How to Set Up Your Startup Legal Entity Structure the Right Way

Two founders walk into a lawyer's office with the same idea, the same product, roughly the same bank balance. One walks out with a Delaware C-Corp and a cap table that will survive three funding rounds. The other walks out with an LLC because it was cheaper and simpler, and spends the next eighteen months explaining to investors why they should invest through a holding company. I have been in the second room. It cost me four months and a few thousand dollars in legal work to fix.

So when people ask me how to set up a startup legal entity structure, I don't start with the paperwork. I start with the question nobody asks first: who is going to give you money, and in what form? Everything else — the state, the filings, the tax elections — follows from that answer.

The entity decision is really a funding decision

Here's the thing nobody tells you before you incorporate: the choice between an LLC and a C-Corp is not a tax question at the start. It's an investor question.

Venture capital funds have limited partners, and those LPs have tax profiles. A US venture fund generally cannot hold an LLC interest without creating unrelated business taxable income headaches for its LPs (that's the UBTI problem, and it's the reason fund docs often simply prohibit it). So when a VC says "we only invest in C-Corps," they're not being difficult. Their lawyers wrote that clause years before you showed up.

A decision tree that actually reflects reality

If I had to compress this into something you can act on this week:

• You intend to raise institutional venture capital in the next 24 months — Delaware C-Corp. Don't overthink it.
• You're bootstrapping a services or content business and want simple pass-through taxation — LLC, and consider an S-Corp election once profits are steady (this changes your self-employment tax picture, so run the numbers with an accountant first).
• You want to raise from angels only, in small rounds — LLC can work, but be honest with yourself about whether "angels only" will still be true in two years.
• You're building outside the US and selling to US customers — you can incorporate in your home country and add a US subsidiary later. Plenty of founders do this and it is not the disaster people warn about.
• You have co-founders spread across three countries — get a lawyer, not a form. This is the case where the cheap option becomes expensive.

The catch? Every one of these paths has an edge case. A founder with a profitable side business wants an LLC for the side business and a C-Corp for the startup. That's fine — just keep the assets genuinely separate, because commingling them is how people lose the liability shield they paid for.

Why Delaware, when you don't live there

Delaware's court of chancery has decades of case law on corporate disputes, and that body of precedent is what makes investors comfortable. You are not incorporating in Delaware because of taxes. You're incorporating there because if your board ever has a fight, everyone knows roughly how it will be resolved.

The annual cost is modest but not zero: a franchise tax that for most small startups lands around a few hundred dollars a year, plus a registered agent fee (typically somewhere in the low hundreds). Compare that with legal fees for a clean incorporation package, which in my experience runs from roughly $1,500 for a standard startup package to well over $10,000 if things get complicated. Those numbers are not from a study. They're what I've paid and what founders have quoted me.

The actual steps, in the order that matters

Most checklists put "choose a name" first. That's wrong, or at least premature — you can change a name later more easily than you can change a structure.

Step one: agree on the equity split and vesting, in writing

I'll say this bluntly: I have seen more startups die from an unresolved founder equity conversation than from any legal filing mistake. Before you file anything, you and your co-founders need a document — even a plain one — that says who owns what, over what schedule, and what happens if someone leaves.

Standard founder vesting is four years with a one-year cliff. Meaning: you get nothing until twelve months in, then 25%, then monthly for the rest. It feels insulting when you're 26 and certain you'll be here forever. It is also the single most investor-friendly thing you can do before your first meeting.

Step two: file the certificate of incorporation

You file with the Delaware Division of Corporations, either directly or through a service. The document itself is short — authorized shares, par value, registered agent, incorporator. The default in most templates is 10 million authorized shares, which is a reasonable starting point and leaves room for a couple of rounds.

The part founders underestimate: you must also adopt bylaws, hold an organizational meeting, issue stock to founders, and get board and stockholder consents on paper. Skipping the consents is common and looks sloppy in diligence two years later.

Step three: file the 83(b) election within 30 days

This is the one that actually destroys value when you get it wrong. When you receive restricted stock subject to vesting, the IRS treats it as taxable as it vests — unless you file an 83(b) election within 30 days of the grant. File it and you're taxed on the value at grant, which for founder stock at formation is usually close to nothing.

Miss the 30-day window and there is no appeal, no extension, no fix. I know a founder who missed it by two days and spent years paying ordinary income tax on stock that had appreciated substantially. Thirty days. Put it in your calendar the moment the shares are issued.

