What do a blockchain-powered toothbrush, an AI that predicts your dog's mood, and a subscription service for bottled air have in common? They all raised millions of dollars before vanishing into the startup graveyard. Today's prompt from Daniel is about ten delightfully ridiculous flash-in-the-pan startups that either never got going or went bust in spectacular fashion.
Herman Poppleberry here. This is a fantastic prompt because we are currently living through a period in twenty twenty-six where the venture capital world is finally cooling off from the AI gold rush, and we're seeing these same patterns repeat. It is the perfect time to look back at the wreckage of the last decade to see what we can learn. By the way, today's episode is powered by Google Gemini three Flash.
It is funny you mention the AI gold rush, because some of these older failures feel remarkably modern in their hubris. We are going to look at some case studies in misreading markets, over-engineering solutions, and essentially ignoring basic human needs in favor of "disruption." Herman, what's our first victim on the slab?
We have to start with the king of Silicon Valley excess: Juicero. This was a company that raised one hundred twenty million dollars from top-tier investors like Google Ventures. The product was a seven hundred dollar Wi-Fi connected juicer that used proprietary produce packs.
I remember this one. It looked like a stormtrooper's lunchbox. And the whole pitch was that it applied four tons of pressure—enough to lift two Teslas—to squeeze these bags of organic pulp.
That is the technical absurdity of it. They spent years and millions on research and development to create a custom-engineered press. But then, a Bloomberg reporter discovered that you could just squeeze the bags with your bare hands and get the same amount of juice in roughly the same amount of time.
That is the ultimate "solution looking for a problem" moment. You have engineers obsessing over the mechanical stress of a custom-milled aluminum press, while a human hand provides perfectly adequate torque for free. Did the Wi-Fi at least do something useful?
It checked the expiration date on the bag. If the bag was expired, the machine would refuse to squeeze it. It was essentially a DRM-protected juice box. This is a classic example of over-engineering a simple process until it becomes a burden. They forgot that the "user experience" of squeezing a bag is already solved by nature.
But wait, how did the investors not see this coming? If you’re Google Ventures, aren’t you doing a "squeeze test" before you hand over millions?
You would think so, but they were sold on the "platform play." The idea wasn't just the juice; it was the data and the subscription model. They wanted to be the "Nespresso of Juice." In the VC world, if you can convince someone you’re a "platform" rather than a "product," the valuation triples. They ignored the hardware reality because the software potential looked so shiny on a spreadsheet.
It is also a lesson in unit economics. They eventually dropped the price to four hundred dollars, but they were still losing money on every unit. When the "hand-squeeze" video went viral, the brand became a laughingstock overnight. It is hard to recover when your high-tech disruptor is beaten by a thumb and a forefinger.
Think about the waste involved. Every single one of those machines contained high-grade aircraft aluminum and custom circuit boards. All of that carbon footprint just to replace a task that takes thirty seconds of manual labor. It’s the definition of "Silicon Valley myopia"—the belief that if you add a chip to it, it becomes inherently more valuable.
Moving from over-engineered hardware to over-funded software, let's talk about Clinkle. This was a mobile payments app founded by a twenty-three-year-old Stanford grad named Lucas Duplan. In twenty thirteen, he raised over thirty million dollars without even having a functional product.
Thirty million dollars for a ghost app? That is impressive even by today's standards. What was the "secret sauce" supposed to be?
That was the problem—no one knew. It was shrouded in mystery. They had a massive office in San Francisco, and reports came out that they spent two million dollars on a launch party and high-end perks before they even had a beta. The culture was apparently toxic, and they kept pivoting the "technology" behind the payments. At one point, they were using high-frequency ultrasound to transmit payment data between phones.
Ultrasound? Because clicking a button or using Near Field Communication was too mainstream? How does that even work in a loud environment?
