It's that time again. Come chat with the Stripe team about building on Connect! Got questions about specific how-tos or pain points? Want to request the feature of your dreams?

Join us on Wednesday, September 16th, 1pm ET-3pm ET.

Can't wait until then? Drop your questions in the thread now, or pre-game the AMA with the Connect Marketplace Mechanics event in the Stripe Community on Tuesday, September 15th!

Stripe community builders around the world are hosting the first "Build with Stripe Community" event series, centered on one shared topic: Simplify building your tech stack with AI.

At each hands-on event, you’ll learn how to provision a development stack from your terminal using Stripe Projects, then put it to work in a hackathon.

Find an event near you:
• Lucknow—September 8
• Dortmund—September 9
• Bogotá—September 12
• Stablecoins—September 16
• AI for Ecommerce—September 19
• Barcelona—September 23
• Apps Developers—October 1

If you build with Cursor, Claude Code, Codex, or another coding agent, you’ll learn a clearer way to move from an idea to a running app.

Btw, Cursor is supporting this Build with Stripe Community hackathon series with $75 in credits for eligible attendees!

Check out the upcoming events 👋

Written by: Abhinav Tiwari, Global Product Lead @ Stripe 👋

As the revenue infrastructure provider for thousands of AI companies, including 88% of the 2026 Forbes AI 50, Stripe data offers a unique view of the AI economy. Our data shows that AI companies are growing their global footprints quickly: the 100 largest by revenue reached 120 markets on average by their third year of operation. But where is market demand actually strongest? Who’s on the other side of the AI growth engine, and what’s the best way to convert their demand to revenue?

To understand, we looked at the transaction data for all AI companies on Stripe and mapped the landscape. We looked beyond simple market size: the countries spending the most on AI on Stripe largely corresponded to high-GDP countries that would make good expansion candidates for any kind of company. More interesting markets emerged when we considered additional signals, like countries with growing AI spending, or AI spending that was disproportionate to their overall payment volume on Stripe.

While it’s easier than ever to sell globally, presence alone doesn’t always translate to revenue. To build a truly international revenue base—like the 100 largest AI companies on Stripe, which collectively draw 48% of their revenue from outside their home markets—it’s often necessary to localize the product and offering. This geographical analysis shows where companies might prioritize launches and investments in local marketing, translations, payment methods, and more. Here’s what we found.

1. Outsized AI spend puts new markets in the spotlight

Unsurprisingly, the 10 markets on Stripe with the highest AI spend—in orange and black on the chart above—are also some of the largest markets on Stripe for general online spending, and generally represent high-GDP countries with high connectivity to global trade.

To identify less obvious high-momentum markets, we looked for atypical spending patterns, identifying the countries with the greatest share of AI spend compared to their overall spend on Stripe. By this metric, India, Mexico, Poland, and the United Arab Emirates all emerged as top markets (in purple). Brazil, Japan, and South Korea were revealed as countries where AI spend is both high in the absolute sense, and high in relation to other online spending on Stripe.

2. Overall market growth is high, with the highest rates found in a mix of small and more established markets

AI spend is growing rapidly in every market we examined on Stripe. Even after limiting our analysis to the 35 markets that had already achieved meaningful AI spend (over $20 million) by 2024, we saw median year-over-year growth of nearly 100%. This means that while many of the largest markets on Stripe grew more slowly than the median, their absolute growth was still quite high: the United States sat at 91% year over year, and Australia at 61%.

Above-median growth was naturally concentrated in already smaller markets, as shown below, but was also found in markets with already considerable spend that show no sign of slowing down.

Canada, Germany, and the United Kingdom all sustained high rates of growth. South Korea emerged as a triple threat: a large market with a high growth rate (134%) and AI spending disproportionate to the country’s overall spend on Stripe. This could reflect a combination of national policies (like the 2026 “Basic Act,” promoting growth amid safety guardrails), a government-supported startup scene, and a population that reports the lowest rate of being “more concerned than excited” about the rise of AI in daily life.

Mexico was the overall standout growth market, with 264% year-over-year growth of AI spend on Stripe. It’s also a market that disproportionately spends on AI, potentially benefiting from its proximity to the US and major investments from US companies. As an expansion target, it offers an opportunity to establish a foothold in a market less crowded than the more established high-growth countries.

3. Launching globally is the first step, while localization drives long-term growth

For many companies, global customer demand drives rapid expansion and growth. At the high end is a company like Manus, which started accepting payments from more than 200 countries and territories just a month after its AI agent platform went viral in early 2025, and achieved a $90 million run rate just four months later.

But we see evidence in our data that localization is needed to sustain high growth and capture meaningful revenue outside a company’s home market. The fastest-growing AI companies on Stripe already use 2x more local payment methods on average than the broader cohort, and previous analyses have shown that surfacing relevant local payment methods can have a significant impact on conversion (up 7.4% on average) and revenue (up 12% on average).

For Gamma, the AI-powered design platform, LPMs have been especially key to success. When Gamma switched on UPI, India’s real-time payment system and protocol, its in-country revenue increased by 22%. Overall, more than half of the company’s revenue currently comes from outside its home market, the United States.

