Many of you readers are probably on vacation. This is my last e-mail send before summer break. SatPost will be off for July and I'll be back in your inboxes some time in August (side note: I'm addicted to X, so you can still find me posting on that app).
Your boy has family trips planned and will be working on new stuff for — **shilling alert** — Bearly AI and Liona AI.
Anyway, today is a quick round-up of my most popular articles in 1H 2025.
Also this week:
How Din Tai Fung Does $28m Per Location
How Eli Lilly Beat Novo Nordisk On GLP-1s
RIP to Legendary FedEx Founder Fred Smith
3 Thoughts About The NBA Finals
Apple’s $40m Sponsorship for F1 film
…and them fire posts (including European vacations)
5 Top Articles and Memes of 2025 (So Far)
I appreciate all you readers.
If I’ve ever made you laugh, say “huh, interesting” or think “wow, Trung has awful takes”…I would love for you to share any of the following articles with friends or family that might like them.
Without further ado, here are the 5 most popular SatPost articles from the first half of 2025 based on views (and general vibes from the e-mail responses I received):
All of [Garmin’s navigation device] progress hit a wall in 2009. Launched in 2007, the iPhone had kicked off a smartphone revolution. Then Google Maps’ came out a year later with a price point of $0, which was slightly more appealing to customers than $300+.
According to multiple anonymous sources at the company, here is a photo of the executives when they saw the launch of the Google Maps app:
Also an image of physical mapmakers when they saw Garmin car GPS devices.
Over the next two years, Garmin’s sales fell from $3.5B in 2008 to $2.7B in 2010. Garmin sales wouldn’t fall any further but — as mentioned earlier — it took a decade for Garmin’s sales to get back to $3.5B. If you adjust for inflation, Garmin didn’t get back to its 2008 revenue high-water mark until 2023. The largest peak-to-trough drawdown for Garmin’s stock was ~90%.
Fortunately, the company’s salvation was already cooking internally. Remember how we talked about Garmin’s commitment to R&D? Just as they had found a way to take the GPS from the ocean (boats) to air (planes) to the street (cars), Garmin’s R&D team had identified fitness as a niche market all the way back in 2003 [with a GPS-enabled watch].
I’m a Blackjack Guy. There’s a little bit of skill involved and few things in life give me more pleasure than saying “double monkey me dealer please” when I have an 11 against a bust card.
This is not even close to the most popular casino game, though. That would be slot machines by a mile. These devices bring in more revenue than all the other popular games combined including blackjack, craps, roulette, baccarat, poker etc.
By the numbers, ~70% of Nevada gaming revenue is from slot machines and the majority of players are locals. In 2024, Nevada casino games won $15.6B from players with $10.5B coming from slot machines. It’s a jarring figure when put into perspective with other forms of entertainment (here’s a comparison: the entire US box office in 2024 was $8.5B).
How? Because slot machines are the “crack cocaine of gambling”.
As a blanket policy for 2025, I implemented a “no new streaming TV show” policy.
Why? I’ve been burnt way too many times in the past 5 years with people telling me I have to watch a new show:
“Dude, The Bear is sooooo good!”
“Bro, you seriously HAVE to watch Ted Lasso!”
“Trung, have you seen Bridgerton? YOU HAVE TO!!”
“Man, this GoT spin-off House of the Dragon is INSANE!!!”
“Trung, I’m telling you. Yellowstone will make you shit your pants.”
“Are you watching The Mandalorian? It’s unreal! You have to! I’m serious!”
“Dude, dude, dude. 3 Body Problem is amazing. You gotta watch it. Seriously!”
So, I did. I watched all of them and every single one had the same outcome: started strong with some laughs, suspense and well-timed cliffhangers…but then storylines became circular or didn’t make a lot of narrative sense (either within a single season or between seasons).
What is going on? The economics and tech behind streaming is affecting the quality.
I calculated OnlyFans dividend distribution in 2023 and compared it to the “net profit” of major tech firms in 2024 to put some context around the business.
