This Week in Startups - $MSFT, $GOOG, $SPOT earnings breakdowns + Coco CEO Zach Rash | E1596

Episode Date: October 27, 2022

First up Jason does a solo breakdown of recent earnings reports from $MSFT (1:38), $GOOG (13:59), and $SPOT. (26:46) Then, Molly is joined by Coco CEO Zach Rash for another edition of The Next Unicorns! (39:09) (0:00) Jason tees up today's segments (1:38) Microsoft earnings breakdown (12:42) OpenPhone - Get an extra 20% off any plan for your first 6 months at https://openphone.com/twist  (13:59) Google earnings breakdown (25:20) LinkedIn Jobs - Post your first job for free at https://LinkedIn.com/unicorn (26:46) Spotify earnings breakdown (33:54) Why now is the best possible time to start a company (37:58) TripActions - Go to https://tripactions.com/twist and get a $500 Amazon gift card after making your first travel booking OR paying off your first $1000 of liquid spend (39:09) Coco CEO and Founder Zach Rash joins Molly Wood to talk about his robo-delivery startup! (55:38) Last mile delivery market size, why Coco is focusing on remote-controlled vs autonomous, and more! FOLLOW Zach: https://twitter.com/zac_rash FOLLOW Jason: https://linktr.ee/calacanis FOLLOW Molly: https://twitter.com/mollywood Subscribe to our YouTube to watch all full episodes: https://www.youtube.com/channel/UCkkhmBWfS7pILYIk0izkc3A?sub_confirmation=1

Okay, everybody, welcome back to this week and startups. We have an amazing show for you today. I'm going to break down the big three earnings reports that came out in the last 48 hours, Microsoft, Google and Spotify. And I'm going to explain to you what's going on inside these companies at this moment. And then Mollywood interviews, Cocoa CEO, Zach Rash, as part of our next Unicorn series, season four of that series. You know, they do those remote control delivery robots, the little R2D2 that brings you your burrito. People are giving up on this space, but Coco is not. Zach, the series. You know, they do those remote control delivery robots. You know, the little R2D2 that brings you your burrito. People are giving up on this space. People are. People are. But Coco is not. Zach, Zach, Zach, founder is a really interesting guy, and I believe in this. I want R2D2 bringing me my casidilla. It's going to be a great show. Stick with us. This week in startups is brought to you by Open Phone. As a startup founder, a lot of mistakes are easy to roll back, but using your personal cell phone number as your company number isn't one of them. Open Phone makes it easy to get business phone numbers for you and your team, right on top of your existing devices. Visit openphone.com slash twist to get 20% off your first six months. LinkedIn jobs.

A business is only as strong as its people, and every hire matters. Post your first job for free at LinkedIn.com slash unicorn. And Tripactions was built to help businesses scale without travel and expenses being a pain point. It's the only all-in-one travel, corporate card, and expense management solution. Go to Tripactions.com slash twist

to sign up for free, and get a $500 Amazon gift card after making your first travel booking or paying off your first $1,000 of liquid spend. All right. First up in earnings, Microsoft reported for its fiscal Q1, 2023. So, you know, the fiscal year is like taking the calendar year of Q3 and Q4 and calling it Q1 and Q2. So right now we have everybody reporting on their Q3 earnings, right, which would be July, August and September, what Microsoft and some other companies call a fiscal year starts with

Q3 being essentially Q1. So we'll call this Q1, but you can think about it as the summer months. Their shares are down after reporting their slowest revenue growth in five years. As we've talked about on this program, we knew going into a recession, going into a down market that the weakest companies would experience the downturn first. What were the weakest companies and products in the world. Cryptocurrency, meme stocks, and maybe high growth tech stocks, perhaps some of the SPACs that were very speculative and didn't have products in the market yet. You can think of a Virgin Galactic, right? Something that is really forward looking, or maybe a electric car company like a Rivian, or even some of the car companies that weren't producing

cars yet. So the first cities to fall in this, you know, war we're going through called the recession are going to be the weakest companies. It could be a startup that doesn't have product market fit. It could be a cryptocurrency project that is built on hot air or it could be an NFT project that has no intrinsic value. Now, what we're seeing, right, if we consider this maybe the third quarter of this down market recession, and a recession is two quarters of negative GDP, but, you know, largely a recession is this kind of down market where the GDP is low and there are job issues where basically people can't find jobs. And that's the strange thing about this recession is that there's still so many jobs available. But wait for it. We always said when we were

talking on this weekend startups and the All In podcast that perhaps the last companies to fall would be the strongest. In other ones, the ones with the most cash position, when they show weakness, when they have the largest amount of revenue, they're still highly profitable, when they start to see their market caps and their stocks come down. And when they start to see the headwinds impact their growth, one will know that the recession is really sort of hitting its final stages. Now, how long it goes sideways, that's a big question. And how long are their headwinds? That's a big question. But here we are. Microsoft has basically their slowest revenue growth in five years. Their current market cap is $1.75 trillion. And their stock is down 30% in 2022. This represents a $750 billion drop in Microsoft's

market cap from $2.5 trillion all the way down to $1.75 trillion today. Now, that doesn't make them like Peloton, where their stocks down 90% or other growth stocks like that. And so let's just talk about their revenue. They're at $50 billion for the quarter. Think about it, $50 billion in one quarter. It's $200 billion a year. It's only up 11% year over year. But remember, 11% on a big number is a large amount of revenue. And it's down slightly quarter over quarter. So when we report numbers, you have to think about two things in your head, year over year. So at this time last year, what were they making? And remember, last year at this time was 2021.

And that was the summer of 2021. That was the boom time in the economy, right? When lots of cash was flowing into the economy because of stimulus spending and the government printing a lot of money trying to get us out of that COVID downturn. Well, quarter over quarter, even though they're up year over year at Microsoft, quarter over quarter. In other words, the Q4, which would be the second quarter of the calendar year, that Q4 were down. So they were at 51.9 billion at that time. So they've actually had less revenue than they did last quarter, but up from a year ago. Microsoft doesn't disclose

specific segments. So if you wanted to know Azure's revenue or LinkedIn's, they kind of bundle these things together. That's for competitive reasons. They don't want people to know exactly how specific product lines are doing. Good or bad, right? They don't want to tip off competitors. So they kind of bundle stuff together. And one of the things they bundle together is cloud services revenue. Then they have search and news ad revenue. And you can look at all of these together. And Azure and other cloud services revenue is up 35% year over a year.

Now, last quarter, the growth rate was 40% year over year. So you're starting to see that the cloud computing revenue is not growing as fast year over year. It's down a little bit, right? But it's still growing. So the revenue number is going up 35%. But the rate of growth went, from 40% to 35% quarter over quarter. So it's a little nuance there, but it does show headwinds on the business. Search and news ad revenue was up 16% year over year now. Last quarter, again,

the growth rate was 18% year over year. So that is slowing down in terms of growth year over year. LinkedIn revenue up 17% year over year, and that's down 9% from last year. And last quarter, it was booming, 26%. But for the first time ever, Microsoft Cloud metrics exceeded 50% of the overall company's growth. The cloud segment includes Azure, which is cloud computing, goes up against AWS and Google's cloud offering. And they're Office 365 subscriptions, which is known as Microsoft Office previously. They call it 365 now because you're subscribing to it month over month. And some parts of LinkedIn, et cetera. The weakest part of all of this is what's called Windows OEM revenue. OEM stands for original equipment manufacturer. In other words, when you buy a PC

from Dell, hopefully, a great company, you will have windows on it, and they pay to put windows on it. That's dropped 15% year over a year. So you're saying, hey, why did that happen? Are people not buying PCs? Correct. People are not buying PCs.