It basically doesn't. Imagine trying to pay for a coffee in a crowded cafe while the espresso machine is steaming and people are shouting orders. The microphone on the merchant's phone couldn't pick up the "secret" sound from the customer's phone. It was a classic "Rube Goldberg" solution. They were trying to bypass the hardware limitations of the time, but it was incredibly finicky.
But why ultrasound? Was there a technical reason, or were they just trying to be different?
The pitch was that it didn't require specialized NFC chips, which many phones lacked in twenty thirteen. But they ignored the fact that sound is an incredibly unreliable data medium. It’s susceptible to echoes, ambient noise, and even the physical orientation of the phone. They spent millions trying to solve the physics of sound when they could have just waited for Apple and Google to standardize NFC, which they did about eighteen months later.
It sounds like they were trying to invent a new physics just to avoid paying a small fee to Visa or Mastercard.
Eventually, the talent started fleeing, and the company burned through the cash without ever reaching a meaningful launch. It is a textbook case of "pedigree" over "product." Investors saw a young Stanford kid and assumed he was the next Mark Zuckerberg, ignoring the fact that there was no actual utility being built.
It is that "founder worship" that gets people every time. If you have the right vest and the right degree, people will throw money at you until you prove you cannot handle it. Speaking of things people could not handle, let's look at Quirky.
Quirky was actually a really cool concept on paper. It was a social invention platform where people could submit ideas for products, the community would vote on them, and Quirky would manufacture and sell the winners. They raised one hundred eighty-five million dollars.
I actually liked some of their stuff. They had that flexible power strip called Pivot Power. It was legitimately useful. So what went wrong?
The unit economics were a disaster. Because they were trying to launch a new consumer hardware product every single week, they could not achieve any economies of scale. They were over-promising on manufacturing and taking on all the inventory risk. If a product flopped, Quirky was stuck with the bill. They ignored the reality of supply chains in favor of a "move fast and break things" mentality that just does not work when you are dealing with physical atoms instead of digital bits.
But didn't they have a partnership with GE? I remember seeing their logo on a bunch of smart home stuff.
They did, but even GE couldn't save them from the "long tail" of bad ideas. For every Pivot Power, there were ten products like a specialized egg separator or a weird citrus sprayer that only five hundred people actually wanted. In hardware, you need a "hit" to pay for all the R&D of the "misses," but Quirky was trying to be a hit factory for everything at once.
How did they manage the quality control? If you're launching a new product every week, you can't possibly be doing rigorous safety testing, right?
You hit the nail on the head. Their return rates were astronomical. When you rush a product from a "napkin sketch" to a Best Buy shelf in ninety days, things break. They had a smart air conditioner called Aros that was plagued by connectivity issues. Because they were spread so thin across a hundred different product categories, they didn't have the engineering depth to fix any of them. They were a mile wide and an inch deep.
So they were basically a giant laboratory that forgot it was also supposed to be a store. It is the danger of "crowdsourcing" everything. The crowd might love a niche product, but that does not mean a million people will buy it at a price that covers the tooling costs in China.
And that brings us to the second segment of our list, where we look at startups that failed not just because of bad engineering, but because they fundamentally misread human behavior or legal boundaries. Let's talk about Aereo.
Ah, the "one weird trick" of the broadcast world.
Aereo was a service that allowed you to stream over-the-air television on your phone or computer. To get around copyright laws, they built massive data centers filled with thousands of tiny, dime-sized antennas. Each subscriber "rented" their own individual antenna.
It was such a cheeky legal loophole. Because the law said an individual can record TV for personal use, they argued that they were just providing the equipment for you to do it remotely.
It was brilliant from a technical-legal standpoint, but it was a direct assault on the retransmission fees that big broadcasters rely on. The case went all the way to the Supreme Court in twenty fourteen. The Court basically said, "If it looks like a duck and quacks like a cable company, it's a cable company," regardless of your tiny antennas. They ruled it illegal, and Aereo was dead within weeks.
