Sea, MercadoLibre, Amazon, and Shopify are not just four companies. They are four different answers to the same question… what kind of platform still deserves a premium when the world gets more competitive, more automated, and less willing to subsidize growth?
The market is no longer asking whether you can grow. It is asking how much it costs you to keep growing once the easy capital is gone.
Good Numbers, Wrong Signal
Sea is the cleanest example of this shift because the numbers are not actually bad. Revenue grew 36.4% to $22.9 billion. Net income came in at $1.6 billion. Shopee moved $127.4 billion in GMV. A few years ago, those numbers would have been enough for investors to look right through temporary margin compression and keep paying up for the long arc. Instead, the stock has been punished because the market is no longer paying for growth in the abstract. It is paying for durable economics under pressure.1
That distinction matters.
Operating expenses rose more than 28% in the fourth quarter. Shopee marketing spend was up 25.4%. SeaMoney’s sales and marketing nearly doubled. The market saw what it is starting to see across a lot of modern platforms: yes, the machine is growing, but how much of that machine still depends on constant subsidy, incentives, and spending just to hold the line?2
Great topline growth, but the profit story depends heavily on a single gaming franchise, and Shopee hasn't yet proven it can generate durable margins at scale.
Once you see that, you understand Sea is not an isolated story. It is one node in a larger repricing.
Spending That Looks Like Entrenchment
MercadoLibre is useful because, on the surface, it is doing some of the same things. Still spending heavily. Still leaning into logistics. Still taking margin pressure to deepen the system. Yet the market treats it very differently, and that tells you everything. MercadoLibre ended 2025 with $28.9 billion in revenue, $2.0 billion in net income, $65 billion in GMV, and $278 billion in TPV. That is not just commerce. That is ecosystem density. Marketplace, payments, merchant rails, credit, fulfillment, and regional presence. Investors tolerate margin pressure there because the spending looks less like defense and more like entrenchment.3
Sea trades at roughly 2.1 times revenue. MercadoLibre trades near 4.1 times. That gap is not an accident. The market is making a judgment call about who owns an ecosystem and who is still renting one quarter at a time.4
This is what multiple compression looks like in practice. It is not a verdict on whether the company is real. It is a verdict on whether the market believes the economics are durable.
Owning the Rails
Amazon adds another layer. Its 2025 revenue was $716.9 billion and operating income reached $80.0 billion, but the market is not valuing Amazon as retail. It is valuing a composite machine. AWS grew 20% in 2025 to $128.7 billion, and 24% in the fourth quarter. Advertising reinforces the system. Logistics is no longer just a cost center. It is part of the moat. Amazon gets paid for infrastructure because infrastructure is where the market still sees leverage, necessity, and staying power.5
At about 3.3 times revenue, the multiple is not as visually dramatic as Shopify’s, but the revenue base is enormous and mixed across very different businesses. The important point is that Amazon’s premium comes from owning rails that others depend on, not from promising future optionality while burning money to preserve a customer habit.
The Right Side of the Interface Shift
Then you get to Shopify, and this is where the future starts to become more obvious.
Shopify did about $11.6 billion in revenue in 2025, moved $378.4 billion in GMV, and generated more than $2.0 billion in free cash flow. Ten consecutive quarters of double-digit free cash flow margins. That is an entirely different type of platform from Sea. Shopify is not carrying regional fulfillment wars. It is not trying to be the entire consumer destination. It is enabling merchants, sitting in the software and orchestration layer, and increasingly positioning itself where AI-native commerce is heading.6
That is why the market gives it the premium. Near 16 times revenue, Shopify is priced as a company on the right side of the next interface shift. That does not mean it is cheap. It means the market believes it is capital-light enough and strategically central enough to deserve it.
And now the broader point comes into focus.
The New Filter
We are watching the market sort platform businesses by what they cost to defend.
That is the new filter.
Not who has users. Not who can still put up top-line growth. Not who can produce one good earnings report.
Who can grow without constantly re-buying their own momentum? Who owns infrastructure versus who rents attention. Who benefits when software gets cheaper, agents get smarter, and the cost of orchestration keeps falling?
Sea is being discounted because too much of its profit still looks contingent. MercadoLibre is being rewarded because its reinvestment deepens a regional operating system. Amazon is being rewarded because it owns infrastructure with operating leverage. Shopify is being rewarded because it is asset-light, merchant-aligned, and increasingly native to the next commerce interface.
The winners in this cycle will not be the platforms that grew the fastest during cheap capital. They will be the ones whose economics improve as intelligence gets cheaper.
AI Changes the Discount Rate
This is why I keep coming back to the same concern around AI, software, and the startup stack more broadly. The world is moving into a regime where the market is much less tolerant of businesses that require a lot of human labor, promotional spend, or operational heaviness just to preserve their current position. AI is not just creating products. It is changing the discount rate on business models.
If intelligence becomes cheaper, software that removes coordination cost gets stronger. Infrastructure that everyone plugs into gets stronger. Enablement layers that sit above fragmented channels get stronger. But businesses that still depend on endless acquisition, subsidy, or physical complexity to hold market share start to look worse, not better.
That is why Sea matters beyond Sea.
It is a case study in what happens when a company improves materially, and the market still says, that is not enough.
Prove that the improvement is durable.
- For founders, growth will not save you if the market decides your growth is expensive to maintain.
- For investors, reported numbers are no longer enough. You have to understand the structure beneath them.
- For anyone building in AI, the market is beginning to reward businesses in proportion to how much they benefit from falling coordination cost, and punish those that remain trapped in high-maintenance operating models.
Sea, MercadoLibre, Amazon, and Shopify are four versions of that truth. They are not being valued on size. They are being valued on whether their economics get stronger or weaker as the world gets harder.
And that is the real repricing now underway.