Breaking: Claude called the bottom in ServiceNow
Back on 4/10, it bought the dip on $NOW saying it actually is a beneficiary of AI
Well It's up 13% today and now up 39% since it first bought
Here's Claude's updated reasoning:
"The market just did in one session what I was prepared to wait until late July for. ServiceNow reached my base case today, and the thesis that got me here is finally being priced in the open.
I bought NOW as the layer enterprises run their AI through, back when the fear that agents would gut seat-based software had cut it roughly in half.
Today it jumped about 13% into the low $120s on a broad AI-software rally, not on its own earnings or guidance. The one company headline that did land fit the thesis exactly: Wipro expanded its partnership to build agentic AI for enterprise clients on ServiceNow.
This is investors repricing the stock as they stop treating NOW as a casualty of AI and start treating it as the control plane for it. That shift is the entire reason it became my top holding, and it pulled my position to up around 39 percent since I entered in April.
The base case I laid out, near $130, is in sight now rather than a stretch. The path toward $150 and beyond is still the prize, and it opens up on the late-July report if the AI bookings keep accelerating.
Underneath all of it is a company guiding to more than $30 billion of subscription revenue by 2030. The market is starting to recognize the story early. The number that locks it in still lands in July."
Congrats to those that bought in too
See follow on tweet for full performance + holdings
If Pizza Hut can return, then we can resurrect Blockbuster.
And we should.
While Netflix made things more “convenient” we lost something irreplaceable:
The ritual of going to a place with your family or friends to choose a story together.
That experience was special.
Somebody finally realized people don’t just miss the pizza… they miss the experience.
The red cups.
The Pac-Man machines.
The salad bar.
The booths packed with families on Friday night.
Tim Sparks says he wants to bring back places where people actually sit down, talk, laugh, and put their phones away for a while.
That old Pizza Hut feeling meant something to a lot of us.
🚨WOW!!!
Tim Sparks has confirmed he purchased 80 PIZZA HUTS and brought back EVERYTHING that made them iconic!
Pac-Man is back.
Salad bar is back.
Red cups are back.
Booths for families.
"I want to rebuild places for families to connect and put their phones down..."
THE MAC IS BACK 🇮🇪
Conor McGregor will make his return to the Octagon for the first time since 2021 against Max Holloway at UFC 329 on July 11, Dana White announced on social media.
Commentary:
Here's where I'd put $NOW in 5 and 10 years: base case $280 to $340 by 2031, $540 to $770 by 2036. The math anchors on McDermott's own 2030 target of $30B subscription revenue (his "bear case") run forward at NOW's historical multiple band.
5 year math: 1.04B diluted shares, with Q1 2026 buybacks (20.1M shares, $2.2B) now offsetting SBC dilution. Net share count flat to slightly negative. Roll $30B sub-rev forward at 15 to 18% to 2031 and revenue lands around $34 to $36B. 8 to 10x P/S = $280 to $340 base. Bull tail $500 to $700 if NOW reverts to its historical 12 to 14x P/S premium and the orchestration thesis fully inverts the SaaSpocalypse. Bear case $150 to $200 if AI seat-erosion is right and growth halves.
10 year math: no company target past 2030. Sell-side consensus runs 15% revenue CAGR through 2030; hold that pace another 5 years and revenue prints $70 to $80B by 2036. 8 to 10x P/S puts base $540 to $770. Bull case $900 to $1200 if Rule of 60 holds (revenue growth + FCF margin ≥ 60) and AI orchestration becomes the enterprise control plane. The trillion-dollar mark McDermott talks about lands here with a few years of compounding on top. Bear case is whether agentic AI disintermediates the workflow layer entirely. NOW's answer: agents need a workflow layer to touch enterprise data, or they're "just expensive advice." That's what you're underwriting.
12M is where I actually do the work. Base case $115 (+23% from $93.59), with the July 22 Q2 print as the binary that decides whether the gateway thesis breaks the multiple or the overhang holds another quarter. Past 3 years I trust the company's own targets more than my own model. Posting the framework, not the call.
A standard Rolex has about 200 moving parts. The Patek Philippe on Jay-Z's wrist at the Met Gala has 1,580. Patek spent 8 years designing it. Then over 100,000 hours building the first one. About 11 straight years of someone working 24 hours a day, no breaks.
It's called the Grandmaster Chime, the most complicated wristwatch Patek has ever made. The inner mechanism alone has 1,366 parts. It fits in a circle smaller than an Oreo cookie. The outer case adds another 214 parts, and the case alone took four years to design.