Why is that? Well, do you remember the COVID boom? And then the post-COVID bust? That had packed in a lot of companies, like Peloton, let's say. Or companies that were doing at home, Zoom would be another one, another company that was booming during COVID would be these pop-in and hop-in rather,

other virtual conference software companies. These companies became darlings because in the boom, in the COVID boom, we're staying at home and everybody bought a new PC. Now, COVID-boom, post-COVID bust had a corollary, which is the reopening companies, Uber, hotels, cruise ships, etc. okay, now we're going to get out in the real world. So you lost revenue. People stopped taking Uber's or going to hotels during the peak of COVID. And then things open up. Okay, you can go on a cruise ship and even the cruise ships today no longer will require you to show a vaccine card.

PC sales are down massively. Not because people are not using PCs, but because a lot of companies said, okay, we'll buy you a computer for home. We're going to send one to your house. You're going to be remote working or people were stuck at home and said, oh, yeah, I'm going to treat myself to a new computer so I can play better video games or get more work done, whatever, whatever your jam is. But PC shipments fell 15% in Q2 of 2022 compared to Q2 of 2021 according to IDC. That's a market intelligence firm. That's not Microsoft's data. That's IDCs. So that shows you if less PCs are being shipped, well, of course, Windows revenue is going to be down. It dropped 15%, which is the same amount as PCs. So you can just see it. It's not a Microsoft issue as much. It's a COVID boom-bust

cycle operating income $21 billion for Microsoft. And so, you know, they are, they have just massive amounts of cash. Microsoft did warn, according to the financial times, that revenue growth from Azure was going to slow by five points in the next quarter. So they're just telling people, hey, listen, there's going to be headwinds, right? So the 40% we saw over a year, over a year went down to 35. Okay, so it's going to go to 30%. Why is that? Why would cloud computing go down? Well, it might be there's less startups being started, companies are tightening their belt and they're looking at their cloud spend. And this is what happens. I saw this happen during the dot-com bust and in 2008. When the market gets challenged and you see your income going down, you see less revenue, you just look all around and you say, what are we spending

money on? Literally today, I'm in my office and we, you know, we became a remote company. And I saw this just sitting in my office. I was like, oh, what's this? A Verizon M-Fi. And I'm like, hmm, I turned it on and it's working. And I said, oh, yeah, we used to have this as a backup in the office in case we lost our internet, we turned this on. At least we get email going, et cetera. So I emailed the ops team, and I'm like, hey, what's going on here? And so, oh, yeah, we have two of those. We spend $100 a month on them when people travel.

I'm like, okay, should we be spending $1,200 a year on this? Turn this off. Okay, so Verizon's going to see $1,200 turned off. And that's what everybody does. I'm looking at all of our expenses, at all of our company, saying, hey, what SaaS software are we really using? What are we not using? What hardware are we using?

What services? Turn that stuff down. Turn that stuff down. Turn it off. So you can be sure every team is looking at their spend in cloud computing and saying, hey, how many service do we have? How much storage do we have? Okay, can we put some of that storage into glacial storage, really, you know, slow storage? And let's see if we can make our databases smaller. Let's re-optimize our code so that we're not using as refactor it, so we're not using as much CPU. All that's happening. And people just are not as disciplined during a market. Why should they be?

Who cares about five or ten percent of costs here and there? When you have, top line growth of, you know, you're tripling revenue every year, you're doubling revenue every year. You want to focus on growing revenue, not your costs in a down market. Everybody clenches up and starts going through item by item, what can we cut? And that includes humans and team members, sadly, includes office space. And that is this process of everybody becoming disciplined all at the same time. And we're starting to see that happen even to mighty companies like Microsoft. And mighty categories like cloud computing, that nobody is going to be immune from belt tightening. And so everybody's tightens the belts, everybody gets austerity measures, as we call them.

You're going to see it impact every single company in all likelihood. And here is a beautiful Sankey chart. We talked about these before on previous episodes. It's like kind of the ribbon chart where you see 50 billion in revenue. When you go left, you see it broken down into cloud, the intelligent cloud, and Xbox, and, you know, where all that revenue comes from, Microsoft, 365. LinkedIn, productivity, software. And then going to the right, you see the cost of the revenue, $15 billion, the gross

profit of $34 billion, operating profit 21, down to the net profit of $17.6 billion. This company just throws off, you know, tens of billions of dollars a year in just net profits. It is extraordinary. Satya has done an amazing job with cloud and moving this company to a new business model where people pay for subscriptions. They've done an amazing job challenging and catching up to AWS. overall great company. I own the stock.

I'll continue to own the stock. Might even add to it in a down market. And let's move on to Google. On the program today is Dorena Kulia. She is the founder of Open Phone. Welcome to the show, Dorena. Thank you so much, Jason. Great to be here.

Now, what mistakes do most founders make with phone numbers in their startups? Really, delegation, right? Because what ends up happening is as a founder when you're starting, you do everything. You are the salesperson, the support person. You make the coffee, you do HR, marketing, sales, recruiting, everything. But then eventually you have people joining the team. And what ends up happening is if as a founder, your phone number, let's forget about the privacy, the spam, all that problem.

Let's say it doesn't exist. But you're not going to want a year into your company, two years into your company, to have all the support calls or all the questions come to you because now you've just hired your support team. Why did you hire them? So that's another reason why. having that separate number makes so much sense because you can always delegate those calls to your team as you grow. All right, everybody, here's your CTA, the old call to action.

Twist listeners, 20% off any plan for your first six months. Just sign up at openphone.com slash twist. And if you got an existing number, no problem. They'll put it right over. Openphone.com slash twist. O-P-E-H-O-P-H-O-N-E dot com slash twist today for 20% off. Google, you know, feels like a software company. It feels like the same as. Microsoft, but it is not. It is an ad business. It is very similar to other ad businesses in the world that preview existed like magazines and TV networks, except theirs is a much better ad network. They also are down 8% today and 33% year-to-date, $700 billion drop in market cap from $1.9 trillion to $1.2 trillion. And the reason is because they missed their expectations ever so slightly.

and the total revenue for Google was $69.1 billion in Q3. They go by the calendar year. They were up 6% year over year. A company like Google, because they're advertising based, is going to have more headwinds. Why? Well, if you're talking about belt tightening, if you're talking about austerity measures, people are going to say, you know what, these ads, let's just do the minimum we need to do to keep our business growing. Let's be less frisky, be more disciplined in our ads.

spend. And so you'll see ads like billboards, outdoor advertising. That category tends to get hit first because you can't track it as well as you can track something like Google Search. So Google Search ads are the best in the business in terms of being able to attribute, a click and a spend, the dollar you spent to a dollar you earned. They don't get hit as hard as TV ads. They don't get hit as hard as outdoor ads. Those things tend to get hit first, newspapers, magazines, the little bit of advertising that still exists there, will get hit pretty hard. But revenue is going to slow

for ad-based businesses like Facebook and Google and the ad-based businesses inside of companies like Apple, which has an ad-based business and LinkedIn inside of Microsoft. Q3, ad revenue, $54.5 billion. That's up only 2% year-over-year. We've seen this movie before. In a recession, the last time we had one in 2008,

Google had similar headwinds against them. And so this is becoming, you know, an acute issue. Google really needs to focus here on what they're going to do to solve this problem. When you look at the $54 billion in ad revenue, $39 billion of it comes from Google search, about $40 billion. And then YouTube ads make up $7 billion. Now, YouTube ads have been booming, right? YouTube has been spectacular for Google. Eventually, we might see a world where YouTube ads become, you know, maybe on par with search ads.

