ExpertsFM(Preview) Three Weeks of Meta Momentum, The Ultimate Muse Upside, A King Kong vs. Godzilla Aggregator War
Sharp Tech with Ben Thompson · September 25, 2026
5 Selected Lessons:
heuristic · 2 reasoning_pattern · 1 framework · 1 case · 1
The 'number one player' slotting effect
Source
Sharp Tech with Ben Thompson — (Preview) Three Weeks of Meta Momentum, The Ultimate Muse Upside, A King Kong vs. Godzilla Aggregator War (September 25, 2026) · at 15:09 ▸ · at 17:34 ▸
When this applies
when a team, portfolio, or lineup of offerings suddenly feels much stronger after adding or fixing just one piece
The takeaway
Look for the single missing 'anchor' element whose absence forces every other component to overperform its natural role; fixing that one gap can transform the whole system's perceived quality.
Why it works
Without a genuine 'number one' (the element that organically draws demand/pull), every other piece is miscast into a role above its natural level, degrading the whole; adding a true number one lets everything else slot down into the role it's actually suited for, multiplying apparent quality.
One level up
System quality is often bottlenecked by role-fit, not by the sum of individual component quality.
Where it stops holding
This stops working when the system's weaknesses are distributed and structural rather than concentrated in one missing anchor role — in that case no single addition will cause a cascading improvement.
Detail
procedure:
- Identify whether the system has a genuine 'number one' — the piece that attracts pull on its own.
- If missing, recognize that every other piece is being asked to play a role above its natural fit, making the whole system look worse than the sum of its parts.
- When the number one arrives, let other pieces drop back into their natural supporting roles (number two, three, four...).
- Reassess the whole system — it should look qualitatively better even though most components didn't change.
benchmarks: {}
rationale: In basketball, adding one true number one player lets a number three stop pretending to be a number two, etc., and the team gets way better not just additively but structurally.
In their words
So the problem is that if you're missing your number two, then your number three seems worse because he's a bad number two. Your number four seems worse because he's a bad number three. Like, all the way it goes all the down the chain. You you could slot just one guy in, and a team gets massively better because everyone is now in their proper position.
Hardware diversification risk for horizontal platform companies
Source
Sharp Tech with Ben Thompson — (Preview) Three Weeks of Meta Momentum, The Ultimate Muse Upside, A King Kong vs. Godzilla Aggregator War (September 25, 2026) · at 13:01 ▸
When this applies
when a company whose core advantage is serving everyone across a horizontal market considers building a vertically-integrated product
The takeaway
A horizontal business entering a vertical category risks converting partners into competitors, sacrificing the breadth advantage that justified its horizontal position in the first place.
Why it works
Owning hardware forces a horizontal company into direct competition with the hardware/device makers who were previously its distribution partners, undermining the neutral-platform position that made it valuable to everyone.
One level up
Strategic moves should be evaluated by whether they preserve or destroy the structural relationships that created the original advantage.
Where it stops holding
This stops working when the vertical category is foundational enough that controlling it becomes necessary to survive, not just advantageous.
Detail
approach: Ask whether a diversification move puts the company into competition with entities who should remain its partners.
signals:
- company's core identity is horizontal/service-based
- new initiative requires building proprietary hardware/devices
- existing ecosystem partners make the complementary hardware
rationale: Ben notes this predicted downside 'all of which happened, by the way' regarding Meta and Reality Labs.
In their words
my discomfort with and issue with Meta doing this whole space and spending all this money and doing hardware when they're a horizontal services company that should be serving everyone and hardware diverts you and sets you into competition with folks that should be your partners. All of which happened, by the way.
Sell hassle elimination, not productivity
Source
Sharp Tech with Ben Thompson — (Preview) Three Weeks of Meta Momentum, The Ultimate Muse Upside, A King Kong vs. Godzilla Aggregator War (September 25, 2026) · at 10:37 ▸
When this applies
when deciding how to position any new capability-enhancing product to end users versus enterprises
The takeaway
Match your value proposition to what the buyer is actually optimizing for: businesses optimize for output metrics, individuals optimize for reduced hassle.