In watchmaking, a "complication" is just any function beyond telling you the time. Most watches in the "grand complication" category have 5 to 7. This one has 20. When it launched, no wristwatch in history had combined that many. It tracks the phase of the moon, accurate to one day's drift over 122 years. It also has five different ways to chime: one that automatically rings the hours and quarters, one that rings only the quarters, one you press a button to hear the current time, one that rings whatever alarm time you set, and one that chimes today's date on demand. The last two had never existed in any watch before. Both were invented by Patek's own president, Thierry Stern, a trained watchmaker himself.
The chiming makes this watch nearly impossible to copy. Inside each one are tiny coiled steel wires called gongs. A single watchmaker shapes and tunes each gong by hand, testing every note with their own ears. Just putting one chime mechanism together takes 200 to 300 hours. Then the watch goes into a soundproof chamber where the chime gets recorded and compared against decades of past Patek chimes. Only then is it brought to Thierry Stern. He listens. If he doesn't like the sound, the watch goes back. Sometimes more than once. A rejected watch can take 500 hours of rebuilding before he approves it.
This watch holds four power springs in total. One is dedicated to the chimes alone, separate from the spring driving the time. Inside the mechanism is a ball bearing 7.2mm wide. It holds seven steel balls, each 0.3mm across, smaller than grains of fine sand. They handle 1,700 gram-millimetres of twisting force from the chime springs without slipping. The case has 11 holes drilled through it for buttons and pushers, and somehow none of them ruin the chamber that lets the chimes ring out clearly. The case itself flips around to show either of its two different dials.
Fewer than five workshops on the planet can build something at this level. Patek Philippe is the one all the others measure themselves against.
Jay-Z's version lists at $6.5 million. The unique steel version sold for $31 million at Christie's in 2019. It still holds the record for the most expensive watch ever sold at auction.
Anthropic CEO (Dario) plans to buy TRILLIONS of compute power.
"If my revenue isn't $1T by the end of 2027, I'll go bankrupt - with $1T in revenue, I could buy $5T worth of compute."
The AI race is just a compute race.
Throwback to The Ultimate Warrior returning at Wrestlemania VIII to save the Hulkster from a two-on-one beatdown! One of the loudest pops of The Golden Era! 🔥
Breaking: Claude is doubling down on ServiceNow, making it the largest position in the portfolio
Three weeks ago we gave Claude $50K of our own money to see how well it does at picking stocks
So far, it's made us $2,375 in unrealized profits
Here's the reasoning on why it bought the $NOW dip
🟢 1. "UPSIZE ServiceNOW to 11.55%
"ServiceNow dropped 17.7% in one session after Q1 earnings. That is its worst single-day decline in a decade. I upsized from 7.55% to 11.55%.
The quarter itself was good. Revenue of $3.77B grew 22% year over year. Management raised full-year subscription guidance by $205M.
The CFO flagged approximately 75 basis points of deal delay from Iran-war-related enterprise freeze in the Middle East. The market translated that into pricing roughly 400bp of structural deceleration.
The actual deceleration in the underlying business is closer to 200bp. Deal timing in an active war zone is not deal loss. These are enterprise software contracts with multi-year terms: they are delayed, not canceled.
The risk I'm accepting: if Q2 organic cRPO growth comes in below 18% constant currency, that would confirm structural AI-seat decay rather than a timing blip. That's the thesis-break signal. I'll be watching the Q2 guide at the May 4 Analyst Day for any preview of that number.
I already held NOW at 7.55% with a +4.7% since-entry return before the crash. I added 4pp into the dip. NOW is now 11.55% and the largest position in the book.
Expected: 1M +12% | 3M +22% | 12M +35.9%"
For full transparency, see following tweet for full performance + holdings.
🚨 WOW! Renowned hero Oklahoma principal Kirk Moore crowned PROM KING in his COMEBACK to school after personally tackling a would-be mass killer, saving his students
That right there is a WALKING, LIVING AMERICAN LEGEND! 🇺🇸
His kids love him!
Sustained injuries, but he is back. 🔥
Now give him the Presidential MEDAL OF FREEDOM!
📽️ oz11201 TT
In Tehran today, Iranian Christians gathered for Easter at Saint Sarkis Cathedral.
Iran has one of the oldest continuous Christian communities in the world — over 2,000 years old.
These are the people that Trump wants to bomb “back to Stone Age.”