That's not certain, but what we're seeing this quarter in Q3 was that search was up 2%, but YouTube ads were down 2% year of a year. Okay, now why would that be? You got one company here. They've got video ads, which you would think, hey, these are really great, and then you have click ads. Well, click ads have better attribution and they're more targeted. This makes sense.

Remember I said before outdoor advertising goes first, then TV and newspaper ads, magazine ads, go next. Well, YouTube is most analogous to TV. And Google search ads, those are the crown jewel. Why? You type in Volvo, Santa Monica. We know you want a Volvo in Santa Monica, as opposed to, hey, Volvo of a Santa Monica is running ads on YouTube against people in Santa Monica who have an interesting cars, right? Or in a certain strata in terms of their income, right?

You're inferring that they would be interested in a Volvo in Santa Monica based on, you know, different signals. But you don't know they type that in. So it makes sense that YouTube would hit those headwinds first. And sure enough, they missed expectations by $400 million, which isn't like the end of the world, but it does show you what happens in a recession. And you'll know the recession is ending when you start to see TV ads, newspaper ads, outdoor advertising come back. And in fact, when those things are sold out, that's the sign of a top of the market.

When the 101 freeway here in the Bay Area, you know, in Silicon Valley, and that goes from San Francisco all the way down to Facebook and Cupertino and Apple and Google when you're taking that highway I always know when we're at a peak market when I see startup companies spending 30,000, those billboards cost

20, 30,000 a month. When you start seeing people buy those and they're sold out or, you know, it's one thing to see Apple have two or three of those billboards for the new iPhone. It's another thing to see Twilio have them. And then some startup that's much smaller than Twilio, you're starting

to wonder why are they spending a quarter million dollars or if they have two or three of these a million dollars a year on billboards. Well, you know, somebody in recruiting or they're just loosey-goosey with their ad budget. When you see those billboards not change, what they'll do is the secret is they'll leave somebody's ad up. So I was talking to a startup. I was like, hey, why are you spending on this? This is in the last downmark. And they said, oh, no, we're not spending on it. They left our ad up for a year because they couldn't sell it. And they didn't want people to think that nobody was buying them. So they don't want to put your ad here. The second of Billboard puts your ad here on the billboard, what happens? People say, oh, those don't work, as opposed

to leaving it up. So somebody literally told me it might have been Zenga that they hadn't paid for an ad for a year because they were just up and running for free. So it's just interesting. When you know the ad market, it's a very interesting signal for how the market is doing. Q3, Google Cloud revenue was $6.9 billion. That's up 37% year over year. And they beat Xx. expectations. Why? Well, Google Cloud, if we're being totally honest, trails, Azure, and AWS. It's kind of an after-not. When you run a company like Google, and most of your revenue comes from ads and the search ads are the crown jewel and YouTube is the up-and-coming, cloud revenue is playing third or fourth fiddle here. You know, they're the fourth or fifth scoring option on the team. And, you know, a CEO and a management

team can only keep so many of these players in short rotation. And, you know, cloud revenue has just not been great for Google historically, but they're working on it now and they're taking it much more seriously. I see that because I see the cloud revenue. I see the Google cloud people out at events. I see them doing some marketing and they're coming up more and more in startups like Azure was maybe five years ago. So when you start taking competition seriously, which Google started taking their cloud, I think seriously, maybe two or three years ago. Microsoft started taking it seriously five to ten years ago, and obviously Amazon took it seriously for the last 10, 15 years. So Microsoft and Google catching up to Amazon, I do see Google taking

it more seriously, and I guess that's why they beat expectations. This is where it gets interesting. And obviously, this is super profitable. The net income for Q3 was 13.9 billion. That's down 26 year over year, and down 12% quarter over quarter. So it's coming down their profitability essentially. Now, why? Because they are addicted to hiring employees. My lord, Google has 187,000 employees. That's up 25% year over year from 150,000.

They literally added 37,000 people in a year. Why? Why would you do that, Google? It makes no sense. They added 13,000 people in the past three months. So if you were like, hey, we think it's going to be a mild recession. Let's add, you know, a bunch of people, you know, we'll have a head start when we come out of this mild recession. Well, we all know it's more severe.

And you would have changed years in the past three months. They still added 13,000 people. Who's running this company? What are they doing? They're just adding tons of people and they don't need that. You can't be increasing 10,000 people without communicating to the market why you're doing that. So Sundar, I think, is, you know, going to be. under the gun because of this. And you saw them cut, you know, like some people in their in-house incubator

called Area 120. Their game service, what's it called? Stadia? Stadia, maybe. They cut some of that, but they're not making serious cuts. They're hiring like this is, you know, 1999, 2006, or 2021. Absolutely unforgivable. And the market does not like it. They really need to start thinking about making cuts. paradoxically, confoundingly, Sundar was doing all that saber rattling the last six months, telling the company, hey, everybody's got to be more efficient. Here's an idea. Get people to come back to the office. Start making cuts.

Do a hiring freeze. This is really hard to take seriously. Now, of course, Google is sitting on a ton of cash. Google has tons of profits. They're throwing off money. So, you know, there's an argument to stay the course and keep adding people if you don't care about public shareholders. But your employees are compensated with the stock.

So if the stock is going sideways or going down, you can have a hard time hanging on to that talent. Everybody knows this company is overstaffed, just like Twitter is overstaffed, Apple, Facebook. Everybody knows these companies are overstaffed. And the employees, let's face it, are coddled. When you see employees going to, you know, alphabet, Google or Facebook, and, you know, having $30 in snacks a day,

coming to the office twice a week. You know, I get it. It's a different world, but it's a bad look. And I think Wall Street is looking at this, and they're going to hold management accountable for spending. And that's going to be the big trend of the next year. And I do think, I've said it from the beginning,

I thought this would be a three to six month recession. I'm sorry, three to six quarter recession. I thought it would be. And here we are in what is probably at the third quarter of a recession, the first two, quarters of the year, this quarter, the third quarter that we're reporting on now, and we're probably soaking in, you know, arguably the third or fourth quarter of this, it's going to be another two or three quarters. I don't think that we'll get back to growth again in GDP until the

second half of next year. That's been my belief since the beginning. And that's not taking to account some crazy event occurring like an invasion of Taiwan, which I put as low single digits, the Russian invasion, the illegal and immoral invasion, and war in Ukraine, Ukraine that Putin has done, you know, that war continuing or escalating, probably put that again, you know, continuing at 20% and escalating at, you know, low single digits, you know, the tactical new concept, you know, probably less than 1% for me, my belief. Other people might differ in that belief. Certainly some people I know do, and they think it's 20%. I think that's not going to happen. But if something like that does happen, that could maybe

extend, you know, a recession and create even more chaos in the markets. And so what this means is you're going to see a hiring freeze, I believe, at Google get announced. They can't keep hiring like this. The market is going to sell shares in Google and find other places to deploy their capital. And you saw friend of the pod, Brad Gersner, say, hey, we want to see a 20% RIF, a reduction in workforce at Facebook. We'll see if Zuck takes that advice. Zuck's revenue is going to get hit even harder than Google's ad revenue and probably in line with snaps, or snaps, as I like to call themselves. All right, everybody, LinkedIn jobs.

