth device, I mean, these things are tiny, right? 'cause like, I mean, and this is one of 'em. Um, they, they just chirp out occasionally because it's Bluetooth, it's low energy. And this technology has been amazing and it's, you know, been being developed for like 20 years at this, maybe that's not true. Maybe 10, 10 to 20 years at this point.

Raj Suri (31:08):

Mm-hmm <affirmative>.

John Bicket (31:08):

But constantly over time, the battery life's been getting better, the range has been getting better. Also, everybody's been under understanding how to build software systems around them and stuff like that. And so what we've ended up finding is that like now we have a big enough network that we can kind of piggyback a new technology like this on it. There's zero chance when we were getting started as a new company, we could have actually like, constructed a product that has these capabilities, if, if that makes sense. It was only once we had something at scale, uh, in a network kind of like out in the field in a real platform, if that makes sense, that we could even consider doing that. So I think, you know, it's kind of one of the, the hard things about some of, some, some of the stuff is like the plays right at the beginning when you're trying to just like find your f first foothold to be able to start scaling up, you know, the wall, if that makes sense. It is like so tricky and so hard to do. And like, the fortunate thing is that that's not really something that needs to be totally repeatable, and there needs to be a reproducible process once the company like gets off the ground. But it's so hard and why, like, I think, you know, um, there are all those analogies for like, just getting off the ground why it's so difficult.

Raj Suri (32:16):

Yeah. Did you find that, like it was, um, you know, when, when you're building this company, um, in the beginning, did you find that like, I mean, you were doing this as like a serial entrepreneur, so you already had a, a success, but for first time founders building like, you know, this type of company, do you think it would be much harder without that track record to attract capital, et cetera?

John Bicket (32:39):

That, that, that's a, that's a good question. I think it's a diff there's a different set of challenges, right? Like, you know, the, the certainly first time around, like you're trying to figure out everything. And I mean, I'm curious to hear like how you guys did it. I think, um, you know, like you need to figure out everything from like, how do you find office space <laugh>, you know, and like what does that look like? And you know, what, what's appropriate to do and things like that, to actually like getting product market fit. And that's really what matters. I think, um, the, the hard thing about when, when you're, you know, going for the, the second time, second time around, and I think one thing we're really cognizant of, uh, there were two things. One is, you, you're right. It's like, it's easier to recruit people.

(33:20):

Uh, but the, the secret thing that I think I would tell folks is like, and the thing that like, was really interesting, right? It's still, it's still really hard <laugh> and like, even recruiting, like, uh, was I, and I remember this early on, like we, we felt like we had an incredibly high hit rate on folks that we were going on. 'cause we were kind of like going back, uh, and, and talking, uh, to folks that we knew were very good and we had worked with before and we had a track record with, and still, like, our hit hit rate was like probably 50%. And part of it was like, you know, it's, and I think it's akin to on product market fit, timing. You know, like there, there are some folks that just like, you know, they may, they may just not be able to come join a startup because like, hey, if it doesn't work out in six months, like they, that's not acceptable to them and their family. Right? Totally understandable. Right? So there were some folks that we like, were like, we need to start recruiting them. And it still took, you know, a year or two to be able to pull them in.

(34:18):

And other stuff. But I, I think our second time around, once we ended up kind of identifying that we felt like there was something, and, and for us, what it, what it does, and it reminded me when we started talking to physical operations and understanding they were gonna go through this technology transition, it reminded me of seeing cloud and seeing mobile, you know, in those waves where it was just like, it was kind of, you know, you could see the graphs understand like what the technology curves were, and you knew it was kind of gonna happen and it was gonna be big. So it felt like that, it was like, didn't wanna miss out on that. So I think that's what propelled us, you know, forward. And then we just started trying to put in needing to put in the hours to like go do all the things that we knew we needed to do. Uh, the hard thing I think was, uh, that we ended up knowing was that sometimes it's easy to fool yourselves, if that makes sense. And, and this get ahead of ahead of your tracks. And we really forced ourselves to do some very hard things early on. As, as in, you know, one of the hardest things I think for early stage startups is you just to get the first deal across the line, we didn't want to dilute ourselves into like going and selling to someone that we knew.

(35:26):

We actually mm-hmm <affirmative>. Forced ourselves to cold call people that we didn't know. 'cause we were like, if we can sell our solutions to folks that we don't have a prior relationship with, that's like a stronger signal than, you know, than like, oh, someone, it's kind of, they might feel like they're doing your favor. We just wanted to like, avoid that, if that makes sense. So we were like, we need to figure out how to get the, like, um, how to develop almost a low noise amplifier for like where, where, where that is. Uh, and that was really, uh, really,

Raj Suri (35:56):

That's interesting insight, right? Do the hard things ear, uh, early and really test the product market fit, right? Yeah.