By this measure, each OnlyFans employee is generating a “profit” of $11m. Only Tether — the stablecoin company that somehow generated $13.7B in profit with only 165 employees — has a higher figure at $83m per employee. Of the Big Tech Ballers, Nvidia is the only one to break 7-figures with a profit of $1m per employee in 2024 (Meta is at $837k, Netflix is at $621k, Apple is at $585k and Google is at $545k).
Ultimately, the higher education bundle we talked about earlier is supposed to prepare students for the real world. Guess what? Knowing how to use AI and how to manipulate the correct models and how to ask the right questions and how make the right prompts is what will be valuable in coming decades. I find a lot of credence in the popular line that “AI won’t take your job, but someone who knows how to use AI will.”
So, grade the workflow (and not the output). Let’s just keep some form of writing to force critical thinking and make sure there’s a signal for future employers.
Is grading the workflow actually worse than grading students copy and pasting Wikipedia? It might not be the most ideal situation but we have to be realistic about the technology. The teacher can also train his or her requirements into an AI and have it mark the workflows, thus freeing up time for more hands-on teaching. Again, is that any worse than a teacher reading 100 papers on The Great Gatsby that are all +/-2% derivations of Coles Notes?
PPS. Check out my Caffeinated Deep Dives podcast, which does...err...deep dive episodes on bangin' topics including the iPhone, Calvin & Hobbes, Hermes, La Sagrada Familia and Sriracha Sauce.
This issue of SatPost is brought to you by Bearly AI
Are you looking for an encrypted and anonymized AI chat tool?
Then, try the Bearly AI research app, which provides an easy-to-use UX to access leading models from OpenAI, Anthropic, xAI, DeepSeek, Gemini and more.
While some AI providers keep chat logs indefinitely even after a user deletes them...all Bearly AI chats are encrypted and requests to LLM providers are completely anonymized.
If you're looking for AI chat privacy, get access to leading LLMs and try one month FREE of the Pro Plan using the code BEARLY1.
…across its 18 locations in America, which is the highest average unit volume (AUV) in the casual-dining space and stacks up very well against some usual suspects according to Restaurant Business:
Mastro's Restaurants (steakhouse, $14.5m)
The Cheesecake Factory (everything, $12.4m)
Nobu Restaurants (Asian, $10m)
Chick-fil-A (chicken, $9.3m)
Benihana (Asian and Onion Volcanoes, $6.5m)
McDonald’s (Big Macs, $3.9m)
Din Tai Fung was founded by Yang Bing-yi and his wife Lai Penmei in Taiwan in 1958 as a cooking oil business but rebranded as a noodle shop in 2000. He passed away in 2023 and his eldest son Yang Ji-Hua now runs the show. The chain has 180 locations in 13 countries slanging that handmade soup dumpling (aka xiao long bao aka "little basket bun") and total annual sales have reached $411m.
“To generate unit volumes of that magnitude, a restaurant generally has to do three things: It has to be big, customers have to spend a decent amount, and it has to be busy,” explains Restaurant Business.
Din Tai Fung checks all the boxes:
Big: “The chain’s restaurants are large, anywhere from 5,580 square feet to 25,000 square feet at its sprawling new Manhattan location.”
Customers spend: “Its check averages are high—about $45 per person, according to Technomic. Though individual dishes are reasonably priced—few are more than $20—many customers order multiple items to share with the table. Soup dumplings come in baskets of 10, and the menu is heavy on apps, wontons, greens and other shareables. It also has a full bar, which can further boost check size.”
Busy: Very.
Din Tai Fung’s pork-filled xiao long baos are legendary. It takes serious game to make these dumplings and the volume is intense.
The Tasty YouTube channel shows how they are made: new chefs train for at least 6 months and go through thousands of dumplings to ensure that each dumpling weighs 5 grams and has 18 folds (a lucky number in Taiwanese culture).
Conversely, it takes me 6 seconds to destroy a basket of 10 dumplings with that glorious dip sauce (1 part soy sauce, 3 parts vinegar) dripping all down my white t-shirt.
Fair warning: if you watch that Tasty video like I did, the YouTube algorithm may hit you with some ridiculous ASMR content a few clips later…like it did to me.