We have to talk about the downturn right now. There is one silver lining for all of us running businesses, especially small businesses or nascent companies. You know, when you got that half dozen dozen people, the talent pool is getting stronger and stronger. And people are looking for interesting companies to go work for. And that's where you come in. And that's where LinkedIn's going to solve all your problems.

When you run a startup, you run a small business, you know every single new hire is high stakes. So you have to ask yourself, what if I hire the wrong person? What is that going to do to my team? My team dynamics? Well, that's why you have to check out LinkedIn jobs. There's so much great talent out there right now. And LinkedIn jobs helps you find the right people for your team. And they help you do it faster.

Now, you want to be 100% certain. You have the right candidate pool that you're looking at. Well, that's LinkedIn. You're talking about over 800 million people are. are there. Use screening questions. Ask people thoughtful questions. Hey, what if you were a podcasting company? What do you love about podcasting? What are your top three podcasters? And why? Hey, if you're an app company, have you downloaded our app? And what do you think of it? And maybe give me some feedback. So that will help you filter the people who really want to come work for you, who have that

passion. LinkedIn jobs helps you find the qualified candidates you want to talk to faster. Post your job for free at LinkedIn.com slash unicorn. That's LinkedIn.com slash unicorn to post your job for free. Terms and conditions do apply. Spotify down 11% today. After reporting larger than expected losses in Q3, they're down 65% year to date from a $45 billion market cap to just a $16 billion market cap. That is extraordinary. It makes me actually, if I'm being honest, maybe want to potentially do a J-Trade in Spotify. I think this is a great company. I think they're a market leader. And I like their forays into other content. I think they are eventually going to be competing with Netflix. I think they're going to be doing more video shows. I think they'll be investing more in advertising-based

content. And they've got 195 million premium subscribers. That's up 13% year over year. And their monthly active users is at 456 million, 20% year over year. Ed supported monthly active users, 273 million, up 24% year over year. So they're still growing briskly in terms of their user base, total revenue 3 billion. And that's up 20% year every year. We converted that from Euros. Remember, you know, Netflix has 220 million paid subs. And the Disney collection of assets, I think, is at about 250 million between Disney plus Hulu and ESPN. So they're right up there with those contemporary. Is that probably a slightly lower dollar amount per user? But that doesn't mean that that you shouldn't think about these as comparable companies. I think they are comparable because

people look at Spotify increasingly as an entertain a place to find entertainment as much as music. And so I wouldn't be surprised if you saw not just podcasting deals on Spotify, but series deals. I'm not saying Game of Thrones, House of the Dragon, Lord of the Rings level. But is it so crazy to think that Spotify might sign a professional sports league and you watch it on Spotify? I don't see why not. Now, they don't have the bankroll to go after the NFL, but maybe a cricket league or something else is possible. And why wouldn't they do a documentary series

or a reality TV show or a TV series? I think that's completely possible, if not probable. Why wouldn't you do that if you were Daniel Ack and the team over there? If you understand how to do different serialized, you know, true crime stuff, it's not such a big jump to have video versions of that and shows. And young people don't have a problem watching a TV show on their phone. In fact, it's kind of their default. So a 20 minute, 30 minute show watching it on your phone or your iPad, why wouldn't Spotify

do that? And that's why I think it's not just about leading in music and now in podcasting. I think it's really going to be about increasingly them coming to do shows. And if I, you know, I'm doing a reality TV show now with a major reality TV producer that I signed with. And I'll have more news about that in an announcement. in the coming months. But when we bring it to series,

me as a creative who, you know, would have a choice of where to put the show in all likelihood, I would totally consider Spotify if they wanted to do, you know, a startup show with me. Why wouldn't I? They do a great job for this week in startups.

They do a great job for All In. They support video now natively on the platform. And my audience is already over there. So that would just mean a company like Spotify would have to get comfortable spending $750,000 to $1.5 million per episode of a TV quality show, buy 10 episodes. You know, you start doing the math on that.

It's a bigger investment than they're probably used to making $10 million. But they already have the advertising relationships from the Joe Rogan podcast and from Call Her Daddy. So those same advertisers will probably have an appetite for larger deals and ad spending on things like The Ringer. And hey, listen, you look at Bill Simmons, he's done some of the greatest sports documentaries ever. Where are those sports documentaries on Spotify? Why wouldn't you have a 30 by 30 series from Bill Simmons? Why wouldn't you have the team at the ringer do not just a podcast about a great director or, you know, they do these like rewatchables? Why wouldn't they do this is the 80s

science fiction, right? Bill Simmons should be doing those type of documentaries with the ringer staff at a place like Spotify, not on HBO, not on Netflix or not just on those kind of platforms. The growth in their ad-supported revenue accounted for 13% of total revenue, which is 19% year-over-year. So their total revenue is $3.1 billion, but they're starting to see ad-supported revenue account for more of their overall revenue. And that's led by podcasting, according to the earnings report. They have 4.7 million podcasts on the platform. I think that's kind of an irrelevant number. What it's about is the number of listening hours, and then which ones they can actually monetize and sell ads inside of.

obviously like I don't have a deal with them for this week in startups. They don't, we sell the ads ourselves. At some point would they come to us and say, hey, we want to buy it out? Perhaps, but you know, they went after some of the anchor ones. What's going to be very telling is do they re-sign call her daddy? Do they re-sign Joe Rogan? Or does Joe Rogan go to Amazon, right? And I could see Amazon, Netflix.

Really, Netflix should have gotten the Joe Rogan show and call her daddy. But I think they were being a little snooty and maybe being a little highbrow thinking they were going to be at the Oscars and the Emmys. really the best thing for Netflix to do would be to look at Spotify's business and Amazon video going over and trying to steal Spotify's podcasting business. And conversely, as I've just discussed, Spotify should be trying to steal Netflix's business. And Spotify should say, hey, if you've got a streaming show, come to us as well. We're all in the same business. Subscriptions, eyeballs, advertising-based revenue. And now with Netflix having an ad-based

product coming soon, you could see them maybe going after podcasters. Spotify's got a ton of cash, cash equivalence, short-term securities. There shouldn't have $3.7 billion. If they want to start buying shows, they can start spending tens of millions, hundreds of millions buying shows. They're basically breaking even on a cash flow basis. Their net loss was $167 million. They signaled that they might raise prices in North America on the call. Heck mentioned, a friend of the pod, that Spotify has, quote, significant pricing power and can fare better than competitors with price hikes. Spotify premium, 10 bucks a month, 16 month for a family plan. I have a family plan. I would pay 25, 20, 25 for my family plan, sure. So you'll maybe be spending more. Here's a chart.

Again, another Sankey chart from App Economy on Twitter. They do a great job. You can follow them. Revenue 3 billion, premium, $2.7 billion, ad supported, $4 billion. They're really trying to get that ad supported up because that gives them some independence from the music industry. Their cost of revenue is pretty high. They are shipping a ton of that money that they make to the music industry. They got to get off the music industry train in order to see that gross profit grow. When you look at the gross profit of Google or Microsoft or these other companies, they've got that really big fat gross profit line and not a lot of operating loss there because their cost of providing the service is much lower because it's software. They don't have to give a huge chunk of their revenue to the music labels. So there you have it. Just some insights into what's happening in the market. It's now the best time ever to start a startup company. That is what you should be thinking if you're listening to this week in startups.

