Technology Archives

Uber’s Long Drive Into Google’s Self-Driving Moonshot

Tiger Moonshot

Google just reorganized its entire corporate structure, putting the business “Google” we all know into a bigger company, Alphabet, which will be the holding company of a number of other Google business lines and projects. Some observers have remarked that this will instill more discipline within all of Google’s famous “Moonshot projects” to be judged somewhat independently under Alphabet’s CEO Larry Page’s watchful eye. Certainly, on paper, it sounds like a more logical structure with which to manage a growing behemoth like Google and all of those Google X and moonshot projects.

I don’t know all the moonshots are, but my first reaction in hearing the news was to think about Uber (where Google invested) and the specter of self-driving cars. I had figured out of all of its moonshots, self-driving technologies may be the one that’s ready for a huge market next. Uber’s rise in part was made possible by leveraging Google’s mobile platform (Android), and across platforms, by leveraging Google’s Maps product and API. Now, as Uber is growing like a weed all over the world, could it be that Uber, just a few years ago considered a “startup,” have taken investment from Google and, in that time, wedged into the most market-ready moonshot project and built a network on top of it (leveraging Google Maps, to boot)?


As I was saying, I don’t know what all the moonshot projects are, but luckily someone sleuthed this and came to an interesting conclusion: That Google X moonshot projects very much have a mixed record. The USA Today’s Jessica Guynn wrote a piece with the headline “Google’s Mixed Record On Moonshots Means Alphabet Could Flunk,” revealing some long-term concerns about the likelihood of a second act as elegant as PageRank. Guynn writes:

Google has been under pressure to look to the future and to the next big moneymaker. The Internet giant relies on its main advertising business which makes nearly 90% of the company’s total revenue. That business is still growing, but it’s growing more slowly — and it’s facing rising competition from Facebook and others.(Google shares closed up 4% to $660.78 on Tuesday as Wall Street digested the Alphabet news.)

When Guynn wrote that Google “relies” on its search advertising business, it reminded me of an old post by Chris Dixon, who asked “What’s Strategic For Google?,” making his point succinctly:

Google makes 99% of their revenue selling text ads for things like airplane tickets, dvd players, and malpractice lawyers. A project is strategic for Google if it affects what sits between the person clicking on an ad and the company paying for the ad…At each layer, Google either wants to dominate it or commoditize it.

If something is strategic to Google, he argues, they have strong incentives to dominate it or let it get commoditized. But, that’s for their core business, and even though the stock has been surging this summer, the best tech analysts in the word raise valid concerns. Ben Thompson of Stratechery came out with a strong post earlier this year, “Peak Google,” subtly arguing that the company has peaked in its dominance:

Google is quite safe when it comes to search, and that they will be a very profitable company for the foreseeable future. I just suspect we will all think differently about that dominance when it’s a small percentage of total digital advertising, just as we thought differently about IBM’s dominance of mainframes in the age of the PC, or Microsoft’s dominance of PCs in the age of the smartphone.


If you believe in Peak Google, and if you believe Google will try to dominate whatever is strategic to them, and if you believe that out of all the moonshots in their portfolio, the next one in line will be self-driving technologies, a curious thing may have happened along the way: A little startup from San Francisco may have leveraged Apple’s (then) new mobile platform and Google’s own mapping technologies to create a network within which they could acquire and own the customer relationship (Uber). Eventually, of course, Ubers will be self-driving, and while it may have been logical to assume that technology would’ve been supplied by Google, it now seems like Uber not only doesn’t want a part of that, it’s also that (to a degree) technologies related to autonomous navigation have started to commoditize a bit. This may put Google on the short end of the “dominate or commoditize” framework — after years of doing it to others, Uber’s wedge into their moonshot portfolio might turn around and do it Google.

