Thanks for subscribing to SatPost.
Today, let's go down memory lane with Skype (which will soon be shut down by Microsoft).
Also, this week:
The (new) largest fast food chain in the world
Citibank’s $81 trillion UX problem
…and them fires posts (including the Oscar’s)
Microsoft announced that it is “retiring” Skype in May.
This is truly the end of an era.
The company was founded in 2003 by a European team including Niklas Zennström (Sweden), Janus Friis (Denmark) and four engineers from Estonia (Ahti Heinla, Priit Kasesalu, Jaan Tallinn, Toivo Annus).
Prior to Skype, this group had built a music file sharing service and one of the best ways to get random viruses on your parent’s desktop computer: Kazaa (the peer-to-peer file sharing platform that Soulja Boy ingeniously hacked — along with LimeWire — by seeding his song “Crank Dat” on the network but only after re-naming the file after a top 10 Billboard song to trick people into downloading it).
The Skype co-founders used Kazaa’s peer-to-peer (P2P) architecture to build a voice over internet platform (“Skype” is short for “sky peer-to-peer”). This was a very practical play because the internet infrastructure that exists today didn’t…err…exist back then. Skype was able to scale quickly around the globe by connecting nodes in each region without having to make massive local server infrastructure investments.
And man was I happy that Skype did what they did…when they did it.
I lived in Saigon in the mid-2000s and the move to Southeast Asia was much more palatable with the existence of Skype. Why? Long-distance phone calls to my family and friends in North America would have been such a pain otherwise.
In the pre-Skype days, making a call across the Pacific was an entire production. I remember one summer I lived with my uncle on the outer districts of Saigon. The landlines in the area didn’t have long-distance access. So, to speak with my parents back in Canada, we had to find a specific time block twice a week. The scheduling itself was a annoying because of the 12-hour time difference. When it was time to chat, I walked about 10 minutes to a PC cafe. Then, I bought a scratch phone card and paid through the nose ($2 a minute or something insane) to check in with my parents and promise them I wouldn’t drink my uncle’s moonshine cobra whiskey (well, I did one time but it tasted so bad that I never dabbled again).
Skype was just a totally different ball game. The service was cheaper, better and faster. They soon added video! It was a pure consumer-friendly win with a simple freemium business model. Any two Skype accounts could chat for free online or users could buy credits for Skype-to-landline/mobile calls. Mind you, the Skype credits were less expensive than traditional long-distance calls (which were the cash cow for telecom firms).
Also, how iconic was that call sound? Beep-bob-beep-bop-beep-bop. Like a sorcerer summoning parental judgement from thousands of kilometers away.
I was far from the only person experiencing this magic.
According to a CNBC video profile, Skype launched the same year that laptop sales outpaced desktop sales for the first time. It wasn’t quite the mobile revolution but personal computing was on a steep trajectory upwards and Skype was at the right place at the right time.
“None of [the Estonian engineers] had any telecoms background,” Skype co-founder Ahti Heinla explained in the CNBC piece. “I think that was the key thing about why it worked so well. It didn't look like telecom. It didn't behave like one.”
Skype was a huge beneficiary of network effects. One really smart hack the Skype team borrowed from Kazaa was to show how many users were online (aka social proof).
By 2004, Skype grew to 1.5 million users. A year later it had over 50 million users and there were over 500 million users by the end of the decade.
I had dabbled with text-based messaging such as ICQ, MSN Messenger and AIM, but Skype was probably the first internet communication technology that blew my socks off.
Long before Zoom, Skype became the verb to do video calls with another person. It’s equal parts sad and embarrassing that Skype fumbled such a lead. The service should be up there with “we became a verb” hall-of-famers including Google, Venmo, GoFundMe, Uber, Photoshop, DoorDash, Hoover, FedEx, Xerox and — maybe this is just me — Wendy’s (give it a try next time you’re craving a spicy chicken product: “yo, let’s Wendy’s it”).
Unsurprisingly, the tech world was seduced by Skype’s massive potential from the beginning. Investor Sheel Monhot has a great summary of the business dealings that culminated with Microsoft’s $8.5B acquisition in 2011 (the company’s largest deal at the time):
2003: [Skype] was founded in Estonia by the founders of Kazaa, a P2P file sharing service used to pirate music. The Kazaa money meant that they didn’t need to raise VC $.
2005: 54M users, $7M revenue in ‘04 to $60M in’05. Acquired by eBay for $2.6B!! eBay claimed it would help buyer-seller communication but that never made sense to me, and eBay wrote it down by $1.7B in 2007.
