Given that both 90% figures are just rough general estimates without looking at the data (that is to say "made up") you can't really compare them. Even if they

Given that both 90% figures are just rough general estimates without looking at the data (that is to say "made up") you can't really compare them. Even if they where both data backed statistics you could imagine that maybe TechCrunch tends to cover preferentially the more interesting and exciting shoot for the moon types of things over the more boring "we're making money now" types which would skew their results. I don't actually read TechCrunch so of course I have no idea and that's just speculation based on nothing.