Company Formation and Legal Basics

You do not need a company to start building, and you do need one before you take money, hire, or sign anything. This chapter is the map of what the pieces are and which questions to take to a professional, not legal advice, and not a substitute for one.

This chapter describes categories and decisions. It is not legal advice. Entity choice, equity, and contracts have consequences that depend on your jurisdiction and your specifics. Use this to know what to ask; take the actual decisions to a lawyer or a reputable formation service.

The concept

"Starting a company" bundles several separate decisions that founders conflate:

 ENTITY        the legal container (LLC, C-corp, Ltd, sole trader…)
 OWNERSHIP     who owns what, and on what terms (equity, vesting)
 IP            who owns the code, the brand, the assets
 CONTRACTS     the agreements that govern all of the above
 BANKING       where the money lives, separate from you

Each can be done at a different time. The mistake is doing them all at once "to feel legitimate," or doing none of them until a problem forces it.

When you actually need to incorporate

You can build, validate, and even pre-sell as an individual in many jurisdictions. You genuinely need an entity when:

• You're taking outside money (investment, or in many cases app-store payouts at scale)
• You want a liability shield between the business and your personal assets
• You're bringing on a co-founder or hiring
• A customer or partner requires contracting with a company
• The tax treatment of a company becomes advantageous (Chapter 41)

Until one of those is true, incorporating early mostly buys you paperwork, fees, and an annual filing obligation. The common exception: if you're near-certain you'll raise venture money, the fundraising-standard structure (in the US, a Delaware C-corp) is worth setting up early because converting later is friction.

📐 Do the cheap, front-loadable legal work early and defer the expensive, trigger-gated work. A trademark search is free and should happen before you commit to a name; incorporation costs money and can wait for a real trigger. Founders reliably get this backwards, skipping the free early check and rushing the expensive late one.

Entity types, in brief

The right answer depends entirely on jurisdiction and intent. The shapes:

If you might raise VC: the market standard matters more than optimisation, investors expect a specific structure and will make you convert. Ask a startup lawyer before you form anything.

⭐ IP: the one that quietly ends companies

Intellectual property is where founders create existential risk without noticing:

• Own your code. If a contractor or agency wrote any of it, the contract must assign the IP to you in writing. "They built it, so I own it" is false by default in many jurisdictions, the creator often owns it absent an assignment.
• Own your brand. Trademark search before you commit to a name (Chapter 13); a rebrand after launch costs your SEO, reviews, and recognition.
• Build on rights-clean assets. Fonts licensed for commercial and embedded use; images you have permanent rights to. A borrowed aesthetic, someone else's characters, a franchise's look, is a takedown away from ending the product.
• Employees and IP. Employment contracts should assign work-product IP to the company. Co-founders especially, get the assignment in writing before there's anything valuable to fight over.

Co-founder terms, in writing, early

The most predictable company-killer is a co-founder dispute with no written agreement. Before you build anything valuable together:

• Equity split, and why. Equal is common and not automatic.
• Vesting, typically four years with a one-year cliff, so someone who leaves in month three doesn't keep a third of the company.
• Roles and decision rights.
• What happens if someone leaves, voluntarily or not.

Vesting protects the founders who stay. It is not a sign of distrust; it's the thing that makes staying worthwhile.

The paperwork that comes with money and hiring

Each capability drags legal obligations, exactly like store capabilities drag store requirements:

📐 Best practice

Validate as an individual until a real trigger requires a company.

If you might raise VC, form the standard structure early, ask a startup lawyer first.

Get IP assignment in writing from every contractor and co-founder.

Trademark-search the name before committing.

Build on rights-clean assets you can hold permanently.

Put co-founder terms in writing early, split, vesting, exit.

Use a reputable formation service or a lawyer, not a template you don't understand.

Keep business and personal finances separate from day one of having a business (Chapter 41).

Keep a signed, dated record of every agreement.

💀 Common mistakes

Incorporating too early "to feel legitimate", fees and filings for no benefit.

Incorporating wrong for your path, an LLC when you'll raise VC, forcing a costly conversion.

No IP assignment from a contractor. They may legally own your code.

No trademark search. A forced rebrand after launch.

Building on a borrowed aesthetic. One takedown from over.

No co-founder agreement. The most predictable company-killer.

No vesting. Someone leaves in month three with a third of the company.

Mixing personal and business money. Pierces your liability shield and wrecks your accounting.

Signing contracts you didn't read or understand.

Copy-pasting terms from another company without understanding what they commit you to.