Weekly Curated Email Newsletter About Uber

As many of you know, I blog about Uber a lot. It’s a fascinating company. I have no financial interest in or connection to the company or any investors. I just think the company is fascinating to follow, so much so that I thought I’d write a book about it, but then life happened, and then I said I would can the book, and, well, I’ve been thinking so much about it (and writing and tweeting more about it), so with the help of a friend, decided the best compromise was to just do a simple, weekly, curated email of links with some commentary about each item. This is really for me to keep tabs on what’s going on, and if you find the Uber trajectory interesting as well, you can follow along, too. Here’s the email sign-up link:

Corrective Medicine, Just What The Doctor Ordered

As it relates to startups (and investing in them), the past week’s stock market correction (which still may be going) is just what the doctor ordered. Who knows why it happened exactly — sending a message to the Fed, international market fears, people realizing P/Es were too high, blah blah blah….

The net effect now after some stability is the correction was medicine everyone needed. More specifically, I think the correction and intense social & traditional media focus on it actually makes this better for everyone in the startup ecosystem:

First, writers, analysts, bloggers, and arm chair Twitter economists now have more to write about that timely, global, and more hard-nosed. They can go up to a founder or investor and simply ask “Well, did you crap your pants?” The press has been somewhat reluctant to balance cheerleading entrepreneurship with asking key fundamental questions.

Second, investors can now leverage recent market gyrations to negotiate down valuations. For years, investors couldn’t do this at the risk of losing a deal or offending a founder, but now everyone understands that a bit more balanced has been restored to the ecosystem, and investors (may) get some lower prices.

And, Third, founders avoided catastrophe. If the market kept sliding, many investors would’ve been fine (with bruised portfolio metrics), and writers/bloggers would’ve had a field day, but entrepreneurs, founders, and very early-stage employees (and frankly many rank and file) would’ve been in a real jam. Runways could’ve started to compress. People could’ve jumped ship to work at a safer job. This jolt was a nice reality check and doesn’t seem to affect the long-term positive outlook for technology seeping into the world.

For these three reasons, the recent market volatility may just have been exactly what the doctor ordered.

We’re Finally Learning More About Uber’s Platform Strategy


In the SF/Valley bubble we live in, we are finally learning more about Uber. For years, we have debated: Will the on-demand industry consolidate? As part of that eventual consolidation, what will Uber’s role be? And, if and when they expand the platform from rides to X, Y, and Z, will those services be branded as Uber’s or a sub-brand? Will they be housed inside Uber’s core app, or as a separate app in a constellation of interrelated apps? (Read this old post on these questions vis a vis Uber.)

Of course, Uber could change in the future, but just in the last few days, we have learned a few things: That Uber may elect to build things themselves (because they can execute) versus paying higher prices and going through the trouble of integrating; that Uber will likely place more buttons on the top nav bar in the app (see picture), perhaps even an integration with mega-popular dating app Tinder (which makes sense); and that those experiences will be managed inside the main Uber app container, controlling the brand experience (and checkout, payment) for the user.

As I wrote a few days ago re: UberEATS launching in San Francisco — which is noteworthy given this is also the epicenter of (too?) many food-delivery startups, many of them richly backed by venture capital. Not since Facebook have we seen a company execute at the level Uber has. They react to market opportunities and PR crises with incredible speed and precision. Now, they’re bringing the Uber brand to you in more ways than one.

Many people by now have heard the line that “Uber wants to have a second business line by the time it IPOs.” It remains to be seen if food will move the needle for them, as this is more of a lead-gen model in the likes of Postmates, DoorDash, and Caviar (powered by Square) versus the full operational costs of an integrated model like Sprig or Munchery. Beyond the “food button” on their app, it’s fun to imagine what other buttons we might envision. Beyond food and transport, it’s hard to imagine what is needed on a daily basis and in the moment — though the suggestion of Tinder is appealing, and one can see the connection from early Uber backer Benchmark Capital, which recently became an investor in Tinder (with Matt Cohler joining the BoD). Maybe after that, they open a store, or provide sundries, or anything you’d order from Amazon, or Target, or Whole Foods?

Facebook scooped up Instagram prior to their public offering during a time when pre-IPO concerns included the company’s mobile exposure. For Uber, which should now go public in 2016, the question to ask is: What is their Instagram? What is the move or market needed to show long-term public investors the promise of the underlying platform beyond transportation? And, will it be acquired or built from within? I don’t know what the answers are, but I do believe these are the questions to ask.