How Eli Lilly Beat Novo Nordisk on GLP-1s
Speaking of eating of 100 soup dumplings, some interesting news came out of the world of GLP-1s aka the weight loss drug aka “the fat shot”.
As the story goes, Danish pharma giant Novo Nordisk originally manufactured Ozempic to treat diabetes but doctor’s soon found an off-label use for appetite suppression and weight loss (triggered by GLP-1 hormones).
Novo made a specific weight loss drug called WeGovy and the valuation went gangbusters: in March 2024, the company hit a peak market cap of $640B and was Europe’s most valuable company. Since then, it’s fallen by more than 50% (to ~$300B) and a major reason why is that American pharma giant Ely Lily is on Novo’s ass like white on rice.
Over the past 5 years, Eli Lilly is up +300% (to $730B) while Novo — after the sell-off — is up a more modest +100% (to ~$300B). The Economistrecently detailed a few reasons that Novo fumbled its 2-year lead on Lilly:
Novo didn’t plan enough manufacturing for the massive demand of WeGovy. In America, if there is a shortage of a drug, other pharma makers can fill the gap and make copies of the branded drug and charge it at a cheaper price until the even if its under patent protection. Novo has finally solved the backlog but allowed alternatives to sell into the most lucrative pharma market.
Lilly entered GLP-1 market with Zepbound, which added an additional weight loss hormone (GIP) and outperformed in head-to-head trials (patient weight loss of 20% vs. 14%).
Lilly found new sales channels earlier (sold low-dose Zepbound online with patients supplying syringes) and partnered with DTC firms first (eg. Hims & Hers, which Novo actually just ended a partnership with because Hims & Hers war marketing generic GLP-1s too aggressively).
Novo is working on an oral GLP-1 pill but patients have to fast before taking it (conversely, Lilly’s oral pill in-the-works can be used right away).
A month of Zepbound ($1,000) is currently cheaper than WeGovy ($1,350).
Those strategic mistakes were compounded by what has to be one of the most costly administrative errors ever. In 2018, someone at Novo forgot to pay a $450 maintenance fee to extend the GLP-1 patent in Canada for a few more years. The American patent is still good until 2032 but the Canadian patent will soon expire. Canada is Novo’s 2nd largest market for the “fat shot” and many Americans buy the drug from up north.
This could easily be a 9 to 10-figure mistake. I hate auto-renews with a passion (I hate you Stamps.com). But this seems like one time clicking the “auto-renew” box would have been smart.
In 2024, Novo’s WeGovy did sales of $8B vs. Lilly’s Zepbound at $5B. The Economist says that by 2030, Lilly is expected to have 47% of the $90B+ GLP-1 market while Novo is at 40%.
RIP Fred Smith, FedEx Founder
The founder of FedEx died last weekend at the age of 80.
His story is absolutely wild.
Smith grew up in a well-to-do Memphis family. His father had founded a motor company later acquired by Greyhound Bus. After high school, Smith attended Yale University at the same time as both US Presidential candidates for the 2004 election (George W. Bush and John Kerry; he was friendly with both).
While studying Economics at Yale, Smith worked as a charter pilot and wrote a term paper pitching the idea of an overnight commercial air delivery business. He received a C- on the paper but would come back to the idea.
In the interim, he served in the the Vietnam War as a platoon leader for the Marines and as a forward air controller. One of his good friends died in his arms and Smith would later receive two Purple Hearts for his tour.
Back in America, Smith took a 7-figure family inheritance acquired a controlling interest in an aircraft maintenance company. He pivoted the business to trade used jets before launching FedEx in 1971 based on that C- term paper (take that, tenured prof).
The business pitch was quite ingenious:
Overnight shipping — particularly from East to West coast — was logistically difficult as America’s airport infrastructure was still developing.
Memphis is near the geographic centre of America and could serve as a connecting hub to deliver parcels across the country.
Another benefit of Memphis is that the weather is quite moderate, which allows for predictable flight plans.
FedEx would use Memphis as “superhub” to efficiently sort incoming parcels and send them out.
Spying an opportunity, the City of Memphis was willing to help Smith’s vision come to light.