Immad Akhund (36:01):

I assume you could only do that because you had enough funding that it wasn't like you were trying to hit, like, you know, like you a really early stage startup without funding, you're just trying to get your first deal done and you take whatever shortcuts necessary to do

John Bicket (36:12):

It. Oh, completely. Right? Like, and I'll tell you, when we started Meraki, we, like, we were out in, we were in Cambridge and we went to like the local, local, local bank. I mean like, 'cause uh, you know, yeah. Things like Mercury didn't exist right at the time. <laugh>, you know, and, and we were like, Hey, we wanna search technology company. And they were like, they basically told us, no, we, we won't eat. You know, we won't give you a, we were trying to find a loan to put down to buy our first, uh,

Immad Akhund (36:36):

Oh. Yeah.

John Bicket (36:37):

And you know what Sanjit and I actually ended up having to do was like scrape and claw. Like, we went out and did a bunch of consulting so that we could actually get money to buy the first hardware unit. So we were like trying to figure out like those kind of straight offs. It's, I think it's kind of interesting. It's just like a different set of constraints that you're trying to solve for. Right? And it's almost like test of if you can get off the ground so hard, so hard.

Immad Akhund (36:57):

Yeah. I think when you are, uh, well, let's just a second or later time founder, you really have to make sure you're not diluting yourself. And like, I think actually money actually makes it easy to dilute yourself, right? You can, if you have enough salespeople or enough like, you know, put fancy enough like, uh, kind of things out there, you will make some sales. Like, actually you kind of wanna make things a little harder for yourself so that you can like really understand whether you have product market fit.

John Bicket (37:22):

Yeah. Yeah,

Raj Suri (37:23):

Absolutely. I think second, I think repeat founders who have had some exit before, like, they often make the mistake of raising too much money. And, uh, yeah. And like, I've seen it, we, you and I might have seen it so many times, right? Like,

Immad Akhund (37:34):

I don't think raising money by itself is a problem. I think spending is the problem, <laugh>, right? So if you have the discipline to not spend it, it's fine. We raised 6 million when I started my career and we, we only grew a team of like eight people for two years, basically, uh, until we had launched and we were sure it was working. Uh, but it's really hard to have that discipline. Like it's very tempting to,

Raj Suri (37:52):

What did you do to keep that discipline? Like how I, I I don't think I've ever seen it done really well. So maybe you can,

Immad Akhund (37:56):

So basically what happened is, like, I talked to a bunch of other FinTech entrepreneurs and like, I heard all the horror stories and like the horror stories are like, Hey, it's gonna take you. Like, and this was 2017, so FinTech was still fresh and the horror stories like, Hey, it's gonna take you like three years to launch this thing. Uh, and, and I was like, okay, you know, no matter what, like I wanna launch this. Like, I don't wanna fail because I ran out of money before we launched. Uh, so the whole path of like getting to market was quite, uh, quite an unknown. I mean, the end, it took us one and a half years to launch. Uh, but I was really prepared for taking, because I was like, if it takes three years to launch, then it's gonna take like six to nine months to prove anything.

(38:32):

Uh, so I needed four years runway. So I very paranoid about spending the money. Uh, I mean, in the end, I guess we could have spent it more and everything worked out. But, but having, but also, like, you know, when you have a team of less than 10, let's say, like everyone knows everything. They're all working like to the goal. It's like super, like there is no collaboration bottleneck. Uh, as soon as you go above 10 20, uh, you end up just slowing things. Like every incremental person is just lower. Uh, and I, I've always been like a believer in that, uh, that helped. Uh, I guess we coming towards the end and I really wanted to ask, uh, John, about culture. Uh, we've hire, we've hired a couple of Samsara people and I always felt like the, you know, the, the, they talk very fondly about Samsara and the culture you built, built. Uh, now you are I think around 3000 people. So I guess, you know, how did you think about building that initial culture and how have you thought about kinda scaling the culture?

John Bicket (39:30):

Yeah, you know, it's been really interesting and I think it was one of the things we, we focused on when we started the company. And uh, I think there, uh, especially the second time around, 'cause we knew that that was actually like as founders, right? That's one of the biggest levers that you have. Mm-hmm <affirmative>. Right? And, and actually like as a founder, no one else can really set, it's actually your job, right? To set, yeah. Set, set the culture for that. I think part of like why and why we communicated the, the, what we were doing with starting the company was actually to have an impact out in the world. So that ends up being a huge, huge part, you know, of our culture. And we are very customer focused. So, you know, uh, we spend huge amount of time, uh, out trying to understand, I mean, 'cause that's really our business. Like if we're, if we're successful, we want to be viewed as a true technology partner with our customers where they say like, here's my business problem I'm trying to solve. And we're like, you know, we'll go find the technology to help kind of bring that, bring, bring, bring that to them. So that ends up being such a, a big, a big part of, of,