But why didn't they just pivot to being a legitimate streaming service? They already had the user base.
Because their entire business model was predicated on not paying the fees. Once the Supreme Court said they had to pay like everyone else, their margins vanished. They didn't have the capital to compete with Comcast or Spectrum on their own turf. It’s a reminder that a "legal hack" isn't a business plan.
Did they really think the Supreme Court would care about the technicality of the "individual antenna"?
They did. Their lawyers argued that if Aereo was illegal, then the entire concept of the "cloud" was illegal. They tried to frame it as a threat to all remote storage. But the Justices saw right through it. Justice Breyer famously compared it to a valet service that claims it isn't a parking lot because each car is in a separate "individual" space. The court doesn't like it when you try to outsmart the spirit of the law with a clever gadget.
That is a recurring theme. You can't just "disrupt" your way out of federal law if you don't have the lobbying power to back it up. Uber did it by being "too big to fail" before the regulators caught up. Aereo just wasn't big enough yet.
Now, let's shift to a social failure: Yik Yak. This was the anonymous social media app that took college campuses by storm in twenty fourteen. At its peak, it was valued at four hundred million dollars.
I remember Yik Yak being the most toxic place on the internet for about six months. It was like a localized version of the worst parts of Reddit.
That was the problem. The "feature" was anonymity based on geographic location. It was great for finding out where the free pizza was on campus, but it was a nightmare for bullying and harassment. Because there was no accountability, it became a tool for bomb threats and hate speech. They tried to "fix" it by forcing users to create handles and profiles, which ended the anonymity.
And once you take the anonymity out of an anonymous app, you're just left with a worse version of Twitter. But didn't they try to pivot into a "local discovery" app?
They did, but the brand was already poisoned. Parents hated it, school boards banned it from their Wi-Fi networks, and once the "edginess" was gone, the college kids moved on to Snapchat. Users fled immediately. They misread the "value" of their own product. The value wasn't the location; it was the ability to say things without consequences. When they tried to make it "safe" for advertisers and investors, they killed the very thing that made it popular. It went from a four hundred million dollar valuation to being sold for scraps to Square in twenty seventeen.
It’s a fascinating case of "unintended consequences." They thought they were building a digital bulletin board. They ended up building a digital burn book. Was there any way to moderate a location-based anonymous app without killing the vibe?
Probably not with the technology of twenty fourteen. You would need a massive army of human moderators or very sophisticated AI—which wasn't ready yet—to catch the nuance of campus-specific bullying. If someone posts "The girl in the red sweater in the library is a thief," a global moderator doesn't know who that is, but everyone on that campus does. The hyper-locality made the harassment more intimate and more damaging.
It is the "anonymity paradox." You need it to grow, but it makes you un-monetizable and radioactive to the general public. It is a hard needle to thread.
Speaking of radioactive, let's talk about Peeple. This was famously dubbed the "Yelp for People." The idea was that you could rate any human being on a five-star scale—your ex-boyfriend, your boss, your neighbor—whether they wanted to be on the app or not.
This is literally a Black Mirror episode. I think it was called "Nosedive." The founders were genuinely surprised by the backlash, weren't they?
They were shocked. They thought they were creating a "positivity" tool for "character references." But the internet immediately saw it for what it was: a centralized hub for stalking and reputation destruction. The global outcry was so intense that they had to go into hiding before the app even launched. When it finally did come out, it was a "neutered" version where you had to opt-in to be rated.
Which, of course, no one did. Because why would you give people a platform to publicly grade your personality? "Oh, look, I have three stars for 'listening skills' from my ex-wife." No one wants that.
The founders, Julia Cordray and Nicole McCullough, faced a literal storm of international condemnation. They were interviewed on major news networks and seemed completely baffled that people didn't want to be quantified like a burrito at Chipotle. They argued that "if you're a good person, you have nothing to fear." Which is the exact argument used by every surveillance state in history.
