Startup Stages: The 5 Stages and the Test That Ends Each One

Most lists of startup stages describe what a company looks like at each point: seed, early, growth, expansion, exit. That is useful if you are an investor. It is not much use if you are the one building it, because it never tells you when you are allowed to move on.

This guide gives you the five startup stages in the order you go through them, the test that ends each one, a worked example that follows one idea through all five, how they differ from funding stages, and why some lists have six or seven.

What Are the Stages of a Startup?

Steve Blank, who developed the customer development method, defined a startup as an organisation formed to search for a repeatable and scalable business model. The stages are the steps of that search. Each one removes a different reason the business could fail.

Ash Maurya, author of Running Lean, describes the middle of this path the same way: every product goes through problem/solution fit, product/market fit, and scale. The idea stage before it and the MVP stage inside it are where a business of one makes most of its early decisions, so they get their own place here.

1. The Idea Stage

The idea stage is where most people think they are further along than they are. An idea becomes testable when you can say who has the problem, in what situation, and what it costs them.

The test that ends it: one sentence about a specific person. "Small businesses struggle with admin" fails. "Solo bookkeepers lose about a day a month chasing clients for missing receipts" passes, because you can now go and find those people and ask them.

What usually goes wrong: starting from a product you want to build and looking for a problem to attach it to. Our comparison of problem-first and product-first covers why that order matters, and 10 questions every startup must answer before building is a useful check before you leave this stage.

2. Problem-Solution Fit

This stage asks whether the problem is painful enough that people will do something about it, and whether your proposed solution would change what they do. It is the cheapest stage to fail, because all it costs is conversations.

The test that ends it: a commitment that costs the person something, such as a pre-order, a deposit or a paid pilot, from a share of people you decided on before you started. Interest alone does not count. A waitlist full of people who will not pay is a polite no.

How to run it: talk to 10 to 15 people who have the problem, ask about the last time it happened rather than what they would do, and look for what they already spend to work around it. The full method is in how to interview customers, and how to test problem-solution fit has six steps and three worked examples. If you need to know whether enough people have the problem to support a business, how to calculate market size belongs here too.

3. Minimum Viable Product

Now you build, but only the smallest version that lets a real customer do the real job. The point of an MVP is to learn whether your solution works, not to launch a finished product.

The test that ends it: someone who is not a friend uses it for the actual task, without you holding their hand, and pays for it. Usage you have to prompt is not usage.

Eric Ries describes this as a Build-Measure-Learn feedback loop: build the smallest thing, measure what people do with it, learn, and change it. Our guide to what an MVP is covers the types, from a manual service behind a simple page to a working prototype, and a lean canvas keeps the whole plan on one page while you iterate.

4. Product-Market Fit

Marc Andreessen, who popularised the term, defined product-market fit as being in a good market with a product that can satisfy that market. In practice, it is the point where the product starts pulling customers instead of you pushing them.

The test that ends it: customers keep using it without being chased, usage holds up after the first month, and new customers arrive because existing ones told them. If you still have to push every customer back, you are not there yet.

Pricing usually gets settled in this stage, because you finally have real customers to test it on. What product-market fit is covers the signs in detail, and how much should I charge covers setting a price you can defend.

5. Growth

Growth is not "more marketing". It is finding a way to get customers that you can repeat, at a cost the business can carry. Before product-market fit, spending on growth mostly buys customers who leave.

The test that ends it: one channel, whether search, referrals, partnerships or outbound, brings in customers again and again, and what it costs to win each one is clearly less than what they are worth to you. Customer acquisition cost is the number that tells you.

For a business without investors, this is also where money decisions get real. Bootstrapping your startup covers growing from revenue, and how much money you need to start a business helps you plan the cash this stage takes.

A Worked Example Through All Five Startup Stages

One illustrative idea, followed from start to finish, so you can see what passing each test looks like in practice.

Notice what each stage cost. Stage 2 took a few weeks of calls. Had the idea failed there, that is all it would have cost. Had the founder skipped straight to building, the same failure would have arrived months later, after the product was built.

Startup Stages vs Funding Stages

Many results for this topic mix two different things. Development stages describe what you have proven. Funding stages describe how a company raises money from investors.

The two line up loosely, because investors fund evidence: a pre-seed round often backs problem-solution fit, and a Series A usually expects product-market fit. But a business you build alone, from revenue, goes through every development stage without ever touching a funding stage.

Why Some Lists Have Six or Seven Startup Stages

You will find lists with four, five, six or seven stages. They are not contradicting each other. They are cutting the same path at different points or adding stages that come after the one you need.

• Lists that add a "pre-startup" stage split the idea stage into thinking about it and committing to it.
• Lists that add "expansion", "maturity" or "exit" continue past growth into what happens to a large company. Those are business life cycle stages: a business is launched, grows, matures, and then either renews itself or declines.
• Lists built around funding follow investment rounds, as above.

The five here stop at growth on purpose. For a business of one, those are the stages where the important decisions are made, and each one has a clear test you can apply this month.

What Goes Wrong Between Startup Stages

1. Skipping problem-solution fit. Building first and testing the problem later is the most expensive order there is. See why most businesses fail before they start.
2. Moving on without passing the test. "Most people liked it" is not a pass. Decide the pass mark before you test, then hold to it.
3. Spending on growth before product-market fit. Advertising a product that does not keep customers buys you people who leave.
4. Writing a full business plan too early. A business plan is most useful once you have evidence from the first three stages to put in it.
5. Treating a stage as finished forever. A new customer group or a new product sends you back to an earlier stage. That is normal, not a failure.

Frequently Asked Questions