2009: 400M users, 100M DAU, $700M revenue. 65% of Skype was acquired by Silver Lake, A16Z, CPPIB, and Index, for $1.9B, valuing the company at $2.75B.
2011: 600M users, 170M MAU, $1B revenue. [Microsoft CEO Steve] Ballmer buys it for $8.5B. Incredible deal for the ‘09 investors, 3x in 2 years.
Skype had 600 million users at the time of the Microsoft acquisition. To give some perspective, that was about the same amount of users on a random website out of Harvard Square called Facebook.
There are now less than 40 million users on Skype.
“Microsoft bought and effectively killed Skype,” writes investor and long-time tech analyst Om Malick. “I could write a Ph.D. dissertation on this — for now, this is all I have to say. Microsoft didn’t know how to nurture Skype, and its bureaucracy killed one of the most iconic brands of the new century.”
A major culprit in Skype’s decline is the worst piece of office-related software ever invented Microsoft Teams. Launched in 2017 as Microsoft’s response to Slack, the messaging and workplace platform sucked all the oxygen out of the room for Skype.
The internal neglect by Microsoft was compounded by the external competitive landscape:
WhatsApp: Founded in 2009, the messaging platform was acquired by Facebook — which also mulled a Skype acquisition — for $16B in 2014, when it had more than 500 million users.
FaceTime: Building off the iPhone's front-facing camera, Apple rolled out FaceTime on iOS in 2010 and has been responsible for more people seeing their own double-chins than every mirror created in the past 5,000 years combined.
Google: Over the years, the search giant entered the communication space with Chat, Meet, Hangout and I honestly have no idea what else. I think I used all of them because they were in the sidebar of my G-Mail account and I’m guessing these took some business away from Skype.
Skype really stumbled with the transition to mobile. WhatsApp and FaceTime were built to win with the smartphone form factor and benefitted hugely from growing their networks off of the device's contact list.
It’s a bit poetic that Skype’s mobile struggles happened under the ownership of Microsoft, which famously struck out multiple times with its mobile strategy (culminating in the bunk $7B deal for Nokia in 2014).
Just as Skype clapped long-distance telephone services by being faster and better, WhatsApp and FaceTime clapped Skype by being faster and better (and in the case of FaceTime, more double-chinny).
Fast forward to the pandemic. Instead of capitalizing on a once-in-generation remote work opportunity, Skype had to watch Zoom become a verb. While Zoom has had a wild ride since its IPO in 2019, the platform’s ease-of-use and ability to handle massive internet meeting rooms has created a $23B company (which will be competitive in the upcoming AI-first workplace world).
Meanwhile, the Skype carcass will now be rolled into the Teams zombie.
Let me end the story with two final thoughts.
First, Lebron James was drafted the same year as Skype and, holy crap, his hall-of-fame career will somehow outlast the communication platform.
Second, the end of Skype is also the end of the golden age of desktop PC sounds.
Are you sitting down? Because I’m about list some glorious audio logos that were all invented in the 1990s:
ICQ’s “Uh Oh”
Apple MacBook startup chime
Windows 95 startup jingle by Brian Eno
"Jacked into The Matrix and learning Kung Fu" sound
MSN Messenger “trying to spit game in high school” sound
Skype obviously came a few years later but does seem to be one of the last bangers from that period.
Recall that the PC consumer revolution really kicked off in the 1990s. Long before notifications ruined our lives, humans had much more bandwidth for audio notifications.
In 2015, The Verge published an interesting history on Skype and notes that the creators of the original sound actually didn’t want to make overly-catchy audio (it obviously became one):
Friendliness, in fact, was central to Skype’s original sound interface. The company’s current design director Steve "Buzz" Pearce remembers the pitch for the program as "the landline of the future," but the team wanted Skype to feel as natural as answering an old phone. When users got a notification that someone was calling, they wanted it to be an intimate, unobtrusive extension of the person at the other end of the line.
Skype’s original sounds, recorded by outside studio Soundtree Music, eschewed the style and tone of the chintzy cellphone ring. "We didn’t want it to become like a brainworm," says Pearce, humming the "Grande Valse" — Nokia’s famous and sometimes tooth-grinding ringtone . "Not like that."
Skype and Soundtree opted to use original sounds as basic building-block elements. "All the actual components [were] recorded organic sounds like wind, water, pops, people’s voices," says Pearce. Wind, he says, provided the white noise in a notification. A bubble pop could be recorded from a ketchup bottle, a glass, or a human gasp or gulp. "We don’t like technical things, even though we are a technical company," he adds.
"If you actually ask people to hum or sing the Skype ringtone, they can’t."