It shows a complete lack of emotional intelligence on the part of the founders. They were so focused on the "data" of human interaction that they ignored the "dignity" of it. It is a common pitfall for tech-heavy founders. They see the world as a series of APIs to be optimized, rather than a collection of people with rights.
We have covered some big ones, but I want to dive into some of the more "delightfully" weird ones now. Starting with Bodega. This was the company that wanted to replace "mom and pop" corner stores with high-tech vending machines in apartment lobbies.
The name was the biggest mistake. They called it "Bodega" and used a cat as the logo—a nod to the famous bodega cats of New York. It was seen as incredibly tone-deaf and culturally insensitive.
It was basically a box with a camera that used computer vision to see what you took and charged your credit card. So... a vending machine. But with more venture capital.
It raised several million dollars, but the PR backlash was so severe they never recovered. They changed the name to "Stockwell," but then the pandemic hit. If your business model relies on people being in shared lobbies and gyms, twenty twenty was a bad year to exist. They folded shortly after. It is a case of "disruption" for the sake of disruption. No one was complaining that their local corner store was "too human."
Do people really want to buy a single roll of toilet paper from a robot in their lobby? I feel like part of the "bodega experience" is the interaction, the sandwich guy who knows your order, the chaos of the shelves.
People actually like their local shopkeepers. Who knew? They tried to automate a relationship, and that’s a hard sell. Also, the logistics were a nightmare. Who restocks the "Bodega" boxes? You still need a human in a van driving around. So you haven't actually removed the human element; you've just hidden it and replaced a friendly face with a cold, gray box. It was an efficiency play that forgot that humans aren't always looking for maximum efficiency. Sometimes we just want a bacon-egg-and-cheese and a "how's your day going?"
Now, let's talk about my personal favorite: "Yo."
Yo is the ultimate "flash in the pan." It was an app where the only thing you could do was send the word "Yo" to your friends. No text, no photos, just "Yo."
And it raised one point five million dollars at a ten million dollar valuation! In twenty fourteen! I remember people thinking this was the end of civilization. "The kids can't even type sentences anymore!"
But wait, didn't it actually have a use case during the World Cup? I remember hearing something about that.
Yes! That was its peak. You could subscribe to a "Goal" channel, and every time someone scored in the World Cup, your phone would just shout "Yo!" It was the most efficient notification system ever built. What is fascinating from a technical perspective is that it actually had a lot of utility as a "zero-character notification." You could use the API to have your server send a "Yo" when a process finished, or have a smart doorbell send a "Yo" when someone was at the gate. It was a very efficient way to send a binary signal. But as a social network? It lasted about three weeks before everyone got bored.
I remember the founder, Or Arbel, wrote the original app in just eight hours. It was a joke that went too far. But it actually opened up a serious conversation in the tech world about "contextual communication." If I send you a "Yo" at 8:00 AM, it means "I'm awake." If I send it at 6:00 PM, it means "Are we still getting drinks?" The context provides the meaning. It was brilliant in its simplicity, but it was never a business.
It was the peak of "app fatigue." We were so desperate for the "next big thing" that we were willing to fund a single word. It is the digital equivalent of a pet rock.
Then we have Quibi. This is the one that still hurts to think about because of the sheer scale of the waste. One point seven five billion dollars raised. Founded by Jeffrey Katzenberg and Meg Whitman. Hollywood royalty and tech royalty.
The idea was "Quick Bites"—ten-minute shows designed to be watched on your phone while you were commuting or waiting in line. They even had this "Turnstyle" technology where the video would seamlessly switch between portrait and landscape mode.
It was an engineering marvel that solved a problem no one had. People don't mind black bars on their phone screens if the content is good. But the real killer was the timing. They launched in April twenty twenty, right when the entire world stopped commuting.
You're stuck in your house with a sixty-inch TV and a Netflix subscription, and Quibi wants you to pay eight bucks a month to watch a show about a "Golden Arm" on your tiny phone screen. And you weren't even allowed to take screenshots!