Immad Akhund (40:29):

But when you say impact on the world, you're not necessarily saying like, Hey, um, we are curing cancer or something. You're saying like, we're, we are helping our customers like make their lives better. <crosstalk>

John Bicket (40:37):

That's right. Yeah. Yeah. And you know, we, we, we, we ended up kind of establishing that there's kind of three mechanisms we have for doing that. Like three vectors I guess we have like, you know, it's either gonna be efficiency, safety or sustainability. And sustainability actually just maps back into usually efficiency and, and, and, and to some aspects safety. Um, so those are like, I mean, I will tell you I was out, you know, we just hear back from customers. 'cause like these, the, the, the physical operation shelves are just actually physically demanding and very hard. And so in some ways just being able to like actually assist folks with making their like jobs easier to do and not require as much cognitive like processing or just like, you know, huge. Uh, and so, but part of it is actually like, it's very hard to understand what, like, if you're doing a ride along with the driver and just like seeing what they're doing all day, if that makes sense. 'cause they're probably out in the field. Like it's actually just very, you can see the stuff that they're

Immad Akhund (41:30):

Working on. You have, do you have a program to have employees kind of ride with, with drivers and like, be it customer sites?

John Bicket (41:36):

Um, yeah. We, we, uh, we have, uh, especially for on our product side, like our, our folks spend a huge amount of time with customers. Um, I think one interesting, I mean the, it has gotten easier to do this. Um, like, I mean, clearly our go to market side, they're almost in the field just a, a lot visiting customers trying to understand like how to help 'em out and stuff like that. On the product side, uh, it actually got easier after the pandemic with Zoom just because people will get on a Zoom with you and, and give you feedback. And then on the engineering side, yeah, we, we try to like drive, uh, basically all the product development back to tying it to customers, uh, that would actually use it. 'cause that's like, kind of like you guys were, you know, we were alluding to earlier about like not diluting yourselves.

(42:17):

Sometimes it's very difficult on the product side to get wrapped up in your vision about like, what, what something to do. And it's always a reality check when you like, show it to a customer and get the reaction. Right. And um, and you know, if they're not like super excited about it right? That usually is some kind of signal. So we try to make sure that we're actually listening to customers all the time. I mean, I probably did, in the last two to three months, I've, I've probably done like, uh, somewhere between 40 and 60 zooms where I just sat on the Zoom, asked the customer about a area, ask them to walk me through, like, and, and then at the, you know, I, I get their dump on how they work today and then I, you know, we walk through product ideas with them and get the feedback on it. And, and like they will, they will tell us, uh, it, it's very nice. 'cause if you do that 30 times, you can, you can kind of have a distribution on like how.

(43:05):

How good your, your stuff is. So we, we focus on that a huge amount.

Immad Akhund (43:09):

Um, so going back to the culture thing, you said like, yeah, talking about impact and customers. What, what else did you do that worked?

John Bicket (43:16):

I I, I think, um, you know, so those were kind of the, the things setting what the, what we thought were the, the values and just being very upfront with, with, with folks, folks about that. I, I think part of the hard thing is that there is the articulation of the values and then there is the, like what does that mean day to day is, if that makes sense. So I think, you know, all, we expect kind of all of our, of our leaders to be like, very involved in the in and a lot of the details and in checking stuff. 'cause I think part of the hardest thing as you're growing, um, a company and, and, and, and, and starting to scale it, is making sure kind of the, uh, customer experience matches what you want. If, if, if that makes sense. And what you say about it by the values and it,

Immad Akhund (43:58):

You know, what was the value that was like, the hardest to stick to? Like is there one value that you're like, okay, this one is like the most challenging?

John Bicket (44:10):

I think the, the one, the ones that are the most difficult to understand, I think for folks, and there's a lot of nuance on is like, and we, we have one of our values, it's like build for the long term.

(44:23):

And it almost is at, at, you know, and I, I think it's, uh, so sometimes when you articulate, you know, you have your values and you have like op we, we have a set of operating principles, uh, and other companies like, have these too. The, the hard parts are when they come into conflict, if that makes sense. Like, you know, one thing I think for this that we've kind of articulated is that speed does matter, you know, in, in, in the streets, especially in tech, right? So having to pit that against doing the right thing for the long term is really tricky, right? Uh, because I think some folks will interpret building, doing the right thing for long term as being like, well, I need to build something totally for scale <laugh>, if that makes sense. And it's okay to take, is it okay? You know, how long, is it okay to work on, work on something without shipping it to a customer? Is it, you know, a month, six months, a year, two years, three, three years? And I start to get more uncomfortable with myself starting to say things like two to three years.