While the idea made sense, the entrepreneurial journey was never easy. Even with a starting nest egg, FedEx was such a capital heavy business. When operations launched in 1973 — with delivery to 25 cities — Smith had bought 14 Dassault Falcon 20 (DA-20) jets.
That’s a lot of fuel just as the Arab Oil Embargo hit.
In his first 26 months in business he racks up $29 million in losses. Desperate to pay bills [and down to the final $5,000] he flies to Las Vegas, wins $27,000 at blackjack and wires it back to FedEx. Yet by 1976 FedEx is flying smoothly and bringing in $75 million in revenue, though it remains heavily indebted. It goes public in 1978.
Incredibly baller but — for the record — none of this is life, financing or professional advice.
Smith would later say, “The $27,000 wasn't decisive, but it was an omen that things would get better.”
At age 35, Frederick Wallace Smith was in deep trouble. His dream of creating a Federal Express had become too expensive and was fast fizzling out. He had exhausted his father’s Greyhound Bus millions. He was in hock for 15 to 20 million more. He appeared in danger of losing his cargo jet planes and also his wife. His own board of directors fired him as CEO. Now the FBI accused him of forging papers to get a $2m bank loan and was trying to send him to prison. He thought of suicide.”
In North America, these wild entrepreneurial stories are mostly from generations past (and when we do see them, it’s usually from immigrants).
I think a major reason is the phenomenon for millennials and younger generations of people picking career tracks and specializations at an earlier and earlier age.
This isn’t just for the most obvious careers such as doctors, lawyers, engineers and bankers etc. Even the arts. You have kids wanting to be a musician or actor from a young age and spending their entire childhood and teens trying to “make it”.
Take country legend Johnny Cash. Before and during his early music career, Cash worked at an auto factory, picked cotton, did door-to-door sales, and was a radio intercept officer in the Air Force.
Obviously, there are a ton of factors leading to this and here some some speculative bullet points:
More competition in everything (higher levels of population growth and education attainment have outpaced the number of opportunities available in the aforementioned industries; aka Boomers won't retire).
The low-hanging fruits of an industrializing society have already been picked.
Majority of wealth in past few decades is from financiali-zation and software, less the physical world…and you’ll just have more crazy stories living in the physical world than on your fake laptop job (although, hard tech is making a comeback and AI is taking all white-collar work, so maybe are pivoting society back to more EQ in-person stuff...I dunno, I'm just spitballing here).
On the financialization point: in 2025, there would literally be 10,866 financing options other than a bank loan that Smith could have tapped before hitting a Blackjack table (eg. venture, corporate debt, strategic investments, crowdfunding, memecoins, OnlyFans).
So yeah, that’s why stories like Smith and Cash will be rarer and rarer.
3 Thoughts About The NBA Finals
The Oklahoma City Thunder defeated the Indiana Pacers to win the NBA Finals last week.
My ability to watch full NBA games basically went to zero as soon as we had our kid. Now, most of my consumption comes from YouTube player highlight packages, my go-to source for procrastination when I’m supposed to be writing this newsletter.
Anyway, here are three general storylines that caught my eye:
1. “The OKC Thunder Championship is Gen Z's Non Drinking Moment”: A major consumer trend over the past decade is people — especially younger generations — drinking less booze. The OKC Thunder were the second-youngest champions ever with an average age under 26.
The numbers [on the decline of alcohol consumption] matter, in a cumulative way, but you can’t really see the difference without the aid of math. If the young are really drinking less, it’s under society’s surface, tucked away from where someone my age might notice.
At least I felt that way until the Oklahoma City Thunder won the 2025 NBA Finals on Sunday night. Right there, right then, we were treated to a generational shock.
We had Chet Holmgren pronouncing “Michelob” in a hilarious phonetic manner. We had OKC big Isaiah Hartenstein relaying that Alex Caruso had to explain how to open a champagne bottle to his younger teammates. Then we had indicators that the similarly named Jalen and Jaylin Williams both were trying booze for the first time, absent much enthusiasm.