Because these large companies are in retreat, because they've got to look inward and clean up the mess of the spending and do this belt tightening and turn off this goddamn me-fi that cost $1,200 a year, right? Because people are doing all that work. What does that mean? Well, that means if they are going to go after your startup and try to copy it, they might not have the time to do that. they might not have the resources when somebody internally says, oh, we should copy this startup or

oh, we should add this feature. They might say, you know what? We need to be more profitable. We need to do more with less. We just laid off 20% of the company. We're on a hiring freeze. Stay focused. Now is the best time for you to start a company. There are people who don't want to be at Google or Microsoft or any of these large companies or Spotify during a downturn. It's kind of a bummer to be there. Maybe they'll get rid of the snacks or maybe they'll have a, you know, a hiring freeze and you've got to do more work and you got to come back to the office. So this is the time for you to start a company. If you want to do that and you want my help, I'm teaching a class called founder dot university.

500 people have signed up already. I'm trying to get to, you know, maybe even 1,000 people. And here's how it works. You give us $500 to come into the 12-week course. Every Monday night at 6 p.m. Pacific, I teach the course for 90 minutes. If you come to all 12 weeks or you have an excused absence, you know, you got a kid's play. I don't want you to miss it or recital. But if you come to, you know, 10, 11 of the 12,

we just refund the $500 so you don't burn the seat. We had 96% of people graduate from the previous classes. So this is extraordinary. And it's essentially free if you come to the classes. And if you don't, we just use that money to throw an in-person party, which we'll be doing on November 14th. I'm going to do the first class in person. I'm going to invite everybody to come out to the Bay Area.

I'll probably do two or three of the classes in person. It's my way of giving back to the community. And our investment team looks at the companies in Founder.com university. If they're making good product progress, my team is just going to give $25,000, probably a dozen or two of the people who come. We've invested 25K and 18 of them. Consider this like your friends and family check instead of asking your aunt, your rich aunt or your parents for 25K to get your company incorporated and get started.

Consider me your rich uncle. And I put the 25K in. for 2.5% of your nascent company or or or, or you can come to our accelerator where we put $100,000 in and or if you get a couple of customers, we might directly invest in your company. So founder.comiversity is our way of meeting founders who just have an idea. You don't even need to be incorporated. And we'll teach you how to use no code software and why startups exist and how to grow them and on the margins, how to raise money. And you join this great online community. We have set up with some software. And there's 50011.

in there, and then we're going to add 500 to 1,000 people. Seek it to hang out with your tribe, builders. I'm really excited about it. This is a fourth cohort of founder. Dot University, and I'm doing it personally because I love startups, and I love founders, and I love being the first supporter of your startup, specifically. Yes, you, and the person who's thinking, eh, maybe I'll quit Google, maybe I'll quit Spotify and start my own company.

Get two or three of your friends. And, you know, if you're at Spotify, you're at Google, you're at Microsoft, and you're like, you know what, I don't want to be part of the riff. I don't want to be part of the, this, you know, grind it out. But the two or three of you, product manager, a designer, a developer, two developers, one designer, whatever it is. Go start a company. Come up with a great idea. And let's get to work. Let's do the work. All right. Next up is the next unicorn series. My pal and partner in crime, Molly Wood is going to interview Cocoa CEO, Zach Rash for the next Unicorn Seasons four. You don't know what Coco is. Those are those mini robots that deliver burritos. And I'm in love with these robots. I think they should exist. Some people are giving up on them and other people are doubling down like Zach. And so I'm really interested to hear this interview that Molly does. Stick with us.

Imagine how much better your life would be if business travel was easy and seamless and drama-free. How much time could you save? How much money could you save? And how much happier would your team be if the travel policies at your company were simple and clear? Well, that's why you need to check out Trip Actions right now. Trip Actions was built to save business as time and money on travel. It's really that simple, folks. They have the single platform solution,

and it seamlessly manages your bookings, the itineraries, corporate cards, and of course expenses, including automated reports, which are a pain in the neck when you have to do those expense reports and then people feel like, oh my God, I'm late, paying my credit card. No more drama, folks. Plus, on top of all that,

you're going to get 2% cash back on every purchase from Trip Actions. Their mobile app makes everything so easy. Trust me, you need Trip Actions. for your business travel. And right now I have a special offer just for startups. Visit trippactions.com slash twist to sign up for free. And you'll receive a $500 Amazon gift card after making your first business travel booking or paying off your first thousand of liquid spend.

To learn more about this fantastic special offer for startups, go to tripactions.com slash twist. Tripactions.com slash twist. I am excited to be talking to a next unicorn, no pressure. No pressure. Zach Rash, CEO of Coco. How's it going? Good. Thanks for having me.

Tell me what you're building. What's happening at Coco? Well, we make, I brought one. That guy. This guy, yeah. So this is Coco. So we deliver food and groceries with these remotely controlled robots. And we're operational all across L.A. and parts of Texas.

How long have you been in operation? A little over two years. Coincidentally, we started this right when the pandemic lockdowns happened, when the biggest selling point of the service was contact free delivery. Our first users were actually the elderly community in Santa Monica getting a discounted grocery delivery during the pandemic. But that was not planned.

So you were one of the companies for whom the pandemic was actually, I don't want to say well-timed, but it was a tailwind. It was a tailwind for you. Yeah, it was a tail-well, actually, it's funny because the tailwinds, the biggest tailwinds coming out of the pandemic were actually more related to the labor side of things because it became quite a breaking point for food delivery as everyone started ordering things online.

There weren't enough drivers to fulfill it or willing to fulfill it. Right. So that actually added a lot of benefits that weren't directly related to the pandemic, but a lot of the indirect consequences were good for us. Wow. Well, congratulations on the accidental good slash terrible timing. Tell me, so tell me where you are now.

Let's see, you've raised, it looks like, about $63 million over four funding rounds. The last raise value due at $551 million in February 2020. Investors Included Founders Fund and Open AI Sam Altman. What do you think is the biggest pitch? Like, this is an idea whose time, it sounds like, has come in lots of ways. But what's your, what's your special, your secret saucer? Yeah, I think this has been attempted quite a number of times. I saw a company, I saw a robot doing food delivery when I was 16 years old. And I remember seeing it and saying this is going to be, this is definitely going to be the future.

I don't know if it will look exactly like this, but these sort of purpose-built electric vehicles should be moving things around a city environment sometime in the future. And that seemed to me at the time to be pretty non-controversial. And I went to UCLA. I met my co-founder there. We both studied computer science and electrical engineering. We've been building robots our whole time there. And we graduated.

And this industry just hadn't really expanded at all. And, you know, if you go and look and talk to a lot of these companies, a lot of them have a very deep tech focus on how do we make these as autonomous as possible? They're basically building like slight simplifications of a self-driving car. Yeah. And as you know, that is, we're notoriously wrong about the timelines on autonomy and when it's actually going to be a commercially viable product. And, you know, they kind of back the envelope mouth with Coco just seemed like you can make this deliver.

you can make this kind of a company profitable without having any autonomy. And so then autonomy becomes this incremental benefit you can layer in over time as your fleet grows as you collect more and more data. And this is kind of like that more Tesla philosophy on autopilot is you have people paying you to label your data, to train your systems, and to collect all of these corner cases out in the world. And so we wanted to be really focused on making a product, giving it to customers, and really learning how to actually operate this and focus on the operations of scaling it up, rather than saying we're going to focus on this kind of deep R&D for the next decade.

