Research · November 10, 2026

MercadoLibre (MELI) Blitzscaling Score: 70/100 (B) — Moat, Speed & Risk

MercadoLibre scores 70 out of 100 (grade B) on the BlitzLab 100-point model, comfortably inside the 65+ Club. It is, on the evidence, the textbook blitzscaling case in Latin America: thirty consecutive quarters of 30%+ growth, a commerce business and a fintech business accelerating at the same time, and a two-sided network that keeps getting denser. The score is not higher for a specific reason — the company is deliberately sacrificing margin to buy that growth, and the market is asking when the leverage turns.

The score at a glance

AxisScoreWhat it measures
Speed (A)17 / 25Growth, capital velocity, productivity, share gains
Moat (B)42 / 60Network effects, distribution, market size, gross margin
Creative destruction (C)13 / 15Order displaced, irreversible behaviour change, cost-structure change
Demerits (D)−2Debt, revenue quality, governance, regulatory cost
Total70 / 100 (B)Grade B is 60–79. Both gates pass. No FLAG.

On the radar axes the score converts to moat 70.0, speed 68.0 and creative destruction 86.7 (each on a 100-point scale). The shape of the score is the story: creative destruction is the strongest axis, the moat is broad, and speed is held back less by growth itself — growth is the strongest single item in the whole score — than by the capital and productivity items around it.

What MercadoLibre does

MercadoLibre, Inc. (NASDAQ: MELI) runs the leading e-commerce and fintech platform in Latin America, operating in 18 countries. The business has two engines that report as one company. Commerce — the marketplace — did $5.8 billion of revenue in the second quarter of 2026, up 50% year on year. Fintech — Mercado Pago, the payments, credit and asset-management arm — did $4.4 billion, up 49%. Together they produced $10.169 billion of quarterly revenue, up 49.8% year on year, the first quarter above $10 billion and the fastest growth rate in four years. Trailing-twelve-month revenue is about $35.2 billion, up roughly 46% year on year.

The scale underneath the revenue is what makes the two engines a flywheel rather than two divisions. Gross merchandise volume was $21.9 billion in the quarter, up 36% on a foreign-exchange-neutral basis. Total payment volume through Mercado Pago was $101 billion, up 56%. There were 89 million unique buyers, up 26% year on year, buying 795 million items, up 45% — and items per buyer rose 14%, which is the signature of a marketplace whose liquidity itself is becoming the product. Mercado Pago had 88 million monthly active users, up 30%, processing 5.2 billion transactions in the quarter, up 44%. Assets under management reached $23 billion, up 68% year on year, and the credit-card portfolio reached $7.7 billion, up 91%.

That combination — a marketplace that generates the traffic and a payments and credit business that monetises the same users — is the reason the company clears both of BlitzLab’s gates. The classification is MATURE: at roughly $35 billion of trailing revenue, this is no longer a young blitz; it is a large company still growing like one.

The gates: both pass

Gate G1 asks whether growth continues or a second growth curve exists. It passes, definitively. Thirty consecutive quarters of 30%+ growth is, among large listed companies, a record without an obvious peer. The two engines are accelerating together rather than one masking the other’s slowdown, and advertising — up 62% foreign-exchange-neutral — is emerging as a third, high-margin layer on top of the distribution the company already owns.

Gate G2 asks whether the ten-year ceiling of the new market is at least ten times current revenue — a bar of about $352 billion against trailing revenue of $35.18 billion. It passes on an estimate. E-commerce penetration in Latin America is still low relative to developed markets, and the unbanked and underbanked population is large; the combined commerce-plus-payments-plus-credit-plus-advertising market has room to clear the bar, though no precise TAM figure was secured and the verdict depends on how the market is defined. That definitional sensitivity is recorded as a model note, not hidden.

Speed (A): 17 of 25 — growth is perfect; the rest is not

Growth (A1) scores a full 10 out of 10. A $10 billion-plus quarter growing at nearly 50% is a maximum-grade result at this scale, and the thirty-quarter streak is the evidence that it is a system rather than a quarter. Share velocity (A4) scores 5 out of 5: MercadoLibre is the clear number one in Latin American e-commerce and fintech, with buyers up 26% and items sold up 45% while Amazon and Sea Limited are increasing their investment in the region — the leader is accelerating into the competition, not away from it.

The other two speed items are where the points leak. Capital velocity (A2) scores 2 out of 5. The market capitalisation is about $94 billion, and a $1 billion bond issued in September 2026 at 5.85%, maturing in 2036, confirms access to capital — but the share price was down 11–12% over twelve months and 23% below its 52-week high of $2,428, the opposite of the doubling the item rewards, and there is no buyback or dividend signal to point to. Hiring and productivity (A3) scores 0 out of 5: about 84,207 employees produce roughly $418,000 of revenue per head, below the $1 million bar. That is in part a structural fact about a logistics-heavy model — moving physical goods needs people in a way software does not — and it is recorded as a candidate for a model revision, but under the frozen v1.0 rules it scores zero.