In theory, this should be good, because alcohol is, on balance, pretty bad. Addiction aside, it makes you sick, but compensates by also making you fat. It kills your brain cells, but compensates by hurting your musculature. Booze has derailed more basketball careers than we even know, because not everyone’s been open about their struggles. The NBA lifestyle is lonely and guys have long opted for a depressant to wind down with after that nighttime adrenaline. You still hear about this happening to certain players, but modern technology has at least allowed them to find sex absent buying out bars. Today’s athlete is less of a drunk. So if this seems self evidently great, why does it also seem like the older generations don’t completely love it?
The online reactions to the muted and dry-ish Thunder celebration weren’t on balance positive. Friends of mine, mostly in my generational cohort, weren’t charmed by what they perceived as sterility when the scene called for raw humanity. Where was the catharsis? Where were the bottles?!
***
2. Tyrese Haliburton Deletes Social Media: Another relatable trend was the affect of social media trolls (just the worst). This story concerns Pacers star Tyrese Haliburton, who had been on a Cinderella run leading his underdog team to the NBA Finals before a brutal achilles injury knocked him out of the deciding Game 7. It was an incredible performance after the rest of the league voted him “most overrated player” in the NBA.
Last summer, Haliburton was on Team USA when they won Gold but didn’t play much and was also hobbled by a minor injury at the time. Fans were roasting him online and he says he read all of it. He took the negative energy into the season and didn’t play well for weeks.
Delete social media and have a great season? Feels like there’s a lesson in there.
***
3. Sam Presti and Long-Term Thinking: For non-basketball fans, the Thunder had one of the best young teams in the early 2010s including three future MVPs (Kevin Durant, James Harden, Russell Westbrook). They made the 2012 NBA Finals and the future looked incredibly bright. That team lost and the team broke up in the next few years.
Fast forward a decade and the Thunder won with a new incredible trio (Shai Gilgeous-Alexander, Jalen Williams, Chet Holmgren).
These two rosters were constructed by Thunder GM Sam Presti. He was previously a front office prodigy with the San Antonio Spurs and took over the OKC job at 29 when they were still the Seattle Supersonics in 2007. Since then, he’s frequently cited as the best GM in the business.
“We can't be reactionary or emotional about it, we just have to keep chipping away every single day, knowing that over time, we will achieve our goals if we have the poise, the patience, and the willingness to adjust to the setbacks that we will inevitably encounter. As they say, shortcuts cut long runs short, and we're going to do everything in our power not to allow that to happen.
“Shortcuts cut long runs short”. That goes hard and the Thunder are indeed one of the best-placed organizations for years to come. DEF a lesson in there.
Apple’s $40m Sponsorship for F1 film
For its “F1” film, Apple sold $40m of sponsorships for the fictitious race team, APXGP (which offset the $200m+ budget).
According to Forbes, Expensify was title sponsor with sidepod car placement and logo on front of the racing suit.
How much did Expensify pay? It didn’t say but main sponsors are typically 25-50% of the total sponsorship take.
That puts Expensify’s spend at $10-20m (other brands in the film include IWC, Geico, EA Sports, Mercedes, Tommy Hilfiger).
The expense and spend management SaaS product saw sales of $140m in 2024.
While the F1 sponsorship pencils out to 7-14% of sales, Expensify CFO says the commitment was over multiple years but has not…err…expensed the “F1” deal in its filings yet (and when they do, the “F1 expense will be large, but not a big impact on cash flow.”)
For comparison sake, here are priciest actual annual F1 sponsorship deals:
Oracle ($90m to Red Bull)
Aramco ($75m to Aston Martin)
Petronas ($70m to Mercedes)
Expensify CEO says the F1 film is “one of the best brand placement opportunities ever” and one main reason they were able to get it: Apple wouldn’t allow brands competing with its own products to bid.
The CEO might be right.
There are photos everywhere of Brad Pitt repping the logo. Damson Idris wore the race suit to Met Gala.
Tens of millions of people will watch the film in theatre and streaming and ask “yo, who dafook is Expensify?”
Example: Expensify saw a 4x increase in product sign-ups the day after Met Gala.