Right. So they are currently still human piloted? Yeah, we have an autopilot product. So pilots begin using the autopilot product. But I think one thing that people forget when we talk about automated delivery is quality is incredibly important. Like food delivery in its current state is not great. I ordered dinner on Saturday night from a restaurant I love. It's 0.3 miles away from my house.

and it came completely cold, and I knew it was going to probably come cold. And that is a common sentiment from a lot of people. There's certain types of food and certain brands that deliver really well, but the vast majority of it is a pretty poor experience, and you're paying a huge premium to get that pretty poor experience. So we've just been super focused on how we want to start with the quality aspect. We want to start with the customer, and we want to get this out there

and say, how do we deliver a better experience than people can today, and then increasingly automated over time. So that autopilot product began, we began introducing it on return trips. So not when you have hot food, hot perishable food in the robot, but start doing it when you're a little bit less time sensitive on the trip. Got it.

Okay, so then tell me about the inside of the robot. Like, how are you using the device itself to optimize for that quality delivery? Yeah, so if you see like a number of big enterprise brands care so much about delivery quality that they actually staff their own drivers. This would be like Chick-Flei does this to some extent. Domino's does 100% of their deliveries like this, Jimmy Johns, even their Uber and DoorDash deliveries, they'll staff their own drivers to fulfill.

It's because they care that much about that quality and that experience of delivery that they can't get from these big delivery marketplaces. So we wanted to give a similar level of experience to every brand that's doing delivery, but in a framework that makes economic sense, because not everyone can staff W-2 drivers at their stores. And so we pre-position these vehicles at the restaurant locations.

And so when that food is ready, it comes out of the kitchen and it goes straight into any robot that is there. So this is like first in, first out from the kitchen. So rather than trying to match the right robot or the right courier with the right order, they can use any robot that's there. And it's kind of like a taxi line that's just pumping food out of the kitchen. And our average delivery time is 14 minutes, And we hit this about 99% of the time to the minute.

So it's extremely reliable and it's fast. And you remove any of these kind of compounding human errors with waiting for a driver to come pick up the order. And not to mention these days is everyone's trying to generate positive cash flow on their deliveries, these big delivery marketplaces are batching multiple orders. And so that kind of deteriorates the quality even further. Well, you can pay extra on top of your giant premium to get your delivery before everybody else, so they fix that for you.

So what's the distance? Like, how far can you be from the restaurant to get that 14-minute delivery? So we do about two miles. That's a driving, driving radius. And in L.A., which isn't super dense, it's fairly dense, but it's not like a Chicago or New York or anything like that. Two miles is greater than 70%, depends on the neighborhood, but it's great than 70% of all the orders. So it's a huge majority at that two-mile range.

And actually don't think that people, you know, at 7 p.m. when people ordering delivery, ordering dinner, actually don't think people really want to be ordering from that much further because you will, you will probably have cold food. So, but we limit the radius to where we can reliably deliver with that level of service. So, and then how often do they have to charge? They get charged once a day. So basically there's storage pods located in each of the neighborhoods we service. And so the robots will relocate themselves back to a storage pod at the end of every day.

And that storage pod shuts and locks. And then we have a field crew that will survey the whole city. And they do everything from like basic inspections. They'll do battery swaps. They'll bring back any vehicles that need to be repaired back to the warehouse. And then there's kind of your on-ground team to just make sure everything's running smoothly. And obviously the economics of this makes sense if you have a sufficient number of robots

that they're responsible for managing. Right. And then tell me about the drivers, the human operators. Like, do they need a license? Where are they located? Are they in the restaurant? And is it like a fancy remote control? Can I drive one?

Can I come on a field trip? Yeah, you can drive one. It's literally an Xbox controller. Awesome. And the pilots, the pilots customize them. They've got stickers on. They all have different colors. and camos.

We have pilots all over the world, actually. You can drive from home or using a computer. We do manage very directly all of the IT and the infrastructure required for this because there's a lot of video streaming. So, you know, we use multiple cell carriers to be streaming simultaneously to make sure that we never lose a video feed. So a lot of work has had to go into the whole reliability of the stack from the pilots end all the way down to the robot and hardware.

Just to reiterate, they could be anywhere in the country, anywhere in the city. I mean, so you're just using an internet, connection and you can hire remote pilots. Yeah, they can be anywhere in any time. Yeah. We do. Yeah, they can be anywhere in the world. They just need a reliable internet connection.

The further away you go, the higher the latency is because it just has to travel further. Right. So we have to be very thoughtful about that piece. But we want this to be a global, a global job, right? Like a lot of the difficulty of running an Uber or DoorDash is you have to, you have to staff, you have to match a very time sensitive, local demand to a very local driver supply, and that's a very hard balance to maintain, right? So we have a pretty massive advantage by having a centralized and global driver supply

that can operate across time zones, across cities, and that's a huge advantage. So we've invested heavily into the technology to allow us to do that reliably. And then what are the fees like compared to a company like Thorndasher Uber-Eats? When we work with merchants, we'll understand it depends on a lot of different factors. it depends like how they currently take orders. So if they have their own website and their own app that does a lot of deliveries, I mean, it can be about half the cost of what they did get charged from DoorDash or Uber for those deliveries. We try to save them, you know, tens of percentage points on their on their deliveries for any orders

we're fulfilling from Uber and DoorDash in the marketplaces. Doordash and Uber still take a margin from that delivery. So it's, it's, there's less, there's less room there, but that's also where a lot of the volume is. So we like to work with very large brands that have their own direct consumer presence because it's where our product can be very differentiated from both the quality perspective and the cost is a lot cheaper on that sort of a channel. And then how do you look at scaling? Are they, I assume they're on the sidewalk, right?

Primarily on the sidewalk. Primarily on the sidewalk. Can they use a bike lane or is that dangerous? They can if the sidewalks are blocked. like in LA, there's frequently neighborhoods that have poor infrastructure. Some have a better infrastructure. We actually have to map all of this. So we have details of the cellular sidewalk.

We have a lot of semantic information about every single sidewalk segment in LA. So we can make sure we can travel on the fastest routes possible. It's pretty rare. We usually use sidewalks. In states like Texas, we can use the roads as well, which is often needed because sidewalks will just disappear. But even then, we're going pretty low speeds and we're using the shoulders. How fast is it?

In Texas, we usually go, we would go like up to seven or eight miles an hour. Okay. Like a really fast runner. It's like a sprint, yeah. Yeah. So then how do you think about scaling to like a New York or D.C. Where the sidewalks are a lot more crowded? Yeah.

So there's kind of a, there's a balance here of as you get more crowded and like Chicago or D.C. or New York, you need to travel a lot less, a lot shorter distances to actually solve a huge part of the delivery problem. So you don't need to travel as far, so it's okay that your average travel speed is much slower because I do anticipate it would be, it would be a lot slower in those sort of markets. But L.A., you need to travel faster because you have to cover much larger distance than you would to do 70, 80 percent of deliveries. Right. you would in New York or Chicago.

Right. So you're saying you could really pick your way through some pedestrians in New York and still get there faster than maybe in L.A. because the distances are so short. Yeah. Like I imagine a lot of places in New York would actually be great. The areas you're going to move very, very, very slowly would be, you know, like Soho or downtown or something.

Probably most of those sidewalk segments are probably pretty busy. But there's plenty of neighborhoods where we basically map this out to look for underutilized sidewalk segments. So these are like the least commercial streets, basically. And so we map this out so we understand, okay, well, most of our businesses are located on these primary commercial roads. So we have to be there for a part of the segment, but we pretty quickly try to get off that street segment so that you can travel much faster. So this changes by hour of day, time of, you know, day of week. time of the year even. So we have to include all of that semantic information so we understand.