Once recorded, the sounds were layered on top of each other, creating something abstract but acoustically natural. Skype’s most memorable element was the five-beat incoming call notice, mixed from recordings of a human breath, water, and voices. "If you actually ask people to hum or sing the Skype ringtone, they can’t, accurately," says Pearce. "We did that on purpose, because we don’t want it sticking."
Skype's founding in 2003 also coincided with the year that the cellphone ringtone business crossed $1B for the first time and iTunes was launched for Microsoft (thus greatly expanding Apple’s iPod business). Competition for our ears was ramping up and really hasn’t slowed.
Now, we’ve become so sick of notifications that audio signifiers are more likely to have a negative connotation. That may change as generative AI voice improves but our ears are pretty tapped out at the moment with Spotify, Apple Music, Audible, YouTube and the 50+ hour Mike Duncan podcast series on the Russian Revolution.
I’m not nostalgic for the Skype sound just yet, though.
I remain part of the platform's dwindling user base because it is still my dad’s video chat platform of choice. For that reason, I’ll be riding Skype out for the next two months. It will be sad when that sound is gone. But I’ll always have the memories (and will be reminded of it anytime someone offers me a shot of moonshine cobra whiskey).
The Largest Fast Food Chain In The World…
…by number of stores isn’t McDonald’s (42k) or Starbucks (40k). It’s Mixue, a Chinese bubble tea and ice cream brand with 45k locations per The Wall Street Journal.
The chain went public last week and has a market cap of $14B. To be sure, this isn’t an apples-to-apples comparison. Founded in 1997, Mixue has much smaller store sizes and most of the menu is under $1. That’s how it grew its store count more than 5x since 2018 (when it opened its first foreign location in Vietnam).
McDonald’s and Starbucks are valued at $230B and $121B, respectively, and take their real estate plays much more seriously (of course, McDonald’s is literally a real estate company).
Mixue’s meteoric rise in recent years is due to two main factors: 1) supply chain mastery; and 2) a viral mascot and jingle.
While Mixue runs a franchise model, the chain’s revenue mix is different than well-known F&B chains. In 2024, over 90% of its $3B+ in sales was from selling ingredients (milk, coffee, syrup) and equipment to franchisees (which pay a low initiation franchise fee).
One wild stat: Mixue is China’s largest buyer of lemons (more than 110,000 tonnes in 2023, which are used in ~10B drinks sold a year). The massive buying volume means the chain can procure raw ingredients at 20%+ discounts. These products are shipped through a sophisticated logistics network in China and throughout the region (Southeast Asia, Korea, Australia).
Mixue has also increased its mindshare by creating a snowman mascot paired with a very very viral jingle. This combo has been so potent that the Mixue Snowman is now comparable to Ronald McDonald and Colonel Sanders in Asia.
I was on the fence on whether or not to link to the jingle. The main reason is that I just wrote that jingles don’t work anymore (well, tech ones anyways). This one def work. The second reason is that it riffs on the song “Oh Susanna” and is guaranteed to give you a case of Last Song Syndrome (LSS). I figure we’re all adults, though. So fine, I’ll link to it. Be warned: this shit will be stuck in your head and you might hate me. Seriously, if you don’t want LSS, do not listen to the jingle here, here or here (and definitely not here).
How Is Erewhon Selling A Single Strawberry for $19?
I was able to get out of the house and take my wife to an Omakase Japanese restaurant for dinner last month. As with any good Omakase, the sashimi was flown straight from Japan and tasted hella fresh.
However, there was one curveball I hadn’t had before: the restaurant also flew in a fresh melon just hours beforehand. It might have been the highlight of the meal.
The reason I tell you this is because Erewhon — the luxury LA grocery chain whose name is an anagram of “nowhere” — has been going nuclear viral for selling a single (as in one) strawberry for $19, which is flown straight from Kyoto.
I personally would not pay $19 for a strawberry with a special plastic casing and handle. But I do understand why someone might and the economics of the Japanese farming industry helps to explain why Erewhon is able to sell such a pricey fruit.
In Japan, there’s a farming concept called “hatsumono” (which translates to “first of season”), per a fascinating New York Times piece from 2023.
Take any produce (cucumber, tomato) and the first farmer to market fetches multiples of the average price and gets all the media fanfare.
Strawberries are an extreme example. The traditional growing season starts in Spring (April-May). All that changed with the race to get to market first. Now, strawberries are a winter-time specialty in Japan (particularly around Christmas).
To grow strawberries in freezing weather, farmers build specially-designed greenhouses and use a lot of fuel (eg kerosene) to grow in the new peak season of November-February. Add on the competition between farmers to pick out the best-looking pieces and some single strawberries will sell for $100+…and that’s how you get Erewhon’s $19 single (as in one) strawberry.