[As a post-script, you may tell I’m back talking about Uber more. For years, I’ve been fascinated by how this company intersects with our physical world. I was going to write a book about this, and then canned the idea, but I’m going to pick up the idea again in some form — maybe not a book, but just more posts about it. I hope you don’t mind.]

The Chilly Freeze For On-Demand Startups

We will talk about this topic among others at The On Demand Conference in NYC on September 15, Register Here. We are also running our second contest for the best B2B on-demand startup, more details here!

My, how things can change. When we did the inaugural On-Demand Conference back in May 2015, everything was hot. Now, gearing up to the second installment in September in NYC, things are — shall we say — a wee bit chillier. And, rightfully so. In a few months, we’ve seen high-profile companies broadly in the sector nearly fold overnight, we have yet to see the big rollup consolidation strategies that people wish would happen, and the culture of copycat models is bringing more and more founders into red ocean markets like food delivery.

Investors are applying more scrutiny toward on-demand startups, asking flavors of the following questions: How frequent is the consumer use case? What is the payback period when opening a new geography? What do the unit economics look like, and can they modeled at scale? Does the team understand they should balance unit economics and growth by city to make sure they don’t run out of money? We haven’t seen much M&A in sector, so can this company go the distance? How can the team expand the product/service offering if the service grows? With UberEATS now in San Francisco, the backyard to many well-funded food delivery apps, could Uber crush the opportunity entirely? And, who might follow into the next round of funding should the team set out to do what they want to do?

These are MUCH harder questions than VCs were asking before. Welcome to the new world. Now, that’s not to say there won’t be some companies which continue to fetch large rounds, or that there won’t be on-demand concepts that attract funding in different parts of the world or in different product categories (like healthcare). The concept is bigger than food and transportation, though those two are clearly the consumer drivers.

Luckily for us, the purpose of these conferences isn’t to promote a sector, but rather to gather the most thoughtful minds among company builders and investors to tackle the most pressing issues of the day. So, in September in NYC, that will be no different. You can learn more about the NYC event on September 15 here, and on the VC panel in particular, friends Steve Schlafman (RRE), Kanyi Maqubela (Collaborative), Anu Duggal (Female Founders Fund), and David Tisch (BoxGroup) will debate these and other issues facing on-demand startups in the seed and Series A/B stages.

Register here for the September event in NYC. Unfortunately, I am unable to attend in September now, as I was very much looking forward to seeing everyone, meeting all the new founders with new concepts, and kicking off the day’s activities with a fireside chat with USV’s Albert Wenger, but I can’t make this trip. So, I will give away my ticket to one person who really wants to attend. What I ask for is this — tweet out a link to this event and mention me @Semil on twitter, and my co-organizers will pick one lucky winner to attend the event in my place!

Reimagining The Restaurant With Robotics

I grew up working in restaurants. I have been a dishwasher, host, bartender, waiter, line cook, and regular cook. I’ve delivered food to people, and then cleaned those tables. But, that was a long time ago. Then, last week, I went to YC Demo Day, and there were food-related robots, and it got me thinking about how the last few years has seen all sorts of innovation in modernizing or entirely replacing the concept of restaurant. This is important because restaurants provide a space and service that people can use many times a day, and critically because it employs many people, both full-time and for those looking for extra income to make ends meet.

Briefly, here are some ways entrepreneurs are reimagining the restaurant experience.

Mobile, web lead-gen: Companies like DoorDash, Postmates, OrderAhead, Caviar (inside Square) and others are funneling more orders to restaurants, and fulfilling delivery either with their own staff or via 3rd party logistics (3PL) providers (more on those below). Like Grub and Seamless, using mobile for call avoidance to order and schedule with a few taps on our phones. (Oh, and now UberEATS is in the game!)

Vertically-integrated: Companies like Sprig & Munchery make and deliver their own food, bypassing the restaurant experience entirely, but getting fresh food to customers. This allows them to control their COGS and potentially have better contribution margins, though it takes longer to rollout to new cities — that’s one big tradeoff.

3PLs: Companies like Homer, Deliv, and Sidecar offered delivery services to other businesses.

Decentralized, peer to peer: Companies like Josephine Meals envisions a future where people become home-cook entrepreneurs (a la Etsy) and sell directly to people in their neighborhood…and they may even use 3PLs to power delivery.