Like, for example, in Hollywood, Hollywood Boulevard is a fine place to drive down during sometimes the day. And then other times the day, it's like the last place you would ever want to drive. So it's helpful to know that information. And so we keep track of it, you know, based on all those factors as well. So when we look at unicorn and unit economics, hardware is tough. Hardware scares a lot of investors. How much does it cost to manufacture a cocoa robot and how do you make those unit economics work over time?

Yeah, that was something really important for us very early on is we wanted to, like, we pride ourselves a lot in being very practical as a business. And that was one of those things, like in the last, you know, two years, it was easy to feel a little silly if you have competitors raising billions of dollars to fund these kind of vehicles. But now, as the market kind of resets back to some reality, having like a 50 or $100,000 vehicle doing delivery, just doesn't. make any sense. Because your depreciation alone on those delivery is going to be more than you're ever going to charge. So it's very important for us early on to have a very low cost hardware. So these are a couple thousand dollars.

And in terms of the maintenance and upkeep that goes into them, it's usually plastic, right? So the most of the cost is in compute, like very ruggedized compute cellular modems, batteries. So these are pretty rugged equipment. So the lifetime on them is pretty long. So a lot of the upkeep's actually just in labor and, and plastic. It's really important to have the hardware be as low cost as possible.

And, you know, with that low-cost hardware, once you start saturating a neighborhood with these vehicles, I mean, they can pay themselves back in months. So that's really important as a hardware business also to understand the payback period on the hardware. This is, you know, on the surface, hardware is really tough to get off the ground, which I think is a huge testament to like our kind of luck in when we started this. Micromability just exploded. So, There's a very developed supply chain for making reliable outdoor power electronics that people funded billions of dollars into. That kind of category didn't really exist. And then now there's, you know, there's been a tremendous amount of growth and like, how do we get batteries and electric motors and aluminum strapped together and left outside for years?

So we were able to leverage a lot of those, that kind of supply chain to get us off the ground faster. But then, like, in a lot of ways, we're a little countercyclical because right now, like, no delivery companies are able to make much money. And that's not necessarily getting better. And these companies are all at scale. And so it's expected, like, when is they going to start generating cash on these deliveries? But it's tough. I mean, you're getting regulatory fees added on top of these deliveries from certain states. Minimum wage keeps going up. The expectation for drivers goes up. And as you start getting more and more demand, that level of driver supply need to maintain is higher and higher. So this is a really difficult equation to make work. And so we have this kind of. different cash flow cycle where we need to buy the actual assets and buy the hardware and develop the hardware. But the business is so powerful because how much it can reduce the actual cost

required to move something around a city by an order of magnitude. So from that perspective, I think it's very needed right now. I think a lot of businesses are looking at how do we actually deliver profitably or more profitably than before. I mean, it seems like, you know,

you're starting with food, but last mile delivery is basically the problem. I think it's, we looked it up, it's expected to be at least $123 billion market by 2030, which if I'm being honest, sounds a little low to me when you start to think about,

you know, all of the things that go into this concept of last mile delivery. Like, how do you see this business enabling delivery of all kinds of stuff? I mean, I'm already thinking, like, you have like a pharma plan, right?

Yeah, we, we want to be delivering everything. At the end of the day, once we put these vehicles out there, the goal is just to have them be moving all day, every day on revenue earning trips. The reason we started with food, and another reason why this would have been very difficult to start a couple years ago, is food is by far the highest frequency item being ordered. And it's by far the hardest thing to deliver, which also makes it the most expensive thing to deliver. So if we're starting with a company, we're saying we don't have any autonomy, we're having, but we have a lot of operations.

operational efficiencies that make this much lower cost than a human driver. You know, this was a natural starting point because we can make it profitable. We can solve a lot of these horrible quality problems with food delivery. And then you can start using that footprint to start expanding into these other categories. I mean, food only happens during this like, you have this little peak at lunch and they have a big dinner peak and it kind of trails off throughout the rest of the day. So you have a lot, you have large periods of underutilization of the asset.

So you absolutely want to be delivering for anyone that, even if it's higher frequency, you want to be delivering for anyone that's off-peat. But it's very difficult to start there because the volume doesn't make up for it. And again, like five years ago, even food was nowhere near the volume to make this sort of a business makes sense. So it's grown so large and so high frequency that we can actually run a, you know, a profitable hardware business with just food and then expand from there. It's pretty exciting.

I mean, not only do I want to get all my food this way, but I want this to be like my mailman, you know. Take those places for me. To what extent do you think of this as also a climate solution and a potentially pretty powerful one? Yeah. A lot of the conversations we have with cities is something they get really excited about.

One of the reasons that so many cities have been so supportive of this is, you know, you have economies like in New York where all the deliveries for the most part being done with people and bikes, but in a city like L.A., that's not happening. Um, it's all car based. And, and this kind of statistic I shared about how many deliveries are happening

within a couple miles, it's pretty disturbing if you think about it because you're like, we're moving in this direction as a society where we want to get everything delivered and we want it all delivered quickly because of how high that, uh, convenience value is for people. But each of those trips is adding a car to your neighborhood driving around. So,

and you're transporting like a couple pounds of goods with, with the, you know, with the gas powered car. Yeah. Um, so that, That doesn't, that's not like a super inspiring future for, for our cities.

And so, um, there's a couple ways in which we're helping there is obviously removing cars from the road, um, but also we're transporting, we're using a much more appropriate vehicle to be transporting this type of, this type of food, right? So if you're transporting a couple pounds of food, we're doing that with like an 80 pound robot. So it's, it's much more energy efficient. For God's sake. Yeah. Exactly.

So it's much more energy efficient and it's electric. So I think that was like one of the original reasons going back to like when I saw a robot for the first time. I was like, this has to be how the future is going to work because the future of couriers and cars driving around your neighborhood just did not seem sustainable or did not seem like a very inspiring. Inspiring future. So that's definitely that's definitely a big area for us to be proud of when we see more and more deliveries getting done by these vehicles. Yeah, absolutely. I mean, zero emissions, deliveries, and reducing congestion, right? Setting aside just the kind of climate part of that, just getting cars off the road seems like a huge deal in these very dense cities where this is feasible.

Yeah, exactly. What has been really hard about this? Like, was there something that you didn't anticipate? I can't imagine you didn't have some supply chain problems. Yeah. Hardware is hard. I think everyone says that. I'd heard that a lot when we started this.