Japan’s strawberry growing season is now so out-of-whack that the country imports the fruit during the summer and the growing process is not great for the environment.
I guess an air-freighted melon isn’t great either, but at least I didn’t use a special eating handle.
This issue is brought to you by Bearly AI
Why are you seeing this ad?
Because I briefly considered raising funds for strawberry-growing operation but realized there's no way I could compete with Kyoto farmers.
Instead, I co-founded an AI-powered research app called Bearly AI that can save you and your team hours of work. We just added two new powerful models from Anthropic (Sonnet 3.7) and OpenAI (GPT 4.5) for coding, reading and writing. You can try those along with image, text and audio models from DeepSeek, Grok, Whisper, Gemini, Stable Diffusion and more.
It’s all available in one keyboard shortcut (and an iPhone app). Use code BEARLY1 for a FREE month of the Pro Plan.
Citibank’s $81 Trillion UX Problem
There’s a saying in the banking industry that “to err is human, but to really fuck up you need computers”.
I know this because a number of X users replied with that quote when I posted a story about Citigroup accidentally crediting a client account with $81 trillion (it was supposed to be $280).
The snafu passed two levels of review and wasn’t caught for 90 minutes. How? Because of an idiotic software user interface.
Basically, the account had been accidentally flagged for suspicious activity. To process the payment, the Citi employee had to input transaction into a “rarely used back-up screen” and “one quirk of the program was that the amount field came pre-populated with 15 zeros, which the person inputting a transaction needed to delete, something that did not happen”, per The Financial Times.
Think about how dumb that is. Seriously, give that a think.
The only number that an “amount” field for bank software should be pre-populated with is a single 0. There is no other acceptable figure. Fifteen 0s is insane. It’s particularly insane because the number is so comically high that it would evade a person’s natural defences. Like, if you saw that number, you would think your brain was playing tricks on you and that there is no way the bank would have been dumb enough to pay some consultant 7-figures to design such a UX.
This isn’t Citi’s first rodeo with banking software UX gone wrong, either.
Back in 2020, Citi was responsible for paying out interest for Revlon, which was going through a bankruptcy restructuring. Revlon creditors were supposed to receive interest payments totalling ~$8m but instead received a $500m principal repayment.
Citi asked for the funds back but the creditors — who made no mistake and were happy to get back the full principal amount (which may not have happened over the course of the bankruptcy) — said “nah, we gonna keep this”. Ultimately, Citi and the creditors went to court and the judge ruled against Citi.
The mistake was the bank’s fault…and a major part of the story was the abysmal back-end UX, per Ars Technica. Below is an image from court case. TLDR: To prevent Citi from paying out the principal, the user had to fill out 3 fields (Principal, Front, Fund). The Citi employee and supervisors (reasonably) thought they only had to fill out the “principal” field and comedy ensued.
Citi employees should obviously be more vigilant and have messed up many times on their own. A different news story from last week revealed that another Citi bank account had accidentally been credited with $6B because “a staffer handling the transfer copied and pasted the account number into a field for the dollar figure.” HAHAHAHAHAHAHA
That’s a skill issue right there. But it’s hard to fully blame the Citi employees for the Revlon $500m payment or the $81 trillion incident. Just very dumb UX.
Anyway, the wildest response my post received was a plan someone pitched to try and keep some portion of the $81 trillion if it hit his bank account (X/The Salon Don):
Wake up, see $81t in my account. Start moving fast. Load up 10 flash drives with $1b in every kind of Crypto and head straight to the airport.
Give my fiancé 10 minutes to decide if she’s coming with me. We will probably see our friends and family again but I don’t know.
Once in Colombia lay low a few days. Distribute flash drives in random hotel rooms. Send one back to my family with the note “I’m alive. Put this flash drive somewhere safe, don’t tell anyone about this”.
Then find a coke dealer. “Let me talk to your boss for $10k. Let me talk to his boss for $20k”.
“I need a new identity and European passport. Can pay any amount”.
Then I fly somewhere in Europe and open a tanning salon / med spa to start laundering money. Within a year start claiming $10m of EBITDA and paying taxes.
Then start an AI company burning a ton of money. Fly to Saudi Arabia for VC funding but in the meetings I put my cards on the table.
“Listen. I’ll give you $5b of crypto if you legitimately buy my AI business for $700m”.
Just need one taker. Buy a gold immigration card. Move back to the states.
You now have a legitimate $700m in the system and a few flash drives you can use if ever needed.