Step four: EIN, bank account, foreign qualification

Get your EIN from the IRS (free, online, takes minutes), open a business bank account in the company's name, and — this catches people — register to do business in the state where you actually operate. Delaware incorporation doesn't give you a free pass to operate in California or Berlin. That's called foreign qualification, and it's a separate filing with its own fee.

The cap table is the structure, not an afterthought

People think of "legal entity structure" as the incorporation filing. It's not. The structure is the cap table and the agreements around it. A beautifully filed C-Corp with a chaotic cap table is worse off than an LLC with clean books.

SAFE notes versus a priced round

Most first-time founders raise on SAFEs or convertible notes rather than a priced round. That's fine and it's cheap — a few pages, no valuation negotiation. The trap is that SAFEs stack up and their conversion mechanics interact. Two SAFEs with different caps and a discount rate can produce a cap table that surprises everyone at conversion, including the people who signed them.

My rule: once you have more than three SAFEs outstanding, pay someone to model the conversion before you sign the fourth. The modeling costs less than the surprise.

Section 1202 and QSBS — worth knowing early

Qualified Small Business Stock can, under the right conditions, let founders and early shareholders exclude a substantial chunk of capital gains at exit. The conditions are specific — the company must be a domestic C-Corp, meet an asset threshold at issuance, and the shares must be held long enough. Which means, and this is the practical point, an LLC converted to a C-Corp late may not give founders the same clock start.

I'm not going to give you a number here, because the rules have shifted over the years and I'd rather you ask a tax professional than trust a blog. The takeaway is simpler: if a large exit is plausible, the C-Corp timing matters for tax reasons you won't appreciate in year one.

Where to file: DIY, Stripe Atlas, or a lawyer

You have three realistic routes, and they differ more in what happens after the filing than in the filing itself.

Route	What you get	Typical upfront cost	Best for
DIY (state portal + free templates)	Filed certificate, whatever you assemble yourself	State fee + agent fee only	Founders with prior experience or a lawyer in the family
Formation platform (Stripe Atlas and similar)	Delaware C-Corp filing, EIN, template docs, partner perks	Roughly $500 one-time plus annual agent costs	Solo founders or a pair, first time, no complex cap table
Startup lawyer (fixed-fee package)	Filing plus tailored bylaws, IP assignment, founder agreements, SAFE review	Low four figures and up	Two or more founders, any IP complexity, imminent fundraising

Is Stripe Atlas the right call for you?

Atlas is a service that incorporates a Delaware C-Corp for you and bundles in an EIN, a set of standard documents, and access to various partner offers. I used it once, for a project I was starting alone. It took an afternoon and the documents came back clean.

What it does not do: review your founder agreement, handle a three-way equity split with vesting nuances, or advise you on IP assignment from a previous employer. It's a filing service with a good user experience, not a law firm. If your situation is a solo founder with a clean idea, that's fine. If your situation is anything else, read the documents it generates carefully — or have someone else read them.

There are also free-and-cheap startup programs from large cloud providers and platforms that bundle credits and some legal perks. They're worth joining. Just read what the perks actually cover before you count them as your legal budget.

The mistakes I see repeatedly, and the migration question

Three patterns keep showing up in my conversations with founders, and I've made two of them myself.

• Issuing shares to a co-founder who leaves in month three with a fully vested 30% stake. No vesting, no buyback. Unfixable without a difficult negotiation.
• Signing a co-founder agreement before assigning IP properly — the code written before incorporation belongs to a person, not the company, until it's assigned.
• Electing S-Corp status too early, before the payroll and reasonable-salary requirements make sense, and paying an accountant more than the tax saved.

And the migration question: can you convert an LLC to a C-Corp later? Yes. It's usually done as a statutory conversion or a merger, it costs legal fees, and depending on the state it can trigger tax on the appreciation at the moment of conversion. It works, it's common, and it's a detour you'd rather not take. Which is why the honest answer to "should I start with an LLC and convert later?" is: only if you're genuinely unsure whether you'll raise venture money. If you're sure, start where you'll end up.

The structure you choose on a Tuesday afternoon, with a form and a fee, decides who can invest in you for years. That's an uncomfortable amount of weight for a filing. But it's the actual weight, and it's why the founders who take two extra weeks on this at the start are usually the ones who don't need to redo it at the worst possible moment.