Moat (B): 42 of 60 — a double network, with a margin cost

Network effects (B1) score 14 of 18, at the top of the anchor band the model sets for this kind of company. The evidence is the two-sided liquidity itself: 89 million buyers, 795 million items, 88 million fintech users and 5.2 billion quarterly transactions, with commerce and fintech reinforcing each other — buyers bring sellers, sellers bring buyers, and the payments arm deepens the relationship with both. Distribution scalability (B2) scores 12 of 18: the company’s own logistics network, MELI Envios, handles 77% of same-day and next-day deliveries within 48 hours, the Mercado Pago app is itself a fintech distribution channel, and advertising growing at 62% shows a high-margin layer being built on top of that distribution. It is not viral, zero-marginal-cost distribution — marketing and delivery investment continues — so the score sits mid-band rather than at the top.

Market size (B3) scores 8 of 12. The Latin American e-commerce and fintech opportunity is large and fast-growing, but at $35 billion of revenue the headroom multiple has compressed relative to a decade ago — the paradox of scale, the same effect the model records for NVIDIA: the bigger you get, the harder it is for the remaining market to be ten times you. Gross margin (B4) scores 8 of 12, and this is where the score’s central tension lives. Gross margin was 40.9% in the second quarter of 2026, down from 45.6% a year earlier — a 4.7-point fall driven by deliberate investment, notably lower free-shipping thresholds in Brazil. Trailing margins sit around 43–48%, respectable for a commerce-centred business, but the trajectory is downward, and that trajectory is what costs the points.

Creative destruction (C): 13 of 15 — the strongest axis

Order displaced (C1) scores 5 out of 5: offline retail and cash-based finance across 18 countries are being replaced by one integrated digital commerce and fintech platform, and the order being displaced is large and clearly identified. Irreversible behaviour change (C2) also scores 5 out of 5: 89 million buyers, 88 million fintech users and a $16 billion credit book mean digital payment and e-commerce behaviour in Latin America has become the default, not an experiment. Cost-structure change (C3) scores 3 out of 5: the company’s own logistics network and digital credit structurally lower the unit cost of commerce and finance, but independent measurement of an industry-wide unit-cost collapse does not exist yet, so the item is marked as in progress rather than proven.

The risks: one demerit, and it is the one to watch

Demerits total just −2, all of it in debt (D1). Total debt is $13.2 billion as of June 30, 2026 — 0.37 times trailing revenue, far below the two-times trigger, with net debt of $6.4 billion and leverage of 1.65 times EBITDA. The debt itself is not the problem. The observation attached to it is: the credit book is $16.4 billion, up 75% year on year — growing faster than revenue at 50% — loss provisions are rising as a share of revenue, and adjusted free cash flow is thin at $214 million. Fintech credit risk, not balance-sheet leverage, is the blitz-failing item to watch.

Revenue quality (D2), governance (D3) and regulatory cost (D4) all score zero — no restatements or audit delays, no abusive dual-class or related-party issues under CEO Ariel Szarfsztejn, and no company-specific regulatory cost or large litigation on record. The macro context is noted rather than scored: a Brazilian presidential run-off was scheduled for October 25, 2026, and Argentina exposure exists, but neither had produced a company-specific cost at the scoring date.

The wider concerns, as the scoring file frames them, are the margin trajectory and the credit book. Operating margin fell from 12.2% to 6.7% in a year — described as intentional investment, but halved all the same. The share price sitting 23% below its high while the company grows at 50% is the market’s way of saying it wants the leverage story before it pays for the growth story again. And competition is intensifying as Amazon and Sea Limited invest more in the region.

What would change the score

Upward, the clearest path runs through margins. If gross margin stabilises and then recovers while growth stays above 30% — the deliberate-investment thesis converting into operating leverage — B4 would rise, and the capital-velocity item would likely follow as the market repriced the combination. A second path is the credit book: if it grows back in line with revenue and provisions stabilise, the D1 observation fades and the −2 could clear entirely. A third is distribution: if advertising keeps compounding at anything like 62%, the high-margin layer on top of the logistics network would strengthen B2.

Downward, the risks are the mirror image. If the credit book keeps outgrowing revenue and provisions keep rising, the −2 in D1 would deepen — that is the single most likely source of new demerits. If margin decline continues without a leverage turn, B4 would fall further. And if growth itself ever broke the thirty-quarter streak, A1 — the full 10 that anchors the whole score — would be the first thing to move. The next quarter’s watch item, in the scoring file’s own words, is the speed of the credit book.

For context, the score has moved a little, not a lot. A blind backtest put MercadoLibre at 73/B; the current 70/B is three points lower, with the direction of the judgement unchanged and the intensity slightly weaker. The backtest’s conclusion — on the success side of the line — still holds. What has changed is the question the market asks: not “how fast can it grow?” but “when does it get its margin back?”

---

Sources