How to Start a Startup

The traditional approach to starting a new business involves writing a business plan, pitching the idea to investors, hiring a team, building and introducing your product, and selling as hard as you can. Unfortunately, this approach has led to startup failure rates of 75 percent.

An alternative approach has emerged that can reduce the risk of starting a startup. It’s a methodology known as the “lean startup,” which emphasizes experimentation over extensive planning, customer feedback over intuition, and iterative design over up-front product development. Today’s lean startups are improving their chances for success by following the principles of failing fast, pivoting, and continuing to innovate.

start a startup

Here are four steps essential for today’s startups:

1. Consider problems first

Many entrepreneurs fail to recognize that a build-first or investor-first approach is backward and seldom leads to a successful venture. This is known as the Innovator’s Bias, or the entrepreneurial urge to build out perceived solutions without fully understanding the customer’s actual problems.

Entrepreneurs need to adopt the mindset of loving the problem, not their solutions. The key is realizing that customers don’t care about solutions per se; they simply want their problems to go away.

And problems, not solutions, drive innovation. Even if you start with a small solution – like a minimum viable product – it won’t get noticed if that solution doesn’t nail some customer problem. 

A better approach is starting with problems before solutions. Identifying and truly understanding customer problems ensure that you avoid building something that nobody wants. The challenge today isn’t building a product but uncovering the right product to build.

Investors today don’t fund solutions just because they may be technically elegant or possess some other “gee-whiz” factor; they are looking for business models that work. In fact, the business model, not the solution, is the product. From an investor’s perspective, traction matters more than anything else. It’s evidence that customers actually care about your idea. And more importantly, it’s also early evidence of a sound business model.

2. How is the problem solved today?

Entrepreneurs must accept the fact that – no matter how bright their ideas might be – those proposed new solutions will usually have to compete with existing solutions, even if those existing solutions are nothing more than slapdash improvisation. Customers are using something to address critical problems, even if it’s suboptimal. Customer discovery is a process that tries to understand customer problems, needs, and contexts: in other words, the existing business model, with all of its warts, gaps and kinks.

A lean-thinking entrepreneur will do more than simply ask customers what they want; he/she will look for data to provide a quantitative picture of the problem. If a customer declares a problem as a fundamental issue, probe deeper: Ask them how they solve the problem today. If they are doing nothing and still getting by – and the complementary data shows no reason for concern (i.e., wasteful costs) – perhaps it falls into the category of a necessary annoyance.  If, on the other hand, they are a) using a home-grown or competitor’s solution, b) not happy with the jerry-rigged solution, and c) wasting time and money, the problem may be one worth solving.

Data on the customer’s key metrics related to the problem will help identify and prioritize specific pressure points that would respond to new solutions. Running one-on-one problem interviews is the best way to understand the human dimension of the situation. One-on-one interviews may not seem efficient, but they pack more learning per unit time than anything else you could possibly be doing. You also don’t need as many qualitative or quantitative data points as you might think to start finding actionable patterns.

3. Define your idea on how to solve the problem

The Lean Canvas is a template for a one-page business plan created by Ash Maurya that deconstructs ideas into their critical underlying assumptions. Adapted from Alex Osterwalder’s Business Model Canvas and optimized for Lean Startups, it replaces elaborate business plans with a single-page business model.

Traditional business plans are like military battle plans: they take a long time to write, are seldom updated, rarely read by others, and quickly outdated. Said another way, they’re useful until the action starts, and then everything changes. Still, documenting your product idea one hypothesis one way or another is essential. Lean Canvas solves this problem using a one-page business model that takes under 20 minutes to create.

You can outline multiple possible business models on a canvas in one afternoon. Lean Canvas forces you to distill the essence of your product. This is invaluable when trying to grab an investor’s attention in an elevator ride or a customer on your landing page.

Successful ideas need to constantly balance three types of risks: customer risks, market risks, and technical risks. The sweet spot for ideas is at the intersection of desirability, viability, and feasibility.

The complete Lean Canvas helps to baseline an idea and understand the business model as the product. However, in the early stages, it helps to focus more narrowly on just two of the outer boxes of the canvas with a tool known as the Lean(er) Canvas. Under Customer Segments, you succinctly define the customer and how you will qualify/identify early adopters. Under Problem, you list your early adopters’ top 3 problems and how they solve the problem today.

These early underlying assumptions are fundamental, and if you get them wrong, everything else in the business model falls apart. Additionally, you can also do an early-phase risk analysis using a customer/problem quadrant to test desirability, viability, and feasibility.  

  • Desirability – Getting a customer’s attention with a compelling, unique value proposition (UVP) is the first challenge. A good UVP narrowly targets early adopters and specifically nails a problem.  You want the customer to say, “Gotta have it!”
  • Viability – Can you make and deliver a solution at a price acceptable to the customer and generates an acceptable profit for you? You want the customer to say, “What a bargain!”
  • Feasibility – Most entrepreneurs start with a solution and then search for related problems. Unless you are content to be a mad scientist working alone in your lab on esoteric and impractical ideas, it’s more effective to reverse the order. You want the customer to say, “It works like a charm!”

4. Host conversations about the problem

It’s important to gain different perspectives to determine the real problem and solve it.  

Steve Blank is an adjunct professor at Stanford University, a senior fellow at Columbia University, and a lecturer at the University of California, Berkeley. He published ‘Why the Lean Start-Up Changes Everything’ in the Harvard Business Review and emphasizes the importance of getting a wide range of perspectives and input on the problem.

Blank suggests that entrepreneurs can ditch the laborious and protracted process of planning and research. Instead, he advises entrepreneurs to accept the cold hard truth that all they have on day one is a series of untested hypotheses—basically, good guesses. And they can use lean processes to test their hypotheses with a series of experiments with potential customers and others in the business ecosystem.

One of these processes is called the “get out of the building” approach. Lean startups go out and ask potential users, purchasers, and partners for feedback on all aspects of the business model, including product features, pricing, distribution channels, and affordable customer acquisition strategies. 

Talking to customers seems simple, but most startup founders find it’s one of the hardest things they have to do. Many founders believe that such talk is a waste of time; they are convinced that they already understand a customer’s problem and only need to spend their time building a solution. But the data proves that belief is almost always wrong. To make products that people want and will use, founders first need to validate the problem/need, then validate their solution solves that problem (i.e., finding product/market fit). Finally, to have a better chance of a viable enterprise, they need to test all the other hypotheses in their business/mission model (pricing, demand creation, revenue, costs, etc.)

Becoming comfortable with customer conversations is essential. The objective is not to ask them what they want but determine their unmet needs or blind spots. Understanding a problem better than your customers is incredibly powerful. As noted by Ash Maurya, “when you can articulate a problem more clearly than your customers, they automatically assume that you must also have a solution. So, they pay more attention to you and share even more. This helps in further prioritizing the obstacles that stand in the way of their desired outcomes and allows you to build a solution that fits.”  

The key is to focus on learning, not pitching. A pitch is based on knowing the right product for the customer or Problem/Solution Fit. And that requires adopting a learning mindset to identify and understand the right customer problem. After setting the context, let the customer do most of the talking. It’s unnecessary to know all of the answers, and every customer interaction turns into an opportunity for learning.


Kent Nutt

About Kent Nutt

Kent is a marketing and communications entrepreneur in high-tech and nonprofit industries. He’s a strategic thinker and hands-on implementer with experience building end-to-end, transformative marketing and communications programs for startups, Fortune 500 companies, growing nonprofits, and academia. Sectors included telecommunications, 3D printing, network software, semiconductors, web analytics, and scientific research.