Starting something

What does the research actually say about starting a business?

This is the research on starting a business, checked against the papers rather than against the articles about the papers. The famous numbers mostly do not come from anywhere. The question the whole argument turns on appears not to have been studied at all.

I am inside the sample. I run two companies and my own outcome is unknown, which is worth saying at the top rather than leaving you to work it out.

What I found

• Where does the claim that 90 percent of startups fail come from? Not a weak study. No study. One report states it on page five with no citation and had no data that could have produced it. The other source never published anything.
• How many new businesses actually survive? About half at five years and a third at ten, from the government table. Two pages on that same government website disagree with each other by five points.
• Does money or character decide whether a business survives? It has been run once, on 653 German founders. Personality explains about 14 percent of what the model explains. What you brought and what you knew explain roughly four times that.
• If you start a business after losing your job, does it go worse? One Norwegian table has their income falling and their firms hiring more, three rows apart. And displaced founders survive longer, which turns out not to mean what it looks like.

Questions

If you are thinking about starting something.

These are the questions the research made me ask about my own situation. None of them has a right answer and most of the data you would need is yours, not published.

• What would have to be true for this to reach ten times your current target, and does your pricing allow that number at all?
• Which of the numbers you are quoting to yourself have you traced to a source, and which arrived already assembled?
• If this closed in three years, what would you expect the honest reason to be, and are you tracking that thing now?
• Are you measuring the business or measuring yourself, and would you notice if a good month for one was a bad month for the other?
• What result would embarrass the dashboard you look at most?
• Who told you the thing you are most confident about, and had they done it themselves or read about someone who had?

One caution about the last one. People who succeeded and wrote about it are a group with the failures removed, and the failures do not write. That is true of my writing here too, in the sense that I am still going. Ask me again in three years.