New Disruptors: There are too many list to here, but just when you think food is saturated, companies like Tapingo, EnvoyNow, and Fooze (among many others – and please don’t ping me about them!) are finding markets to attack.

Franchise Model: Like Domino’s Pizza, a franchise network perhaps isn’t a terrible idea to get around the cost of expansion. A company in this recently YC batch (Wheely’s) wants to help others create mobile coffee shops.

Robots and Automation: Now, this is why I wrote the post. Restaurants and the food service industry employs LOTS of people. But, now, Robots are coming! TeaBot, coffee bots, sandwich robots, and other robots or automated devices (like e la Carte). KQED had a feature on Eatsa in San Francisco, an automated restaurant. Next time you’re in the coffee shop, count how many employees are there. Probably 7-10 per 500 square feet. Why? We are used to it, but it’s entirely possible for the next Blue Bottle to just be a station of iPads and robots making your stuff. Zero human interaction. The only question is — how long will it take to come? Based on what I’ve seen this year, perhaps sooner than we all think.


Seed Founders, Beware The Syndicate Shenanigans

As any reader of this blog would know, I am a huge AngelList fan. I have run a few Syndicates now over the past two years. Most of them have cleared, but a few haven’t. I could’ve run a lot more, but I intentionally elected to be very careful about how and when I bring syndicate opportunities to the platform (1) because I’m simply new to investing overall, and I didn’t want to bring people into the fold with me too early; and (2) because I needed to learn AngelList’s platform myself, over time, and learn in between each syndicate to get better.

When I present the AngelList option to a founder, I always educate them first on all of their options for their fundraise. I like to lay out the options and let the CEO decide. Many times, the CEO has elected to forego the AngelList option, citing either privacy concerns, or signaling concerns, or wanting to wait; at other times, they dive right in, want to learn the process, and have so far been very happy.

Lately however, and why I’m writing this post, I’m encountering investors in the angel and seed stages who are up to some tricks, what I like to call “The Syndicate Shenanigans.” The classic move goes something like this, with the investor offering to the founder: “I would like to invest X, but conditional upon me taking this to AngelList and putting it in my Syndicate for 4-6x.” The founder in this case is faced with a dilemma, and this post is targeted both at the founder and also those who participate and follow Syndicates on AngelList. To be clear, these can work well if done right, but it’s smart to understand the risks before jumping into the fray:

To Those Who Follow Syndicates: First, determine if you want to automatically follow any deal of a Syndicate leader, or whether you want to pick deal by deal. If you blindly follow, your money gets priority, so just know you’re along for the ride; if you go deal-by-deal, try to understand why the Syndicate lead or founder wasn’t able to fill up the round privately. Second, generally speaking, you may want to monitor the deal pace at which the Syndicate lead is bringing opportunities to the platform.

To Founders Offered Such Terms: This may put you in a tough spot. You want the money, but the conditions attached to it could be confusing. Sometimes investors will say they have a “minimum check size” to make the math work in their own fund model, so when an investor offers to commit with the condition of bolting on a Syndicate, this is another flavor of the line: “We have a minimum check.” In this case, they likely have an economic arrangement on AngelList that gives them a percentage of carried interest on top of the money they collect for you. You have some options here. First, learn about AngelList and Syndicates, and make sure you know whether the particular Syndicate offered to you would be private or public; understand the process and how long it could take (2-4 weeks to clear); and understand how the Syndicate lead (who will be named on your cap table directly) will market your company to those other investors that you’ll never meet.

I want to be clear that (1) I love AngelList and (2) have done this myself, but (3) that making AngelList a condition of an early-stage check isn’t bad behavior, but should be done so in a way that educates the founder about his or her option, and less about throwing down a conditional on the team. Happy to answer any questions you may have and/or refer you to CEOs I’ve worked with who’ve gone through the platform for a Syndicate.

