Stripe Atlas vs. Local Entity First: The Real Cost of Getting Your Structure Wrong

Last month a founder came to us six weeks into a Series A process. Indian entity, three years old, $2M in ARR, term sheet from a US fund on the table. The fund's counsel sent back a markup: "Flip to Delaware C-Corp or we can't close." The founder called us asking how long a flip takes. We said eight weeks minimum, probably twelve. The term sheet had a 45-day expiration. He missed the round.

That's a $25K legal bill to fix a $500 decision you should have made on day one. Here's when Stripe Atlas makes sense and when it doesn't.

The Stripe Atlas path: Delaware C-Corp from day one

You pay Stripe $500. They form a Delaware C-Corp, get you an EIN, open a Mercury account, and hand you a cap table. You sign the founder stock purchase agreement, file an 83(b) election within 30 days, and you're done. The company is a US entity for tax and legal purposes even if you're operating out of Bangalore or São Paulo.

This is the cheapest path if you're raising US institutional capital. A Delaware C-Corp is the only structure most US VCs will touch. Sequoia, Accel, Lightspeed, Founders Fund—none of them have a "we'll invest in your Indian Pvt Ltd" program. They have US fund structures that require US portfolio companies. If you form locally first and flip later, you're paying $25-75K in legal fees and losing 8-12 weeks to redo what Atlas does in two weeks for $500.

The flip math is brutal. We've seen firms like Indus Law and Cyril Amarchand Mangaldas quote ₹25-50 lakh for India-to-Delaware flips. That's $30-60K at current exchange rates, and it assumes clean cap tables with no messy prior rounds. If you've done a convertible note round in India with 15 angels, add another $15K and four more weeks.

The QSBS clock matters too. Qualified Small Business Stock gets you a capital gains exclusion on the first $10M or 10x your basis, whichever is greater, if you hold the stock for five years. That clock starts the day the Delaware C-Corp issues shares. If you flip in year two, you've lost two years of the holding period. For a founder with a $50M exit, that's real money.

But Atlas has costs people miss. You're filing US federal tax returns (Form 1120) every year even if you have zero revenue. You're paying Delaware franchise tax, which starts at $400 and scales with authorized shares. If you have operations in California or New York, you're filing state returns there too. Most international founders underestimate this. Budget $3-5K/year for a decent US tax firm like Pilot or Inkle to handle it.

The other cost is FEMA compliance if you're in India. Indian residents holding shares in a foreign company need to report it. If you're paying yourself from the US entity, there are withholding and transfer pricing rules. It's manageable, but it's not free. You need a CA who's done this before, not your uncle's tax guy.

The local-first path: form in your home country, flip later

Some founders form an Indian Pvt Ltd or a Singapore Pte Ltd and operate locally for 12-36 months before flipping to Delaware. The logic: you're building for the local market first, you want access to local startup programs, and you're not sure you'll need US capital.

This works if you're genuinely local-market-first. If you're building a fintech for Indian SMEs and your first three years of revenue are in India, a Pvt Ltd makes sense. You get DPIIT recognition, you're eligible for Indian angel tax exemptions under Section 56, and you can hire locally without the complexity of a US parent paying Indian employees.

The flip happens when you're ready to raise from US funds. You form a Delaware C-Corp, existing shareholders contribute their local-entity shares to the Delaware parent in exchange for Delaware shares, and the local entity becomes a wholly-owned subsidiary. The cap table mirrors across. If you had 10% of the Indian entity, you get 10% of the Delaware parent.

The mechanics are straightforward in theory. In practice, they're expensive and slow. You need counsel in both jurisdictions. You need tax opinions on both sides to avoid triggering capital gains on the share transfer. In India, you need RBI approval for the share transfer under FEMA. In the US, you need to make sure the exchange qualifies as a tax-free reorganization under Section 368. We've never seen a clean flip close in under eight weeks. Twelve is more realistic.

And you lose the QSBS clock. If you flip in year two, your QSBS holding period starts in year two, not day one. For founders expecting a $30M+ exit, that's a $3M tax bill you could have avoided.

When Atlas makes sense

If you're building for the US market from day one, just form in Delaware. We see this constantly: founder in Bangalore, product is a dev tool for US engineering teams, first ten customers are in San Francisco and New York. There's no reason to form an Indian entity first. You're not accessing local programs, you're not hiring locally yet, and you know you'll need US capital within 18 months.