(45:14):

<affirmative>. I think the hardest thing is to try to figure out, like creatively, you know, how do you bridge the gap, right? How do you stair-step something so that you're getting value out of it, but also working towards a long-term vision. And these are problems that like the, the kind of devil is in the details on them, right? And there's a lot of hard decisions, like whether it's on the go to market side, whether it's on the product development side, or whether it's on the engineering side about like when it, you know, how to make these trade offs and how to, how to think through them. I think those are probably the most difficult. Uh, especially when, you know, you're having people join the company from another company that might have had their own ideas. They may even use some of the same language mm-hmm <affirmative>. Uh, and mean very different things, uh, which is really tricky to work, work through organizationally.

Immad Akhund (45:58):

Yeah. I think those types of things where like speed versus like R one that's often a conflict is like we are, we are all about quality. Like build something of a really high quality that you can be proud of. Uh, but then you're like, okay, well that's gonna take an extra <laugh>, you know, two x more time to build it to a high quality standard. And then you kinda have to say actually, like, let's ship the thing that's, you know, maybe not quite at that standard, but like, when you learn from customers, you can kind of, uh, you know, your definition of quality actually changes when you have like customer feedback and you're like, oh, actually, like it doesn't work for them. So that's not a high quality product.

John Bicket (46:31):

One. One of the, I'm I'm curious if you have any tactics for this. The, the, the best one that I've heard of, you know, or one of the examples is, is some folks will say something like, Hey, we have, we have an idea. We know we want to do this, and it would be a big project if we want to do it tactically. I think sometimes it ends up being an interesting thought experiment. Say like, what is the, so we, we may have the like two year we one year version of what that project would look like, but what would be the one month version and then what would be the one day version? Like, and I think that the fascinating thing that I've seen from folks is like, even saying the one day version usually actually ends up taking like a week to, or two weeks to do. Yeah. But it was so simple. It was almost elegant. If, if, if that makes sense. Yeah.

Immad Akhund (47:16):

Yeah.

John Bicket (47:16):

And I'm not saying that that's always like the right answer, but it's an interesting thought process to try to figure out how to kind of

Immad Akhund (47:21):

Bridge the, I have, I have like two ideas around this. Number one is like similar to what you're saying, like cut the scope down, like make, make it do the simplest possible thing that solves a problem. Right? Uh, and then the second thing that I try to push alongside that is you cut the scope all the way down, but then you try to do one thing like so well that it feels magical to someone. Uh, 'cause I think like the way people perceive quality is like something that's irrationally good, right? If you're just like, good, like everything else, that's not really high quality. It's like something where you go like, like I sometimes call it like a magic feature or something like that. Like you take one thing and you go like very, this is obviously in a bigger product that's taking like a few months to develop, but taking one thing and going like really deep and like, ideally it's something really meaningful to someone. Uh, I think that gives the impression of, uh, they're like, okay, you know, it's missing some features, but this one thing is just like crazy good. Yeah. And that like actually leaves like a really strong kind of quality signal in people's minds. Uh, and you can fill, fill out the features over time and with customer feedback.

Raj Suri (48:19):

Yeah. The, the one metric I have is like, when you're demoing something to someone, it, it should elicit a wow, you should actually get that reaction. If you don't get that reaction. It's not good. It is not like magical enough, you know?

Immad Akhund (48:32):

Yeah. That's a good way to put it as well.

Raj Suri (48:33):

Yeah. That's,

Immad Akhund (48:33):

But I do think the, the bit that I care a lot about quality isn't like these kind of attention to detail things. Uh, and I actually don't think those take that much time, right? Like, if it's like, you know, the button color is awful. Like, it's like just not in the right place and like, you just, I just feel like people shouldn't take those shortcuts. Like, yes, it takes a little longer to have like the attention to detail, but like, I think it's also a training thing like, uh, once you care a lot about attention to detail, like you just default to doing it in the correct way. Uh, so that bit, like, we are very stickler about, like, I think having things slip when it comes to attention to detail is like kind of like a, a road that like is a slippery road. Uh, and you kind of have to like, be, be thoughtful about it. Um, all right. We're kind of coming to town any time. Any question, any last questions? Raj?

Raj Suri (49:19):

I have tons of questions for John <laugh>, but I think we'll need another hour, John. We'll need to have you back. You know, this was fun and, uh, I, I learned a lot and I love, I feel like there's a lot more we can talk about.

John Bicket (49:30):

Yeah. A awesome, I really enjoyed the time and, and love the podcast. Thank you. Thank you for, uh, thank you for having me on.

Immad Akhund (49:36):

Yeah. Thanks for joining. Uh, and thanks for listening everyone. Uh, if you are interested, please leave us a review. That always helps and subscribe everywhere we are. Uh, and thanks for coming, John.