The Confluence Of Trends Making esports Go Mainstream

Like many others nowadays, I’m fascinated by the phenomena around “esports,” which can mean watching other people play video games or code or do both, either in person, or online, or on one’s phone. I had to laugh when Startup L Jackson tweeted “If one more VC discovers eSports this week & starts tweeting about it, I am going to come up with an ultimatum. And it will not be pleasant,” because it’s true…but there is good reason. I was catching up on some reading today and realized there are actually a confluence of non-connected trends which form to make the present day fascination quite logical:

One, the youngest millennials are now right in the middle of their teenage years, right around when they can almost drive a car. This generation entered the world when mainstream gaming consoles hit the market and started to invade the home, with the Nintendo Entertainment Systems coming to the U.S. in 1986, the Sega Genesis in 1989, Sony’s Playstation in 1994, and the XBox in 2001. Like I did, many folks owned or tried every system, and they popularized casual video game play.

Two, the culture around these home gaming systems is that only 2-4 people could play at a time. If you grew up around these consoles, you would clearly remember waiting your turn, and while you waited, you would watch other people play the same game you wanted to play. You’d just sit there. You didn’t have a phone to distract you or the Internet at home to jump on. Turns out “watching other people play video games” is randomly engrained in millennial muscle memory because everyone exhibited the behavior. This turned out to be inherently social behavior, waiting in tournaments to play the next round of EA Sports Madden or NHL, or trying to win Zelda or Metroid.

Three, over half of millennials grew up in an age where network TV was on the decline. Today, for many of them, network TV is wholly irrelevant. Instead, they stream content online, pay for subscriptions, buy individual shows, and watch shows on-demand. While mobile phones and networks take the place (in terms of attention) of TV, there are hours in the day folks can devote to this and they certainly do.

Four, the infrastructure required to support a range of esports activities has become robust enough to handle the traffic. People can now broadcast and/or consume esports media from home, networks like and others have cropped up and amassed large audiences. Mobile phones, data networks, and video channels are also more robust, of course. MLB recently announced it was spinning out its quietly robust digital streaming and hosting business, a unit which could be worth many billions of dollars more than it is today. As livestreaming technologies have demonstrated this year alone, social networks like Twitter are ready to provide users with live content (a la Periscope) and to support those streams as a piece of audience development and engagement.

Five, the cultural interconnectedness from a tech point of view between Asia and the U.S. is also responsible for the trend hitting the west. The culture of watching others play video games in larger audiences began in parts of Asia, which made sense because that’s where the home consoles were originally developed and released, before shipping to the west. Like messaging platforms today, the esports behavior is older and just took a bit longer to incubate here in the west, but now these micro-cultures have converged.

Sixth, the global rise in the importance of and celebrity around computer literacy is timed in such a way that it doesn’t make the act of participating in esports seem like a waste of time. Rather, it can be seen as both educational and social, as both entertainment and interactive.

And, there you have it, all of these forces put into a blender, and it all explains why esports is so popular, why sports and media networks are not only allocating mindshare and budgets to it, but also doing this for derivate products, such as gaming networks which let fans bet on esports participants, among other parlays. It’s been almost a year since Microsoft acquired Minecraft and when Amazon acquired Twitch, and with most technology investors being a bit older than the millennial generation, most of them have enough nostalgia to understand the behavior and recognize the power in the mass audience aggregation when they see events like this one. Like Facebook and Instagram and Snapchat aggregate audience attention on the web and on phones, esports is starting to do the same thing in the same channels, but also in real life, where kids of all ages collide, compete, and make new friends. It’s almost like an entirely new social network, and that’s what gets people excited — and rightfully so.

The UberEATS Test Arrives In San Francisco

We will talk about this topic among others at The On Demand Conference in NYC on September 15, Register Here.

Uber has been testing its food delivery service, UberEATS, in a few markets. Now, they’re gearing up to launch it in San Francisco, the current home of all sorts of food delivery startups. It is the land of “convenience tech” — and, as some would argue, too many of them are running around.

In the time that Uber started exploding around the world and this week, when UberEATS is set to launch, a whole range of food-related startups have sprung up. Food, like transport, presents the ultimate “daily active use” case, and founders (and investors) have been anything but shy in pursuing those channels.

Over the past few years, one of the biggest debates in the on-demand sector centered around centralized or federated apps — would consumers prefer to (a) use single-purpose and single-branded apps to execute certain functions; or would they prefer to (b) have everything ordered and fulfilled through one app that ruled them all?

