NGraph

Stripe: company structure analysis (Arche)

S

Stripe

Startup·USAUSA, 2010
60
/100

Society

8/10

Customer

8/10

Self

6/10

13 above the survival line

Survival 47Excellence 75

An Irish brother duo — 19 and 21 — bet that 'a payment API should fit in 7 lines.'

12 criteria

37/60

+

3 axes

22/30

+

Bonus

+1

=

NCG

60/100

Bonus: [Insight] Articulated a DX revolution — 'a payment API should fit in 7 lines.' Pre-traction after Y Combinator S09

Context

Patrick and John Collison (21 and 19) dropped out of MIT and Harvard and joined Y Combinator S09 (summer 2009). The initial name was '/dev/payments.' In 2010 — when PayPal and Authorize.net's 'hundreds of lines of SDK plus a corporate underwriting wall' was the industry default — they shipped 'payments in 7 lines of code.' Sequoia, Peter Thiel, and Elon Musk wrote checks immediately. Customers numbered in the dozens, there was no moat, and from PayPal's perspective the company was a rounding error.

TelesTeles' Insight

In 2010, Stripe had nothing yet. A few dozen customers, no moat, ARR not even a million. All it had was one bet: 'a payment API should fit in 7 lines.' But the density of that bet was different. Patrick was the prodigy who'd already sold Auctomatic at 19; John was a Harvard dropout and math olympian. Into a market PayPal had walled off with corporate underwriting and hundreds of lines of SDK, they brought 'a payment you can curl.' Sequoia and Peter Thiel cut checks instantly — not because of traction, but on the sharpness of the redefinition: 'payments isn't a finance problem, it's a developer experience problem.' Below the 47-point survival line, but the kind of startup that survives via the NCG exception clause: Insight 5 plus Timing 4.

Key to survival

The sharpness of the insight that bet against the industry default (payments lives in underwriting and contracts) with 'payments is changed by developer experience' — and the technical excellence to crystallize that bet into a 7-line API. With zero traction and no moat, this is the entire reason Thiel and Musk wrote checks on the spot.

12 Criteria (37/60)

StrengthsB Problem, C Solution, L Insight

WeaknessesE Traction, F Moat, G Revenue, H Financial, I Efficiency, J Exit

AMarket3/5

The online payments market is huge, but PayPal owned the castle. The SAM — the layer where developers connect directly — still looked thin.

BProblem5/5

Web developers' pain with payment integration was a daily occurrence. By 2010, everyone agreed Authorize.net's experience was awful.

CSolution5/5

Payments in 7 lines of code — dozens of times simpler than existing SDKs. A 10x+ improvement in developer experience.

DTeam4/5

Patrick had already sold Auctomatic at 19 (2008); John was a math olympian. Exit experience, yes — payments domain experience, no.

ETraction2/5

After YC S09, in beta. A few dozen paying customers, ARR still under a million.

FMoat2/5

Just one API. No network effects, no scale economics yet. Technically copyable.

GRevenue2/5

A clean 2.9% + 30¢ take rate. But the cost of payment processing plus regulatory compliance was untested at scale.

HFinancial2/5

A seed round of around $2M. There was a plan, but as of 2010, zero execution history.

IEfficiency2/5

Operating on a small seed — capital efficiency was healthy. Proof at scale was still to come.

JExit1/5

No exit design. IPO or M&A were years away. The founders had already started saying out loud that they refused to rush.

KTiming4/5

The Web/SaaS app explosion. AWS was sticking, and APIs were starting to feel more native than SDKs — the doorway moment.

LInsight5/5

In 2010 they articulated 'payments isn't a finance problem — it's a developer experience problem.' Competitors at the time were built around merchant underwriting; developer-first eventually became the industry default.

See other cases

View all cases

NGRAPH — AI ADOPTION SUPPORT

We start AI adoption by understanding how your business is built

At NGraph, engineers come into your workplace, look at your work, decisions and the data you use, and then build AI in. We don't stop at teaching you how to use it: we leave AI in a form your team keeps using.