And, you know, you're kind of like, okay, well, we'll figure out how to solve it. But the reasons it's harder, you basically, like in our sort of business, you constantly have this tradeoff of CAPX and OPEX, like, do we want to keep retrofitting the vehicles as we learn more, right? Because if you look, if you, you know, if you signed to one of our product meetings every week, we talk about, everything we talk about is basis points. We're talking about like single basis point problems. Because if you have one delivery out of.

thousands in this neighborhood for this week, that goes, you know, 15, 20 minutes late because the vehicle had a breakdown needs to be restarted or you need to send someone to go and to go and restart it or whatever might have happened. That customer is going to be very upset because you ruined their dinner and you ruined their day and it doesn't matter at that point that 99.9% of your deliveries were much better than a courier. For that person, they, they hate your robots and they hate your service. And food is a very like emotional thing. When you order, especially if you have family, right, you're compounding the anger across all of your family members. So it's a tough, you know, I've, I've been there. It's, you know, when you're hungry and you want to eat and the food just doesn't get there on time or if it comes cold, especially when you're paying so much, right? It's frustrating. So you have this tradeoff of like how much more kind of operational expense do you put into this to make sure that those small corner cases get dealt with by sending a human to recover the delivery or, or constantly like, you know, maintaining the fleet to make sure everything's perfect all the time versus more long.

term fixes with the hardware. And, you know, your understanding the problem grows dramatically. Like I said, we're only two years old. So you learn a lot by doing, which is why I feel like actually doing real deliveries and operating is so important because you wouldn't know any of this stuff. Yeah. You know, I wouldn't have guessed any of these things that we're working on. But that's tough, especially in this market where you're trying to be as efficient as possible

with all of your capital. And, you know, there's always that tradeoff of how much do you invest? back into research and development on the robots and a new generation of robots. And it would have been much, you know, much easier to do this in the kind of last decade. I mean, you saw a lot of the micro mobility companies, right? Their philosophy was just like, we'll ship a new vehicle out every couple months. And it doesn't matter how fast those are the ones to appreciate because investors are going to care about the run rate of the business.

But that's, yeah, that's tough in this sort of a market. It's not very efficient. It's not a very efficient use of capital. So that's, that's a tricky part about the business. business that you wouldn't have to deal with if you're for using couriers, but that's where a lot of the complexity comes from. And like I said, food delivery, the reason we're starting there is because it's so hard to do, but that means you do as a startup this early, you do have to care about every basis point of problems. So that's like a level of detail that I imagine most companies don't

need to do so early on. But I think that's like our core competency and that's what we've been really good at. It's just really caring about the details and having an incredible, reliable quality service. And then talk to me about competition. We actually found out about Coco. One of our producers found out about Coco after reading about Amazon sun setting a similar project focused on home delivery, the robot called Scout, which was, it's like a cooler side. I mean, it looks really similar. When you hear about Amazon shutting down a competitor, are you like, oh no, or great, you know, clear the sidewalks out for us. What they were working on was really cool.

And they had a really cool approach to it. But if you think about how efficient, it just kind of goes back to, like, again, like while we're starting with food, if you try to do this with package delivery, the problem is much harder. You need to get your cost per delivery really, really low for that to make any sense for a company like Amazon. on. Like your hardware costs, and these things go against each other a lot, right?

Your hardware cost needs to be really low, but the thing needs to be full autonomous because any labor you put into the system, you're going to start losing the efficiency of an Amazon driver who's just covering a whole neighborhood time insensitively, right? Drop off the package by the end of the day and you cover this section of the city. That is far more efficient than a courier saying, okay, go to any arbitrary point A and then any arbitrary point B with food and make sure you do that in a way where the food's arriving hot. So that's a much harder problem, which makes it much more expensive. So, you know, I think with Amazon, I think it's one of those problems of like just looking forward

of what is the time horizon of this tech working. And, you know, I think they're probably revealing through this decision that they think it's pretty far out. And I think there's been a couple articles recently just about self-driving cars and I think it goes through these like cyclical hype cycles. But it's a hard problem to predict because it's all about solving a massive long tail of corner cases

and then kind of like a societal buy-in to when is it ready. So we wanted to be free from that constraint. We wanted to make sure that the business can operate profitably and can solve a problem for a customer on day one without any autonomy. And if you're able to do that, then long term you should be the best position to win in.

the autonomy race as well. So since we started this, we thought that approach wouldn't work unless you have a company willing to spend on R&D for an unknown time horizon. Right, which might be Amazon, but you'll be hopefully well-entrenched since then. I mean, it's fascinating that really like your big unlock was this can be remote control. It does not have to be autonomous. and then Amazon and several other companies, it sounds like we're trying,

Amazon was trying to solve two hard problems, package delivery unit economics and autonomy. And then other companies who have competed in the space have tried to solve this like, so far somewhat unsolvable problem about autonomy. Like, how do you think about your roadmap to autonomy and how necessary is it? Yeah, for, the way to look at it is for food delivery,

it's not necessary immediately, but what we care a lot about is once you start getting at some scale in these markets, you care a lot about the incremental cost of a new delivery being added on top of that fleet. So, you know, at first, when you're starting to cover restaurants in a neighborhood, the kind of the costs being distributed across all your deliveries for the warehouse and the maintenance team and field operations and all the charging and depreciation of fleet and stuff, right? those add up to being a pretty similar cost to what you pay pilots. They're pretty, that can be pretty expensive.

But as you start getting saturation, the ultimate goal for us is to have new types of businesses that don't deliver today or that don't deliver the same frequency as food, have it be economical for them to start delivering and having being economical for the consumers in that neighborhood to start ordering from there instead of something like Amazon, where everything's kind of centrally optimized. when a lot of those goods are already in your neighborhood. It's just cost prohibitive today. And so we want, you know, once you have a profitable fleet in a neighborhood,

you want the cost of a new business coming on to be really low. So that's going to require this incremental process of making pilots as efficient as possible of managing multiple vehicles. And we want that cost to eventually get significantly below a dollar because I think that's how you fundamentally reshape the way local commerce works. And to do that, you're going to need some. form of autonomy. So, but if you have this profitable fleet and you continue to grow it, you're,

you're collecting a lot of real world data. You have an expert in the loop doing the deliveries, right? Your pilot is a trained pilot so they can help monitor your autonomy system. They can help label all of your data during periods of underutilization. They're an expert, they're like an expert in the system that is already baked into your economics of doing delivery. It doesn't come off as a new R&D line item. And if you look at how companies like Waymo would operate

is they have these large teams that need to go out and collect that data to validate the algorithms and you'll see engineers sitting in the cars with their laptops, right? That adds just a ton of operating costs to, on top of all the R&D you're doing. So having that baked into your economics, I think, is really powerful. What's next for you guys? Do you want to make any news about a new city here today

or tell us what's in the roadmap? Yeah, well, we have a couple new cities coming up next year. And you asked about competition. Like we, you know, I think in the beginning of this year, we thought our future was probably going to be rapidly expanding with some large enterprise brands. And we felt like new companies were popping up left and right and we needed to go expand really quickly. I think it's much harder to start like a hardware and delivery company today than it probably was a year ago just because they're notoriously unprofitable. and hardware is expensive. So right now, this has given us kind of the breathing room to say,

we're going to take the markets we are operating in, and we want to make them generate, actually generate positive cash flow, which I don't know if any company's ever generated positive cash flow on delivery in a city like Los Angeles. Don't, or maybe anywhere. Or maybe anywhere. I have a little bit of goose.

I was around for the first dot-com bus, so I have some like goosebumps hearing you talk about a profitable food delivery business. I was like, what? Yeah, so it can, it can definitely,

it can definitely happen. This is just a good environment to be really disciplined about that, and that's what we're doing. So we're just really focused on saturating the neighborhoods we're in, and if you live in a neighborhood where we deliver to, I mean, you feel our presence.

We're doing most of the deliveries in these neighborhoods. And we really want to show that this, like, delivery, and that, this 15-minute quick delivery can't actually be profitable, but the paradigm that everyone's been pushing for the last decade It doesn't really work in a state like California or state like New York. I think it can be more successful elsewhere, but it's really tough to get those economics to work in these sort of environments.

Zach Rash is the CEO of Coco. We're calling it now next unicorn. Zach, thank you so much for the time today. I really appreciate it. Yeah, thank you, Molly.