Forbes has some great history of film sponsorship, which only recently has been used to offset production budgets:
Paid sponsorships were still a relatively new phenomenon, and the money was fairly small. Famously, Steven Spielberg had originally written M&Ms into the script for 1982’s E.T., but Mars, the parent company of M&Ms, passed on the opportunity. So the producers approached Reese’s Pieces after Hershey’s agreed to pay $1 million in a cross-promotional deal. Within a few weeks of the movie’s release, sales of Reese’s Pieces reportedly tripled.
In another famous example, Ray-Ban saw a spike in sales after reportedly paying $50,000 per year to put its glasses on dozens of TV and movie characters, including Tom Cruise in 1983’s Risky Business. Racing movies, even then, were the biggest prize, with Exxon reportedly paying $300,000 in 1990 to be featured in the Cruise-led Nascar movie Days of Thunder.
Leener says the product placement business really took off in the 2000s, when the commercial potential of blockbuster movies was realized across the globe. He worked on National Treasure, Herbie: Fully Loaded, Pirates of the Caribbean: Dead Man’s Chest and eventually the Transformers franchise, which he says unlocked the multimillion-dollar earning potential of brand integrations.
In one memorable example, Leener recalls pitching a Chinese milk company to Bay for Transformers: Dark of the Moon, using the absurdity of the product’s plastic straw to create a funny moment with actor Ken Jeong.
After realizing the product placement money he was bringing in was going into the studio’s coffer rather than the actual production budget, he helped negotiate for Bruckheimer and Bay to be able to spend against any money he raised in product placements. It’s a contract stipulation Leener says has spread to all A-list filmmakers and further aligned the incentives to create these branded moments.
On a big-budget movie, for example, a day of production might cost anywhere from $200,000 to $400,000. If Leener sells a single three-second placement, it could mean an additional day of filming to perfect an action scene.
The practice has become particularly important in recent years, with movie budgets rising to astronomical levels. Summer blockbusters routinely cost $250 million or more to produce, plus another $100 million or more to promote, making it very difficult for a movie to turn a profit.
We’ve come a long ways from Tom Cruise’s NASCAR flick Days Of Thunder. In that 1990 film, Cruise’s Cole Trickle wore a race jacket sponsored by Melllow Yellow (!!!) while Exxon paid the studio $300,000 for some product placement.
F1 is tracking quite well for Apple and should later lead to Apple TV Plus sign-ups. If this film doesn’t hit in theatres, the iPhone maker may pull out of making big-budget films and that would be very bad for Hollywood.
All in all: pretty hilarious that other than $3T Apple, the company that may benefit most from film will be a $225 fintech SAAS firm.
Liam Neeson is re-booting The Naked Gun…and, damnit, this trailer is funny AF. It’s old school dumb 1980s-90s satire humor that Leslie Nielsen perfected. Nielsen pokes a lot of fun at his Taken persona.
This 10-part podcast on Fidel Castro…is very worth your time, especially if you have a long road-trip coming up and need some incredible Cold War history content via Noiser’s Real Dictators show(one wild fact: Castro tried to get Che Guevara out of his hair by briefly appointing him to run the equivalent of Cuba's Central Bank).
Aaron Sorkin will write and direct a sequel to The Social Network…the original was almost perfect (Quentin Tarantino called it the best film of 2010s). It’ll be about the 2020 election and Facebook’s broader effect on teen mental health. David Fincher won’t be directing and there’s no word on if Jesse Eisenberg will return. Honestly, there’s really no reason for this sequel to exist, so just enjoy this video from Insider explaining how Sorkin made dialogue in his original script “musical”.
New York City Mayoral Upset: 33-yearDemocratic Socialist Zohran Mamdani won the Democratic primary. Lots of ink being spilt on how New York — the global centre of finance — may elect a socialist candidate this fall. Two takes I’ll flag are this one on how Mamdani used video so well and this e-mail from Peter Thiel to Mark Zuckerberg in 2020 predicting the rise of socialist politicians in America.
…and here some extra memes for your weekend meme needs (including this first one based on news that Denis Villenueve — director of Sicario, Arrival, Blade Runner 2049 and new Dune Franchise — will direct the new James Bond film for Amazon):