Look, I probably obviously wouldn’t do that but, wow, great plot for a Netflix caper film. The larger lesson is that opening a Citi bank account might be one of the greatest asymmetric investment opportunities in the history of capitalism. It’s free and there’s a 0.0000000000000000000000002% you could walk away with $1 quintillion.
Links and Memes
Shane Gillis was on SNL last week…and dropped a hysterical skit about a phama product called “CouplaBeer”. It was a throwback to the most demented fake SNL ads during the show’s “give zero F” heyday. (X/SNL)
Pinterest Is Being Strangled by AI Slop…the image-heavy social network is getting flooded by AI-generated images and there’s a whole network of YouTube creators shilling “plans” to make $10k+ a month filling the platform with said slop. (Futurism)
Scientists created ‘woolly mice’ in a step to bring back the mammoth…this is either the best idea ever or the worst idea ever. (CNN)
If you ever played the board game Risk…and lost to your sibling or high school buddy after a 6-hour session, you will want to watch this meme. (X/Smubpublius)
Vice President JD Vance…is the subject of a wild meme trend called “Rare Vances”. The entire internet — on all sides of the political spectrum — are sharing photoshops of his face like collecting "rare" Pokemons. (Know Your Meme)
The Government Knows AGI is Coming (?)…a fascinating Ezra Klein podcast with a top AI advisor from the Biden administration talking about how a lot of the Trump administration’s actions with the Federal bureaucracy may be in preparation for the arrival of some form of AGI in the next few years. (The Ezra Klein Show)
FTX founder Sam Bankman-Fried…did a pretty surreal 40-minute video chat podcast with Tucker Carlson. SBF is two years into a 25-year sentence over the FTX implosion. He’s in the same cell unit as Diddy and talks about life on the inside, regrets and how he views the crypto space. Wild part of the story is if somehow kept his fraud scheme going through the 2022 crypto bear market without anyone catching him, FTX would now be sitting with ~$90B of assets against ~$20B of liabilities. (Tucker Carlson Show)
Amazon paid ~$1B for the creative rights to James Bond…on top of the $8.5B it paid for MGM studio. New details from the story show that the deal was struck after producer Barbara Brocolli was quoted in WSJ calling Amazon creatives “f—ing idiots”. Bezos read that and said “I don’t care what it costs, get rid of her.” (The Hollywood Reporter)
King Charles is doing a radio show…with Apple Music about his favourite songs, proving that the male urge to podcast is irrepressible. There is no amount of money or status acquisition that can quench the thirst to speak into a Shure mic about a personal passion or hobby. (X/Apple Music)
…and them wild posts:
Finally, here’s some Oscar’s content for you. I didn’t watch the event but I did watch Best Picture winner Anora after and my review is that you should definitely NOT watch that film with your parents. As an aside, here is an insightful podcast from The Town on how that winning an Academy Award still gives talent a ton of leverage within the industry (even if viewership is down huge, from 44m in 2000 to 18m in 2025, which isn’t bad if we adjust for the time). The Ringer has a good winners and losers list.
Skype closing down is one of these great examples of an acquiring companies relentless dedication to failure (maybe like MySpace with Fox).
I worked at Microsoft for a few years from 2013 and I remember they had this iconic London headquarters, they were the only "cool" part of Microsoft where nobody wore suits 😂
Everyone loved them, then one power struggle later Microsoft lay down the law by shutting it all down and fired everyone in London (mostly). The new plan was to abandon the hugely successful consumer side and build an enterprise product in LA (later just a reskinned "Lync" called "Skype for Business").
Despite this blunder Skype was still huge and the consumer "Skype" product was actually very successful in the business world. I saw in 2020-2022 it was still by far the #1 comms channel for mobile game developers to work with partners. Why on earth Microsoft didn't take advantage of this and build Slack before slack or use it to make a 10x better version of Teams I don't understand.
I originally thought it was one of Microsoft's worst acquisitions as they acquired gold and turned it to dust...but I had a think and realised even thought Skype seemed a great deal at $8.5b and is now closed down Microsoft have had some stinkers and a track record of making good deals with horrific execution:
- Danger, Kin smartphone: only $1b but shut down after 48 hours due to horrific reviews (qyute similar to Hololens)
- Mixer/Beam: shut down 4 years after buying/gifted the whole thing to Meta for free
- Nokia: $7.2B destroyed a large chunk of Finlands tech economy with this one, but maybe not Microsoft's fault
- aQuantive: the WOAT, $6.3B, 5x cost of YouTube & 2x the cost of DoubleClick by Google, wrote down a 100% loss and sold the only good part to Meta for $100m
this was like 3 or 4 posts worth in one...and the chuckles just kept on coming. Multiple high-fives for the ode to 90s audio cues