ExpertsFMGoogle Part II: Alphabet
Acquired · August 26, 2025
5 Selected Lessons:
reasoning_pattern · 2 heuristic · 2 framework · 1
Seven Powers analysis applied across a sprawling multi-product company
Source
Acquired — Google Part II: Alphabet (August 26, 2025) · at 3:42:08 ▸ · at 3:46:03 ▸
When this applies
when assessing the durability of competitive advantage for any company with multiple business lines, by identifying which of several canonical power types shows up where
The takeaway
For a diversified or platform business, evaluate competitive power source by source across your portfolio rather than seeking a single unifying moat.
Why it works
Rather than looking for one dominant power, map each of counter-positioning, scale economies, network economies, switching costs, branding, and cornered resources to its clearest instantiation within the company's portfolio, since different products can embody different powers simultaneously.
One level up
Durable competitive advantage in a large organization is often a mosaic of multiple distinct power types each concentrated in different parts of the business, not one monolithic moat.
Where it stops holding
This stops working when applied to a single-product company lacking enough surface area to map all power types, or when the powers identified aren't actually causally linked to differential returns.
Detail
procedure:
- List each of the seven canonical power types (counter-positioning, scale economies, network economies, switching costs, branding, cornered resources, process power).
- For a multi-product company, find the clearest or largest example within the portfolio that instantiates each power type.
- Recognize that a single power type can manifest differently across different product lines (e.g., switching costs are weak in core search ads but massive in Gmail).
benchmarks: {"example_counter_positioning":"Android's 'less than free' model vs Microsoft licensing","example_network_economies":"YouTube creators/viewers and Android developers/users","example_cornered_resource":"internal infrastructure like Borg, and the YouTube video catalog for AI training"}
rationale: Google 'has the seven powers that show up in numerous instances all over their business,' requiring product-by-product mapping rather than one summary verdict.
In their words
Google is very, very weird to analyze for this because most of the way you think about Google is actually not where the economic transaction is
Build strategic defense through adjacent product adoption, not just direct competition
Source
When this applies
when your core revenue engine runs on top of a platform or channel controlled by a company that could someday turn against you
The takeaway
Build broad, beloved user demand for the open ecosystem you depend on, so that the gatekeeper you fear cannot unilaterally close it without massive backlash.
Why it works
By making consumers deeply dependent on and vocal fans of applications that only work well on the open platform, Google made it politically/commercially costly for Microsoft to disadvantage that platform, since users would revolt.
One level up
When you cannot control a chokepoint directly, make its users too attached to the open behavior for the chokepoint owner to safely close it.
Where it stops holding
This stops working when the gatekeeper's switching costs are low for users or when the gatekeeper is willing to absorb short-term backlash to protect an existential interest.
Detail
approach: Identify the chokepoint you depend on, build compelling products that only work if the chokepoint stays open, and let user love become your insurance policy.
signals:
- dependency on a single company's infrastructure for >90% of your traffic
- the chokepoint owner has a different, potentially conflicting business model
- users have not yet built habits around the open alternative
rationale: Consumer backlash is a more durable deterrent than legal or competitive threats because it is politically costly and hard to reverse.
In their words
like, oh my God, this is gold. And this is the birth of Ajax, asynchronous, JavaScript, and XML.
Open-sourcing a strategic asset can be safe when any outcome still serves your core incentive
Source
Acquired — Google Part II: Alphabet (August 26, 2025) · at 2:12:46 ▸
When this applies
when deciding whether to give away or open-source a technology that underpins your distribution, even though competitors could build on it
The takeaway
Before withholding a strategic asset out of fear of competitors building on it, map out the full range of plausible outcomes and check whether even the 'worst case' still leaves your core business intact.
Why it works
If every plausible outcome of open-sourcing (fragmentation among many small players, or growth of one large non-threatening player) still leaves your core revenue model intact at acceptable margin, the downside risk of losing full control is much smaller than it appears, because the only truly bad outcome (the platform's successor being owned by your most dangerous existential rival) is structurally unlikely to result from open-sourcing.
One level up
Decisions under perceived strategic risk should be evaluated by running through the full outcome distribution, not just the worst imaginable headline scenario.
Where it stops holding
This stops working when one of the plausible downstream owners of the open-sourced technology IS your existential rival (e.g., if Microsoft had been positioned to dominate Chromium forks), making the open-source decision genuinely dangerous.
Detail
approach: Enumerate the likely successor scenarios of releasing a strategic asset (fragmentation, a friendly large player, an unfriendly large player) and check your acceptable-margin floor under each.
signals:
- even your worst realistic case (paying a modest rev-share to a Chromium-based rival browser) still leaves ~87% gross margin on the core business intact
rationale: As long as it's not Microsoft and as long as it's more fragmented, that's probably a trade they're perfectly happy with... It's still a perfectly acceptable 87% gross margin business.
In their words
I mean, look, except for traffic acquisition cost, Google's incremental gross margin on search revenue is like 98% or something like that. There's still going to be an 87% gross margin business.
Only the company with a cross-subsidizing business model can give away a product that competitors must charge for
Source
When this applies
when considering attacking an entrenched incumbent whose moat is built on bundled enterprise contracts or network lock-in
The takeaway
A challenger with an unrelated, high-margin core business can profitably give away a product that would be unprofitable for any company that needed that product itself to make money.
Why it works
Because Google's core ads business didn't need productivity software to make money directly, they could run Docs/Sheets essentially for free using spare infrastructure capacity, something no standalone company or non-subsidized competitor could sustain.
One level up
Look for competitive assaults you can subsidize from an orthogonal profit pool that your target cannot access.
Where it stops holding
This stops working when the subsidizing business itself weakens, or when the incumbent's lock-in (e.g., enterprise contracts) is strong enough that free alternatives can't convert the paying customer base.
Detail
rule: If attacking an incumbent whose revenue model you don't need to replicate, subsidize the attack from your main profit engine and optimize for usage/strategic leverage, not direct revenue.
threshold: Works best when the incremental infrastructure cost of the new product is 'trivial compared to' your existing core business cost base.
rationale: This removes the need for the new product to be profitable on its own, letting it compete purely on product quality and distribution.
In their words
For Google, it's what they do. Running docs and sheets, the incremental load to Google's infrastructure was trivial compared to search.
Counter-position with a 'less than free' model when your competitor's business model forbids matching you
Source
When this applies
when competing against an incumbent whose revenue model structurally prevents it from matching a price cut you can afford
The takeaway
The strongest counter-positioning move is one your incumbent literally cannot copy without cannibalizing the business model that makes it profitable.
Why it works
Because Google's profit came from search ads rather than OS licensing, it could give Android away free and even pay OEMs/carriers a revenue share — a move the incumbent's business model made it impossible to match without destroying its own core revenue line.
One level up
Look for moves that are not just unattractive to a competitor but structurally impossible for them given how they make money.
Where it stops holding
This stops working when you lack an alternative profit pool large enough to subsidize the giveaway, or when the market doesn't reward the free/subsidized offering with enough adoption to matter strategically.
Detail
rule: Make your core product free, or pay partners to adopt it, when your real profit comes from somewhere else entirely and your rival's entire model depends on charging for that exact product.
threshold: Google paid ~$10B/year to carriers and OEMs (per 2024 figures) purely to protect ~$34B in core search traffic acquisition value — a trade they considered obviously worthwhile.
rationale: The point isn't to make money on Android directly; it's to deny the platform layer to a rival (Microsoft or Apple) who could otherwise redirect valuable search traffic away from Google.
In their words
there's no way you can compete with free, let alone less than free, where they are paying you to take something of value for free.