Building a startup means turning an idea into a functioning business that can grow, and it follows a roughly predictable sequence: validate the idea, set up the legal entity, build a minimum viable product, secure funding, and find a repeatable way to acquire customers. Each stage has specific costs, timelines, and decisions that shape everything that comes after. Here’s how to move through them.
Validate the Idea Before You Build Anything
Most failed startups don’t fail because of bad execution. They fail because they built something nobody wanted. Validation is the process of proving demand exists before you spend months and thousands of dollars on a product. You can do this in a few weeks if you’re disciplined about it.
Start by sizing the market. You need to estimate your total addressable market (TAM), which is the total annual revenue available if you captured every possible customer in your space. Then narrow that to your serviceable available market (SAM), the slice you could realistically reach given your product’s focus and geography. Finally, calculate your serviceable obtainable market (SOM) based on your actual resources, pricing, and expected conversion rates. If your SOM is too small to support a viable business, the idea needs to change before anything else happens.
You can approach market sizing from the top down, starting with industry reports and narrowing to your niche, or from the bottom up, starting with individual customers you’ve identified and extrapolating. Bottom-up estimates tend to be more honest because they force you to name real people or companies who would pay.
Beyond the numbers, talk to potential customers directly. Run 20 to 30 interviews asking about the problem you want to solve, not your proposed solution. If people don’t recognize the problem or already have a good-enough workaround, that’s a signal. You can also test demand by putting up a landing page describing your product and measuring how many visitors sign up for a waitlist or click a “buy” button. These signals won’t guarantee success, but they’ll prevent you from spending six months building something based on assumptions.
Choose and Set Up Your Legal Structure
If you plan to raise venture capital, most investors expect a C corporation. This structure allows you to issue multiple classes of stock (common shares for founders, preferred shares for investors) and is the standard for startups planning to scale and eventually exit through acquisition or IPO. If you’re building a smaller business you intend to own long-term, an LLC offers simpler taxes and more flexibility.
Many founders incorporate as a C corporation in Delaware regardless of where they operate, because Delaware’s corporate law is well-established and familiar to investors. The state filing fee for a corporation with up to 1,500 shares of no-par-value stock is $109. You’ll also owe an annual franchise tax, which starts at $175 for corporations authorized to issue up to 5,000 shares and can reach $200,000 for companies with millions of authorized shares. Since you won’t operate out of Delaware, you’ll need a registered agent there to receive legal documents on your behalf, which is an additional recurring cost that varies by provider.
Beyond incorporation, you’ll want to file an 83(b) election with the IRS within 30 days of receiving your founder shares if they’re subject to vesting. This lets you pay taxes on the shares at their current (low) value rather than their potentially much higher value when they vest. You’ll also need a founders’ agreement spelling out each founder’s equity stake, vesting schedule (typically four years with a one-year cliff), roles, and what happens if someone leaves.
Build a Minimum Viable Product
An MVP is the simplest version of your product that lets real users experience the core value. It’s not a prototype or a mockup. It’s a working product, just stripped down to the one or two features that matter most. The goal is to get it into people’s hands quickly so you can learn what works and what doesn’t.
A typical MVP takes about three to four months to develop and release. But your timeline and cost depend heavily on how you build it. No-code platforms like Bubble, Webflow, or Glide let you assemble functional web apps using visual builders and pre-made templates. They’re cheaper and faster because you’re working with existing components rather than writing code from scratch. For many B2B SaaS products, marketplaces, or simple consumer apps, a no-code MVP is perfectly adequate for testing demand.
Custom development gives you far more flexibility and is necessary when your product requires complex logic, real-time processing, or integrations that no-code tools can’t handle. But it takes longer and costs more because every feature is programmed, tested, and deployed from the ground up. If you’re a non-technical founder hiring developers, expect custom MVP costs to run anywhere from $15,000 to $75,000 or more depending on complexity.