We’ve also seen that offering payment in local currency improves both initial conversion and lifetime subscription value for AI businesses with recurring revenue. In a previous analysis, subscription businesses using Stripe’s Adaptive Pricing feature saw a 4.7% boost on average to initial conversion, and a 5.4% boost on average to lifetime subscription value.

Some AI companies can experience even greater uplift: Runway saw up to 17.7% more lifetime value per subscription after introducing Adaptive Pricing.

What’s next

As AI companies reach the ceiling of global expansion (there are only so many countries in the world), sustaining growth will require laying roots in the places they’ve launched. To plan your global growth strategy and to discover the right infrastructure to support it, get in touch.

Founders in San Francisco: on June 16, join us for Stripe Startups Build Day, featuring Wen Sang, co-founder/COO of Genspark.

Stripe Startups brings together founders and builders across the world through Build Days, a hands-on learning and coworking day where you can meet the Stripe founder community, get 1:1 product support, and see demos of our latest products.

We'd love to see you at the next one in San Francisco.

Link to register here: https://luma.com/cli3p75t

Hey everyone! Stripe Team here. We've noticed this question “Standard vs. Express vs. Custom” resurfaces quite a bit, and given how much has changed since the original post back in 2022, we wanted to share some current information. The original thread is still directionally right but wanted to share a few things we have changed before you make a decision.

The Core Tradeoffs (still accurate)

Standard:
  - Connected accounts have a direct relationship with Stripe (not your platform).
  - Stripe handles compliance, KYC, fraud, and disputes for those accounts.
  - You use "direct charges", the connected account is the merchant of record.
  - You cannot split the payment at charge-time or collect a platform fee on
Standard accounts in the same way Express/Custom allow.
  - Lower operational burden on your platform, but less control.

Express / Custom:
  - Your platform controls more of the funds flow using destination charges
or separate charges + transfers.
  - You can collect an application_fee_amount on each transaction.
  - The platform bears more dispute/fraud liability depending on the charge type.
  - Express gives Stripe-hosted onboarding/dashboard; Custom gives you full UI
control but full compliance/KYC responsibility.

The comment about needing ~$1M+/year in volume to negotiate IC+ (cost-plus) 
pricing is a real ballpark figure for Stripe historically, though that threshold isn't officially published. Reach out to Stripe sales directly for an actual quote.

What's changed since 2022

1. ACCOUNTS V2 (General Availability: December 2025)

This is the most significant change. Stripe has released Accounts v2 as generally available, which replaces the rigid Standard / Express / Custom "type" model with a flexible, configuration-based approach.

Instead of picking a fixed account type at creation, you now configure roles on an Account object:

• merchant  → can accept payments
• customer  → can be used like a v1 Customer object (eliminates duplicate KYC)
• recipient → can receive payouts
• storer    → can hold a Stripe balance

A single account can have multiple configurations. This is a meaningful architectural shift -- you no longer have to create a separate Customer and Account object for the same person, which was a major pain point with v1.

Docs: https://docs.stripe.com/connect/accounts-v2

2. PRICING

The $2/active connected account/month for Express was a real concern in 2022. Connect fees have evolved so verify current pricing at: https://stripe.com/pricing/connect

Do not rely on figures from old Reddit threads for pricing decisions.

3. ONBOARDING

Account Links v2 (July 2025) now supports collection_options and configuration-based onboarding, giving you more control without going full Custom. This blurs the line between what was previously Express vs Custom.

Practical Guidance

IF you're starting a new integration today:

• Read the Accounts v2 docs before defaulting to the old type-based model.
• If you just need simple marketplace payouts with minimal overhead, a merchant-configured v2 Account + destination charges may be the right path.
• If you need maximum control over the UX and onboarding, v2 Custom-equivalent configurations exist.

IF you have an existing Standard integration:

• The OAuth-based Standard flow still works and is still supported.
• V2 migration is not required but is worth evaluating for new builds.

NOTE: OAuth-connected accounts (Standard) and accounts using recipient service 
agreements currently still require v1 flows. v2 is not a universal drop-in replacement yet for every use case.

Relevant docs:

Connect account types (v1 overview): https://docs.stripe.com/connect/accounts

Accounts v2 overview: https://docs.stripe.com/connect/accounts-v2

Migrate existing integrations to Accounts v2: https://docs.stripe.com/connect/accounts-v2/migrate-integration

Connect pricing: https://stripe.com/connect/pricing

Choosing a charge type (direct vs destination vs separate): https://docs.stripe.com/connect/charges

We published our roadmap today at Sessions. Now we want to hear from the people we're building it for: you!

Here's what we're building and where we're headed, but want to hear from you. What are you most excited about? What should we build next? What questions do you have about what's coming?

https://stripe.com/roadmap

Our product team will be here to answer any questions you have starting now until 4:30 ET / 1:30 PT. We'll prioritize the most upvoted questions, so vote up what you most want answered.

Ask us anything. Go.