That was the most "boomer" tech decision in history. They didn't want people making memes or sharing clips, which is how things go viral. They tried to control the "user experience" so tightly that they choked the life out of it. It lasted six months before they shut it down and sold the library to Roku for pennies on the dollar.
Is there any world where Quibi works? Like, if the pandemic didn't happen, does it survive?
Probably not. Because TikTok and YouTube already owned the "short-form" space for free. Quibi was trying to sell premium, high-production-value short-form content to a generation that prefers raw, authentic content from creators. They were fighting the tide of how the internet actually works. It is a reminder that "pedigree" does not equal "product-market fit." You can have the best directors in the world, but if you're forcing people to consume media in a way that feels unnatural or restrictive, they will just go back to TikTok.
Let's talk about a hardware disaster that started on Kickstarter: The Coolest Cooler. This was a cooler that had a built-in blender, Bluetooth speakers, a USB charger, and LED lights. It raised thirteen million dollars from sixty thousand backers.
It was the most-funded Kickstarter ever at the time. I remember seeing the video—it was every "bro" dream combined into one plastic box. But "feature creep" is a real thing, isn't it?
It is the "Quirky" problem on a smaller scale. They promised so much for such a low price—one hundred eighty-five dollars for the early birds—that they couldn't actually manufacture it. The motor for the blender was too expensive, the shipping costs for a heavy cooler were astronomical, and they kept running into production delays. They ended up asking backers for an extra ninety-seven dollars just to get their coolers shipped.
That is a bold move. "Hey, thanks for the millions, can you give us more so we can actually give you the thing you bought?" How did the backers react?
About as well as you’d expect. It turned into a legal nightmare. The Department of Justice got involved, there were lawsuits, and the company eventually shut down in twenty nineteen, leaving thousands of people without their blenders. It is a cautionary tale for any hardware startup: "Simpler is better." Every feature you add is a new point of failure in your supply chain. If the blender broke, the whole cooler was basically trash.
And let’s not forget the logistics of shipping sixty thousand massive coolers. Most Kickstarter creators don't realize that shipping is a specialized skill. If you miscalculate the weight by two pounds, and you're shipping sixty thousand units, you just created a million-dollar hole in your budget. They were trying to be a lifestyle brand, but they were actually a logistics company that didn't know how to do logistics.
Speaking of simple, how about "Ship Your Enemies Glitter"?
This one is hilarious. It was a website where you could pay ten dollars to have an anonymous envelope full of glitter sent to someone. The "glitter bomb."
The founder, Mat Carpenter, launched it as a joke on a Tuesday. By Wednesday, it had gone so viral that he had thousands of orders and millions of hits. He famously posted on social media: "Please stop buying this horrible glitter product. I am sick of dealing with it."
Why would he want people to stop? He was printing money!
He realized that "going viral" is actually a nightmare if you are a one-man operation trying to stuff envelopes with the most annoying substance on earth. His entire apartment was covered in glitter. He couldn't eat, he couldn't sleep, and he had a backlog of thousands of angry customers. He sold the site for eighty-five thousand dollars just a few weeks later. This is the "Success Trap." If you don't have the infrastructure to scale, your own popularity will destroy you.
It’s a great example of a "micro-startup." It did one thing perfectly, it captured the cultural zeitgeist for forty-eight hours, and then it became a liability. Mat was smart to get out when he did. The buyer, however, probably spent the next three years trying to get glitter out of their carpet.
We have to mention Theranos, even though it is more of a "fraud" than a "ridiculous startup," but the central premise was certainly ambitious. A single drop of blood for hundreds of tests.
The "Edison" machine. It was the "Juicero" of healthcare. They spent years pretending they had this revolutionary microfluidic technology, when in reality, they were just diluting blood samples and running them through standard Siemens machines in the back room.