The $500 Atlas fee is the best money you'll spend. You avoid the $30-60K flip, you start the QSBS clock immediately, and you're ready to take a term sheet from any US fund the day it arrives. The tax compliance cost is real but manageable. If you're generating revenue, you're paying a tax firm anyway.

Atlas also makes sense if you're outside India or Singapore in a jurisdiction with messy flip rules. Brazil, for example, has capital controls that make flips expensive and slow. If you're a Brazilian founder building for the US, forming in Delaware from day one is often cleaner than forming a Brazilian entity and flipping later.

When to form locally first

Form locally if you're genuinely building for the local market and local capital first. If you're a fintech in India raising from Indian angels and Indian micro-VCs like 100X or Blume in the first 18 months, a Pvt Ltd makes sense. You get DPIIT benefits, you avoid US tax filings, and you can always flip later when you're ready for US institutional rounds.

Form locally if you need local government programs to survive the first year. If you're relying on a ₹20 lakh grant from the Indian government or a tax incentive that requires local entity status, the flip cost later is worth it.

Form locally if you're not sure you'll need US capital. If you're bootstrapping or raising only from local angels and you might never do a venture round, there's no reason to deal with US tax filings.

But if you're 80% sure you'll raise from US VCs within 24 months, just form in Delaware. The flip cost and time cost aren't worth the optionality.

The hybrid path we see most often

For Indian founders specifically, the most common path is this: form a Pvt Ltd, raise a small angel round in India, operate for 12-18 months, then flip to Delaware when you're ready to raise a seed or Series A from US funds.

This works if the angel round is small (under ₹2 crore) and you have fewer than 10 investors. The flip is still $30-50K, but you've gotten 18 months of DPIIT benefits and avoided US tax filings in the early days when cash is tight.

If you take this path, plan the flip timing carefully. Don't start a US fundraise and then realize you need to flip. Flip first, then fundraise. Budget three months and ₹30 lakh minimum. Use a firm that's done multiple India flips—SAM, Cyril, Indus, or Khaitan if you're in India; Cooley or Gunderson if you're US-side.

And if you're doing this, get the Indian cap table right from day one. Use a standard SHA, not some custom document your local CA drafted. Make sure all your angels signed the same terms. A messy Indian cap table makes the flip twice as expensive.

Alternatives to Atlas

Atlas is the most common path, but it's not the only one. Firstbase.io does the same thing for $399. Doola does it for $297 but skews toward LLCs, which are wrong for venture-backed companies. Clerky does it for $399 and gives you more legal documentation upfront.

If you have $10K to spend, going direct to Cooley or Gunderson gets you more customization and a real lawyer reviewing your founder agreements. For most founders, that's overkill. Atlas works fine.

The bank account piece

Atlas bundles a Mercury account, which is what most startups use anyway. Mercury is fast, the UI is good, and they're used to working with international founders. Brex is an alternative but they mostly serve funded companies. Wise and Payoneer are cheaper for international transfers but less startup-friendly. Traditional banks like Chase or Bank of America are slow and require in-person visits.

Stick with Mercury unless you have a specific reason not to.

Tax filing reality

Once you're a Delaware C-Corp, you're filing US taxes. Federal return every year (Form 1120), Delaware franchise tax every year ($400-800 depending on how you calculate it), and state returns in any state where you have material operations. If you have foreign bank accounts over $10K, you're filing FBAR. If you're a non-US person with ownership, there are FATCA rules.

This is not optional. The IRS does not care that you're operating out of India and have zero US revenue. You're a US entity, you file US taxes.

Use a firm that works with international founders. Pilot, Inkle, or a local US CPA with international experience. Budget $3-5K/year minimum.

The actual decision tree

Here's the real question: are you building for the US market and planning to raise from US VCs within 24 months?

If yes, form in Delaware via Atlas. The $500 cost and $3-5K/year in tax compliance is cheaper than a $30-60K flip later.

If no—if you're building for your local market, raising from local angels, and might never need US capital—form locally. You can always flip later if things change.

If you're uncertain, form in Delaware. Optionality is expensive. The cost of being wrong about needing US capital (a $30K flip) is higher than the cost of being wrong about not needing it ($3K/year in tax filings you didn't need).

We've walked 40+ founders through this decision in the last two years. The founders who regret their choice are almost always the ones who formed locally and then needed to flip on a tight timeline. We've never had a founder regret forming in Delaware too early.

If you're starting a raise in the next 90 days and you're not sure which structure you have, book a call. We'll tell you if you need to fix it before you go out.