I can't get past the paywall, but the info looks a lot like the survey that circulated this summer. In that study, the researchers went through tax data and use

I can't get past the paywall, but the info looks a lot like the survey that circulated this summer. In that study, the researchers went through tax data and used tech companies that hired at least one employee as their proxy for a "startup." This seems wildly flawed. It's the same methodological quirk that makes Kauffman data not so helpful. I'd like to do a study that looks at some of the more common metrics of success in the startup community. + Top 100 companies tracked by Gartner + Top 100 Apps on Apple Store + Top 100 Websites measured by Quantcast I've looked at sample that's pretty close to that and the average age is much closer to 30, and even artificially inflated by serial entrepreneurs who are on their third companies.