An informed rationalist could argue (a) or (b) and make sense. The problem with (a), however, is each company then would have to fight for distribution and fight a CAC war to keep going; the problem with (b) is would consumers have the mental model to go to the Uber app for things like a salad, and would the platform provide enough choice for consumers, given that so much about food is choice and variety?

An initial glance at UberEATS in SF (click here) shows a scheduled menu with limited choices, reasonable prices, and a “reminder” button. The visual design reminds me of a mix between Sprig and Munchery, and that’s likely intentional (and smart). While companies like Sprig and Munchery have to do their work while also acquiring new customers, theory says Uber can “route” their active users to this behavior and piggyback on the behavior change pioneered by these two startups. We will have to see if it actually happens.

Additionally, companies like Postmates, DoorDash, OrderAhead, and Caviar (now part of Square) are trying to provide leads to restaurants, paired with some component of on-demand or scheduled delivery. These companies offer more variety to their consumers (relatively speaking), but don’t have the luxury on a unit-economics basis to control their cost of goods sold and, therefore, their margins, in the same way that Sprig and Munchery could. This is critical to understand because almost certainly UberEATS will not be a centralized model and therefore, could be a loss-leader for quite some time…unless their delivery platform is even more efficient than what exists today.

I do not pretend to know how this will all shake out. In a way, Uber is essentially a big incumbent company that just happens to be private, yet it also an execution machine, showing time and time again it can roll out software, services, and products to compete with breadth and depth. We will eventually, finally, get to answer so many lingering questions: Will Uber acquire a big startup before going public? Will Uber extend into other on-demand verticals on top of its platform? And, if so, will those be done in one app container, or federated into a family of apps branded separately but connected within a holding company (a la Google and Alphabet)?

Moving Stress And Mobile App Medicine

After a little over 5 years living in the same little apartment, we finally moved. If you’ve been following on Twitter, I’m sure you’ve seen — it’s a painful process and when you have a lovely spouse (who is also a packrat) and a toddler in the mix, it’s downright stressful. I tried to manage as much of it I could up front, but in the hours and moments leading up to the final day, there was little control to be had. The process is just bigger than most individuals.

I ended up using a lot of local and on-demand services during the move, so I thought I’d list them here:

MoversDelancy Street Movers. They were recommended 3-5x over by our friends on Facebook. We, too, had a nice experience and they were fast but also patient.

Food — No surprise, but I used Instacart and DoorDash a lot the first few days of transition. With a toddler around and in the middle of work, there was little time to cook or shop. We are finally back to a routine of cooking, but I’ve always felt on-demand services for us was less about opulent convenience and more about providing a support network that many folks have who live close to family or friends.

Labor — I used TaskRabbit a lot. Great experiences. I didn’t realize TR charged 30% rake on the transaction, but then lowers the rake % if you keep hiring the same rabbit. I get why they do it, and every rabbit we used liked the service and wanted to be hired through the interface to earn reviews and keep their platform bosses happy.

But the big winner for me was Move Loot. We need to get ride of old furniture and needed a lot of new furniture, but I had no time or patience to actually order online and deal with shipping and assembly, or to go to a physical store. I just wanted it to be easy and was happy to buy second-hand stuff, especially living with a small criminal (I mean, child). When you buy on Move Loot, it strips away the nearly every pain point associated with furniture. You can browse on your phone or the web. You can get alerts when things hit the market that you want. The app is easy to navigate. You can pay for stuff, schedule delivery, and checkout with a few taps. Most stuff comes assembled already. The drivers come and deliver the new stuff, and then they rip off the plastic protection, and put it where you want the stuff. Then you tip them via the app, and they’re off. Very friendly. I didn’t have to go to IKEA and then hire someone to lug the stuff back for me, or to pay someone to spend 2 hours putting it together. It was just done and scheduled.

I have no connection to Move Loot and don’t know much about the company, but it solved a problem for me many times and I’m sure relieved me of tons of additional stress that would’ve arrived otherwise.

Haywire is written by Semil Shah, and is published under a Creative Commons BY-NC-SA license. Copyright © 2015 Semil Shah.

“I write this not for the many, but for you; each of us is enough of an audience for the other.”— Epicurus