The professional workflow

 1. STAY AN INDIVIDUAL until a real trigger (money, hire, liability, VC)

 2. IF VC IS LIKELY → talk to a startup lawyer BEFORE forming anything

 3. CHOOSE THE ENTITY for your path (bootstrap vs venture), in your jurisdiction

 4. FORM IT via a reputable service or lawyer

 5. IP
    trademark-search the name · assign contractor/co-founder IP in writing ·
    rights-clean fonts and assets

 6. CO-FOUNDERS (if any)
    equity split + reasons · vesting (4yr / 1yr cliff) · roles · exit terms

 7. BANKING — a business account, separate from personal

 8. AS YOU ADD CAPABILITIES → the matching agreements
    (ToS, privacy, employment, contractor, DPA)

 9. KEEP a signed, dated record of everything

Tools, websites & costs

Realistic formation cost: $100-600 plus jurisdiction fees. IP and legal review add more as needed. You can validate for $0 as an individual.

Alternatives & trade-offs

Incorporate early vs late. Early gives you a clean structure and a liability shield and costs fees and filings. Late saves money and risks a scramble when a trigger arrives. Late for bootstrappers; early for venture-track.

LLC vs C-corp (US). LLC is simpler and tax-friendly for a business you keep; C-corp is the venture standard and taxes differently. Choose by whether you'll raise institutional money.

DIY formation vs lawyer. A formation service handles the standard case cheaply. A lawyer is worth it for anything non-standard, multiple founders, foreign ownership, unusual equity, or an imminent raise.

Trademark now vs later. Searching is free and should happen before you commit. Filing costs money and can wait until the name has value worth protecting, but search first, always.

Home jurisdiction vs Delaware/other. Founders sometimes incorporate away from home for investor familiarity or tax; it adds a foreign-registration and compliance burden. Default to your home jurisdiction unless a specific reason (usually VC) says otherwise.

Checklist

• I am building as an individual until a real trigger requires a company
• If VC is likely, I've spoken to a startup lawyer before forming
• The entity type fits my path (bootstrap vs venture) and jurisdiction
• Every contractor/co-founder has assigned IP in writing
• The name is trademark-searched
• Assets (fonts, images) are rights-clean and permanent
• Co-founder terms, split, vesting, exit, are in writing
• Business and personal finances are separate
• Each capability's matching agreement is in place
• I have a signed, dated record of every agreement
• I understood everything I signed

📓 Case Study: one real IP decision, and an entity that never formed

Project: SOLIS, built solo. This is a thin case study on purpose, a solo, pre-launch, pre-revenue project barely touches company formation, and pretending otherwise would be dishonest. Two things it did exercise are worth pulling out.

A genuine IP decision, made for the right reason. The product's visual identity was built entirely on public-domain artwork, museum open-access paintings with permanently, unambiguously waived rights, plus original generated assets. The reasoning was recorded in the product plan as a competitive advantage:

An original, IP-safe identity, everyone else in the category is anime skins; this is oil paintings. That's a launch-blocker the imitators all fail.

The insight generalises: competitors built on a borrowed aesthetic (a franchise's characters, an anime's look) carry an existential risk, a rights-holder appearing ends the app. Building on public-domain assets removes that risk permanently. Rights permanence is a business decision, not a design detail (Chapter 13). This is the one place a solo pre-launch project genuinely intersects company-formation concerns, and it got it right.

⚠️ What was not done, and why that's instructive rather than negligent. No legal entity was formed. No trademark search was run on the name, despite the app being about to go to a global store under a common word (Chapter 13). The code lived in a private repository under an individual account.

For a pre-launch, pre-revenue, solo project, not forming an entity is a defensible default, this chapter's own advice is to stay an individual until a trigger arrives, and none had. The developer-account enrolment and the Paid Apps agreement (which needs bank and tax details, Chapter 41) were the first real brush with the machinery, and they arrived exactly when the model says they should: at the point of taking money.

The trademark gap is the real miss. Unlike incorporation, a trademark search is free and should happen before committing to a name, and it didn't. For a product about to publish globally under a common word, that's an open, avoidable risk carried into launch.

The honest summary: the project correctly deferred the expensive, trigger-gated legal work (incorporation) and skipped the cheap, front-loadable legal work (a trademark search), which is precisely the wrong way round, and a common one. Founders defer the free early checks and rush the expensive late ones.

Lessons

1. You don't need a company to start. You need one at a real trigger, money, hire, liability, or a raise.
2. Do the cheap, front-loadable legal work early, trademark search, and defer the expensive, trigger-gated work.
3. Get IP assignment in writing from every contractor and co-founder. Default ownership is often not yours.
4. Rights permanence is a business decision. A borrowed aesthetic is a takedown away from over.
5. If you might raise VC, form the standard structure early and ask a lawyer first, conversion is friction.
6. Vesting protects the founders who stay. Put co-founder terms in writing before there's value to fight over.
7. Separate business and personal money from the moment there's a business.
8. Don't sign what you don't understand.
9. Match agreements to capabilities, selling, hiring, contracting each drag obligations.
10. This is a map, not advice. Take the real decisions to a professional.

Next: Chapter 40: Privacy, Terms and Compliance →