What's wild to me is that they were valued at nine billion dollars without ever having a peer-reviewed paper or a third-party audit of their tech. It was entirely built on the "aura" of Elizabeth Holmes—the black turtleneck, the deep voice, the board of directors filled with former Secretaries of State.
It is the ultimate example of "fake it till you make it" going horribly wrong. In software, if your beta crashes, people lose their saved games. In healthcare, if your "beta" gives a false positive for cancer, people's lives are ruined. The technical failure here was fundamental: you cannot physically get enough data out of a single drop of capillary blood to run hundreds of distinct assays accurately. The physics just don't allow for it.
And yet, people wanted to believe it so badly. They wanted the "Steve Jobs of Medicine." It shows how desperate we are for a technological savior. We are willing to ignore the most basic scientific red flags if the narrative is compelling enough. Holmes didn't just sell a machine; she sold a vision of a world without needles.
And finally, let's talk about the shortest-lived "flex" in history: the "I Am Rich" app.
Oh man, the nine hundred ninety-nine dollar red gem.
It was an iOS app that did nothing. It just showed a glowing red gem on your screen. It was a pure status symbol. Eight people actually bought it before Apple pulled it from the store twenty-four hours later.
Those eight people are the ultimate early adopters. "I have a thousand dollars and I want the world to know I'm willing to waste it." It is the purest expression of the "App Store Gold Rush" era. It’s almost performance art.
Did they ever get their money back?
Most of them did. Apple refunded the purchases, but the developer, Armin Heinrich, reportedly made about five thousand dollars from the stunt after Apple took their cut. It’s the ultimate low-effort, high-reward startup. No R&D, no manufacturing, no customer support. Just a JPEG of a gem and a very high price tag.
So, what are the takeaways from all this wreckage? For me, the biggest lesson is "Validate the problem, not the solution." Most of these companies—Juicero, Bodega, Quibi—built incredible technology for problems that didn't exist. No one was crying out for a Wi-Fi juicer or a ten-minute show they couldn't screenshot.
And "Unit Economics are King." You can have all the venture capital in the world, but if you are losing money on every "Quick Bite" or every "Cooler," you are just accelerating your own demise. You cannot "subsidize" your way into a profitable business model if the fundamental math doesn't work.
I also think there's a lesson in "Respecting the Atoms." Hardware is hard. Quirky and The Coolest Cooler found out that physical logistics and manufacturing are a completely different beast than writing code. You can't "A-B test" a plastic mold that cost fifty thousand dollars to create.
For our listeners who are thinking about joining a startup or investing in one, I'd suggest a "pre-mortem" analysis. Imagine it is three years from now and the company has failed. Why did it fail? Usually, the answer is right there in the pitch deck if you look hard enough. Is it a "solution looking for a problem"? Is it "fighting the laws of physics"? Or is it "ignoring human nature"?
If the answer is "all of the above," you might be looking at the next Juicero. Or worse, the next Theranos. It’s important to remember that just because a founder is charismatic and has a lot of money doesn't mean they have a viable business. In fact, sometimes the more money they have, the less they are forced to confront the reality of their product.
Well, this has been a trip down memory lane. I'm going to go squeeze an orange with my hands just to feel alive. I don't need four tons of pressure to enjoy my breakfast.
And I'm going to go send a "Yo" to our producer, Hilbert Flumingtop. Thanks for putting this together, Hilbert. I hope it doesn't annoy you too much.
Big thanks to Modal for providing the GPU credits that power this show. If you're building the next great—and hopefully not ridiculous—startup, check them out for your serverless GPU needs. They won't make you buy a seven hundred dollar juicer.
This has been My Weird Prompts. If you're enjoying our deep dives into tech history and absurdity, a quick review on your podcast app really helps us out. It's the only way we know people are actually listening and not just "Yo-ing" us.
Catch us at myweirdprompts dot com for the full archive. See you next time.
Goodbye.