Whichever route you choose, resist the urge to add features before you’ve gotten feedback on the core ones. The MVP exists to generate learning, not to impress people with polish.
Fund the Business
Most startups begin with some combination of personal savings, friends-and-family money, or credit. This phase, often called bootstrapping, keeps you in full control but limits how fast you can move. Once you have a working MVP and early traction (users, revenue, or strong engagement metrics), you’re in a position to raise outside capital if the business model calls for it.
Startup fundraising typically follows a progression. Pre-seed rounds are often raised from angel investors or small funds, usually on a SAFE note (a simple agreement for future equity) rather than a priced round. This means you’re not setting a formal valuation yet. Instead, investors give you money now in exchange for equity that converts at a discount when you raise a larger round later.
Seed rounds are larger and often the first priced round. Founders selling equity at the seed stage give up about 19.5% of the company on average. That means if you and a co-founder split 100% of the equity before the round, you’d collectively own roughly 80.5% afterward. Each subsequent round (Series A, B, C) dilutes your ownership further, so understanding how much you’re giving up at each stage matters.
To raise a seed round, you’ll typically need a pitch deck covering the problem, your solution, market size, traction, team, and how you’ll use the funds. Investors at this stage are betting on the team and the market opportunity as much as the product itself. The process usually takes two to four months of active fundraising, including meetings, follow-ups, due diligence, and legal paperwork.
Find Your Go-to-Market Strategy
A go-to-market (GTM) strategy is how you acquire and retain customers. The right approach depends on who you’re selling to and how much your product costs.
If you’re selling to individual consumers or small businesses at a low price point, a self-serve model often works best. This is product-led growth: customers find you through search, content, or word of mouth, sign up for a free or freemium version, and upgrade to a paid plan when they hit the limits of the free tier. Your product does the selling. This model requires investment in the onboarding experience and in-product prompts rather than a sales team.
If you’re selling to mid-market companies, inside sales is the typical approach. You’ll need sales development representatives (SDRs) who reach out to prospects, book demos, and guide buyers through a structured sales process. The sales cycle might be a few weeks to a couple of months.
For enterprise deals with six-figure contract values, direct sales with dedicated account executives is standard. These deals have long sales cycles, often six months or more, and involve multiple stakeholders on the buyer’s side. You’ll need case studies, security documentation, and the ability to customize your offering.
Many startups begin with one model and layer on others as they grow. A company might start self-serve to build a user base, then add an inside sales team to convert larger accounts. The key early decision is which motion to invest in first, because each one requires different skills, tools, and spending.
Hire Carefully and Slowly
Early hires define your startup’s culture and capability more than any other decision. The first five to ten people you bring on will set the tone for how the company communicates, solves problems, and handles pressure. Hire for versatility over specialization in the early stages. You need people who can wear multiple hats, not someone who only does one narrow function.
Equity compensation is a major tool for early-stage startups that can’t compete on salary. A common structure is to offer below-market cash compensation paired with stock options that vest over four years. Be transparent about what those options are worth today (likely very little) and what they could be worth if the company succeeds. Early employees typically receive between 0.25% and 2% equity depending on their role and seniority, though this varies widely.
Set Milestones That Matter
Startups that survive tend to operate in focused sprints toward clear milestones rather than vaguely “working on the business.” At each stage, define what success looks like and build backward from it.
Before fundraising, your milestone might be 100 active users or $5,000 in monthly recurring revenue. Before hiring your first salesperson, it might be closing 10 deals yourself to prove the sales process works. Before expanding to a new market, it might be reaching a specific retention rate that shows your current customers stick around.
Track your visibility and traction with measurable indicators: web traffic, social media engagement, conversion rates from free to paid, customer acquisition cost, and monthly revenue growth. These numbers tell you whether your strategy is working and give you concrete evidence to share with investors, advisors, and future hires. The startups that build momentum are the ones that know their numbers cold and adjust their approach based on what the data actually says.