Why it works
Enterprises are organizationally rewarded for productivity gains, but individual consumers are not motivated by abstract productivity — they respond to removal of friction and annoyance in daily life.
One level up
The same underlying capability needs a different story depending on who bears the cost of adoption.
Where it stops holding
This stops working when the buyer genuinely is an enterprise or professional whose success is measured by productivity metrics, in which case selling 'convenience' undersells the value.
Detail
rule: Don't sell people on being productive; sell them on making annoying things in their life go away.
threshold: Enterprises want productivity; individual consumers want convenience/hassle removal — treat these as two separate pitches.
rationale: People don't intrinsically want to be 'productive' the way organizations do; they want friction gone.
In their words
People don't want to be productive. Enterprises want to be productive. What you sell them on is making annoying stuff in their life go away. Making life more convenient.
Live, unscripted events beat polished marketing videos
Source
Sharp Tech with Ben Thompson — (Preview) Three Weeks of Meta Momentum, The Ultimate Muse Upside, A King Kong vs. Godzilla Aggregator War (September 25, 2026) · at 7:38 ▸
When this applies
when choosing between a live/unscripted format and a polished, risk-free produced format for communicating to an audience
The takeaway
Formats that carry real risk of failure are more compelling to audiences than risk-free, over-produced formats, even when the produced format showcases a 'better' product.
Why it works
The possibility of failure in a live format creates genuine stakes and attention, while a risk-free produced video removes tension and invites disengagement.
One level up
Perceived authenticity and stakes drive engagement more than production polish.
Where it stops holding
This stops working when live execution actually fails (a botched demo), which can destroy credibility rather than build engagement.
Detail
rule: Prefer live, time-boxed, audience-present formats over polished pre-recorded marketing videos when the goal is sustained audience attention.
threshold: A tight ~50 minute live format with successful demos was enough to fully hold Ben's attention, while Apple's produced videos lost him within ~30 seconds.
rationale: Live's inherent risk of failure makes it 'compelling' in a way safe, produced content is not.
In their words
Live is so much better. Like, I just blew through it. Like, wait a I I was gonna watch it. You know, I just started watching at night. Locked in. Just watched the whole thing. Why? Because live's compelling. Apple puts out these marketing videos. I'm, like, drifting off into space within, like, thirty seconds.
Reframing glasses as 'Muse delivery devices'
Source
Sharp Tech with Ben Thompson — (Preview) Three Weeks of Meta Momentum, The Ultimate Muse Upside, A King Kong vs. Godzilla Aggregator War (September 25, 2026) · at 17:34 ▸
When this applies
when a hardware or infrastructure product only becomes compelling once paired with the software/experience that people actually want to pull toward them
The takeaway
Reposition a hardware product not as a category exemplar but as the delivery vehicle for the software/experience that already has organic pull.
Why it works
People don't want 'AI glasses' or 'VR glasses' as abstract categories; they want the agent/experience inside, and the device becomes desirable only as the delivery vehicle for that pulled experience.
One level up
Products are not sold in isolation; a weak standalone product can become the housing for a strong pulled product, inheriting its demand.
Where it stops holding
This stops working when the underlying software/experience itself lacks genuine consumer pull — a delivery device for an unwanted service doesn't inherit demand it doesn't have.
Detail
situation: Meta had VR headsets and AI models, neither of which was a true demand-driving 'number one' product.
decision: Meta positioned its glasses (VR and AR) as vehicles for delivering the Muse AI agent rather than standalone AI/VR hardware.
outcome: The keynote and product lineup felt focused and exciting; glasses 'without a camera' and other quirks suddenly made sense as features of a Muse delivery device.
lesson: A hardware category without a compelling standalone pitch can be reframed around the one product people actually want (the agent), making the whole lineup suddenly cohere.
In their words
And once you have Muse and Muse is the thing, now you're not selling VR glasses. You're not selling AR glasses. You're selling Muse delivery devices.