Country × State Decision Tree: Pick Your US LLC State

Answer capsule

The decision tree asks four questions: country of residence, business type, revenue stage, and primary goal. It outputs a recommended US state (Wyoming in roughly 90% of non-resident cases, occasionally Delaware or Florida), a banking strategy (Mercury, Relay, or Wise as primary), and country-specific tax notes (treaty access and Form 5472 obligations). It is free, requires no signup, and takes about 60 seconds.

Open the decision tree → · Form your Wyoming LLC for $397 (state fee included) →

Why a decision tree beats a "best state" listicle

Search for "best state to form an LLC" and you get a wall of articles - LLC University, doola, Northwest Registered Agent, Firstbase - most of which hedge with "it depends." They are not wrong. The problem is that it depends is useless to a founder in Lahore or São Paulo trying to make one decision today. A decision tree forces the dependencies into the open: it asks you the questions that actually move the answer, then commits to a pick. That is the difference between content that ranks and content that decides.

The four variables we ask about are the same four that a US CPA would walk through on a paid call. We are not hiding the logic; the rest of this page is the logic.

The four questions, and what each one branches on

Question 1 - Country of residence

This is the highest-leverage input. Your country determines three things at once:

• Tax-treaty access. The US has income tax treaties with roughly 60 countries. A treaty does not change where you form (federal tax is identical in all 50 states), but it changes the withholding rate on any US-source FDAP income (dividends, royalties, certain interest) your LLC receives. India residents drop dividend withholding from the default 30% to 25% (and to 15% only where a corporate shareholder holds at least 10%) by filing Form W-8BEN-E (India–US treaty, Article 10). Brazil residents get no reduction - Brazil has no ratified US income tax treaty, so the default 30% stands.
• Bank approval odds. Mercury, Relay, and Wise each maintain their own country risk tiers. Mercury clears UK and EU founders most easily, keeps India/Pakistan/Bangladesh/Brazil/UAE founders approvable with clean paperwork, and restricts or rejects sanctioned countries - though approval is never guaranteed. The tree routes high-friction countries straight to Wise Business, which accepts almost every country.
• Local anti-deferral rules. Some countries (Brazil's CFC regime under Lei 12.973/2014, for example) tax foreign-company profits at the owner level. This does not change the state pick, but it changes the tax note you receive.

The dropdown covers 60+ countries and branches by treaty status, country risk profile, and banking acceptance tier.

Question 2 - Business type

We ask: SaaS, agency/services, e-commerce, content creator, freelance, crypto, real estate, or other. Business type branches on two things - regulatory exposure and banking compatibility:

• SaaS, agency, services, freelance, content are the cleanest. No inventory, no licensing, high bank-approval odds. Wyoming is the default.
• E-commerce / Amazon FBA adds sales-tax nexus considerations and marketplace W-8/W-9 handling, but the formation state stays Wyoming for most non-residents.
• Crypto narrows your banking options - Mercury and many neobanks restrict crypto-native businesses - so the tree flags Wise or a crypto-friendly alternative.
• Real estate is the one type that can flip the recommendation. If you are buying US rental property, you often want the LLC in the state where the property sits (or a Wyoming holding LLC over state-level subsidiaries). The tree surfaces that branch instead of a blanket "Wyoming."

Question 3 - Revenue stage

Pre-revenue, $0–$50K, $50K–$250K, or $250K+. Revenue stage branches on banking qualification and tax-planning complexity. Brex, for instance, is built for funded or higher-revenue startups and is a poor fit pre-revenue; Mercury and Relay are stage-agnostic. At $250K+ the tree adds a note recommending a US CPA review, because effectively-connected-income (ECI) analysis and multi-payer withholding get genuinely complex at scale.

Question 4 - Primary goal

Lowest cost, maximum privacy, fundraising-readiness, or asset protection. This is the final branch to a state:

• Lowest cost / privacy / asset protection → Wyoming. Wyoming carries no state income tax, no franchise tax, a $60 minimum annual report, strong charging-order protection, and no public member listing.
• Fundraising-readiness (priced equity imminent) → Delaware. If you are about to raise a priced round from US VCs, they will expect a Delaware C-corp, not an LLC. The tree says so plainly rather than selling you a Wyoming LLC you will have to unwind.
• US real estate / Florida residency angle → Florida in specific cases.

The state shortlist, with real numbers

For 90%+ of non-resident founders the answer is Wyoming. Here is the cost and feature comparison the tree draws on. Year-2+ recurring cost is where states separate.

Sources: Wyoming annual license tax is the greater of $60 or two-tenths of one mill on Wyoming assets, due your anniversary month, per the Wyoming Secretary of State / Business Division. Delaware's flat $300 LLC franchise tax is due June 1 every year, with a $200 penalty plus 1.5%/month interest for late payment, per the Delaware Division of Corporations. BOI exemption for US-formed domestic entities follows FinCEN's March 26, 2025 Interim Final Rule.

The takeaway the table makes obvious: over five years, Wyoming's recurring cost (~$60/year) beats Delaware's ($300/year) and Nevada's ($350+/year). New Mexico is technically cheaper than Wyoming at $0/year recurring, and the tree will tell you so if your only goal is cost minimization - but New Mexico's asset-protection caselaw is weaker and it is less established for non-residents, which is why the default still points to Wyoming.

Why state choice does NOT change your federal tax

A point the decision tree hammers, because so many founders get it wrong: your US federal tax is identical in all 50 states. Forming in Wyoming versus Delaware versus Florida changes your state-level obligations and your annual fees - it does not change a dollar of federal tax. A foreign-owned single-member LLC is a disregarded entity for federal income tax in every state. The state pick is a cost-and-privacy decision, not a federal-tax decision. Anyone telling you Wyoming "saves federal tax" is wrong.

The non-resident layer the listicles skip: banking, privacy, Form 5472

State choice is only one of three decisions the tree bundles. The other two matter just as much.

Banking - Mercury, Relay, Wise

You cannot run a US LLC without a USD bank account, and approval is country-dependent, not state-dependent. The tree outputs a banking primary and a fallback:

• Mercury is the default primary for most founders - no minimum balance, no monthly fee, fully remote. It accepts a meaningful share of applicants, clearing UK/EU founders most easily and remaining approvable with clean paperwork for India, Pakistan, Bangladesh, Brazil, UAE, Egypt, Philippines, Vietnam, and Indonesia (approval not guaranteed).
• Relay is the standard fallback when Mercury declines.
• Wise Business is the universal backstop - it accepts almost every country, so the tree routes higher-friction nationalities here first.
• Brex appears only for the $250K+ / funded branch.

All four open remotely with your Articles of Organization, EIN (IRS CP575 letter), passport, and a clear business description. No US visit, no US address.

Privacy

Wyoming does not list member or manager names on public filings. For founders who do not want their name in a searchable public registry tied to a US company, this is a real differentiator - and it is the reason the privacy branch resolves to Wyoming, not Delaware (which is also private but costs $300/year) and not a public-disclosure state.

Form 5472 - the federal filing that is mandatory regardless of state

Every foreign-owned single-member US LLC must file Form 5472 with a pro forma Form 1120 every year in which it has a reportable transaction with a related party - and an initial capital contribution counts. This is true in Wyoming, Delaware, Florida, New Mexico, every state. The penalty for failing to file, or filing substantially incomplete, is $25,000 per form under IRC §6038A, with an additional $25,000 for each 30-day period the failure continues after IRS notice - and there is no statutory maximum cap, per the IRS Form 5472 instructions and the IRS international information reporting penalties page. The 2026 deadline for 2025-tax-year, calendar-year filers is April 15, 2026. The decision tree surfaces this on every output node, because no state choice exempts you from it.

Worked examples - three real branches

These are the kind of outputs the tree produces. Each is a full recommendation, not just a state.

India + SaaS + $50K revenue + lowest cost. → Wyoming LLC via WyomingLLC ($397, state fee included). Mercury as banking primary (Indian founders are approvable with clean paperwork; approval not guaranteed), Relay fallback. Tax note: India–US treaty is active; Article 7 keeps operating profits out of US tax with no US permanent establishment; file W-8BEN-E with each US payer to drop dividend withholding to 15%. Form 5472 + 1120 due annually.

UK + bootstrapped SaaS + fundraising goal + pre-revenue. → Two-path output. If a priced equity round is not imminent: Wyoming LLC now, convert to a Delaware C-corp when you actually raise. If a priced round is imminent: go Delaware C-corp now (e.g., Stripe Atlas) and skip the conversion friction. Mercury primary (UK founders clear most easily; approval not guaranteed). Tax note: UK–US treaty active; no US tax on non-ECI operating profit.

Brazil + agency + $250K revenue + lowest cost. → Wyoming LLC. Banking primary: Wise (Brazil has no US treaty, so there is no FDAP-withholding advantage to optimizing the bank, and Wise is the cleanest for Brazilian-resident USD operations); Mercury still works (approval not guaranteed) if preferred. Tax note: no Brazil–US treaty means default 30% on US-source FDAP - so avoid routing US dividend income through the LLC; operating agency revenue stays at 0% US tax. Flag Brazilian CFC rules (Lei 12.973/2014) for a local CPA. Form 5472 + 1120 mandatory; $250K+ triggers a US CPA review note.

Pakistan + freelance + $0–$50K revenue + lowest cost. → Wyoming LLC ($397, state fee included). Mercury as primary (Pakistani founders can be approvable with a clean, specific business description, but Pakistan is a tightened-review tier and approval is not guaranteed), Relay then Wise as fallbacks. Tax note: the US–Pakistan treaty is old but in force; ordinary freelance services income is already outside US tax with no US permanent establishment, so the treaty matters mainly for any US-source passive income, claimed via W-8BEN-E. Form 5472 + pro forma 1120 due annually even at low revenue, because the founder's own capital contribution is a reportable transaction.

UAE + e-commerce (Amazon FBA) + $50K–$250K + asset protection. → Wyoming LLC for the strong charging-order protection and privacy. Banking: Mercury primary (UAE approval is strong), Wise as the multi-currency backstop for cross-border supplier payments. Tax note: the UAE levies 0% personal income tax, so the founder's main exposure is the UAE's own corporate-tax rules on business carried on in the UAE, not US federal income tax on non-ECI; an FBA seller should also map US sales-tax nexus by state, which is a separate question from the formation state. Form 5472 + 1120 mandatory.

The pattern across all five examples is deliberate: the state answer is Wyoming almost every time, and the variable that actually changes the recommendation is rarely the state - it is the banking route and the country tax note. That is the whole reason the tree leads with country.

Why country is the highest-leverage variable

It is worth dwelling on why the first question carries the most weight, because it is counterintuitive: founders assume the state is the big decision, when in practice the country is. The state pick swings your recurring cost by a couple of hundred dollars a year and your privacy posture - meaningful, but bounded. Your country of residence, by contrast, swings three things that can each dwarf the state decision: whether a tax treaty lets you reclaim withholding on US-source passive income (potentially thousands of dollars a year), which bank will actually approve you (the difference between an operational LLC and a dormant one), and whether your home country's anti-deferral rules tax the LLC's profits at your level regardless of the US treatment. A founder who optimizes the state to save $240/year while ignoring a 30% withholding leak or a failed bank application has optimized the small lever and missed the large ones. The tree front-loads country precisely so the high-leverage variable is settled before the low-leverage one.

How to use the tree yourself - step by step

1. Open wyomingllc.xyz/tools/decision-tree (no signup).
2. Select your country from the dropdown.
3. Pick your business type (SaaS, agency, e-commerce, content, freelance, crypto, real estate, other).
4. Pick your revenue stage (pre-revenue, $0–$50K, $50K–$250K, $250K+).
5. Pick your primary goal (lowest cost, privacy, fundraising, asset protection).
6. Read the output: recommended state, banking primary + fallback, and country tax note.
7. If the output flags "consult a US CPA" (complex / $250K+ / real estate / fundraising), take that seriously - those are the ~10% of cases where a generic answer is not enough.

When the tree's answer is wrong

We are honest that a four-question tree cannot cover everything. It will be wrong, or incomplete, for:

• Multi-LLC and international holding structures (e.g., a Wyoming holding LLC over state-level operating subsidiaries).
• US real estate beyond a single rental, where state-of-property and 1031/FIRPTA planning dominate.
• M&A or exit planning, where buyer preferences and QSBS (a C-corp feature, not an LLC one) override the cost logic.
• Crypto businesses with token issuance, which raise securities questions the tree does not touch.

For these, the tree explicitly recommends a US CPA or attorney consultation. The generic recommendation is reliable for roughly 90% of non-resident founders; the remaining 10% deserve paid, tailored advice.

The honest framing we hold to is that a tool like this is a starting point, not a verdict. It narrows four interacting variables into a concrete, defensible default and tells you why that default holds - which is far more useful than a listicle that hedges with "it depends." But it is built on the typical case, and a typical-case tool cannot know that you are mid-acquisition, that your home country just changed its CFC rules, or that a specific investor has a specific entity requirement. When the tree flags one of those edge branches, the right response is to treat its output as a well-reasoned first hypothesis to bring to a professional, not as a final answer to act on alone. For the large majority whose situation is genuinely typical, the tree's answer and a professional's answer converge - which is exactly why publishing the logic openly, rather than gating it behind a paid call, is the honest thing to do.

Bottom line

The country × state decision tree turns the most-asked, most-hedged question in non-resident company formation into a single concrete answer in 60 seconds: a state, a banking plan, and a tax note specific to your country. For the large majority, that answer is a Wyoming LLC ($397, state fee included), a Mercury or Wise account, and an annual Form 5472 filing. For the edge cases, the tree is honest enough to send you to a professional. Run it free at wyomingllc.xyz/tools/decision-tree.

Open the decision tree → · Start your $397 Wyoming LLC →