"You can see that other than the top-line metric of total signups, the other metrics are quite solid." Try as you might, I don't think it's possible to downpla

"You can see that other than the top-line metric of total signups, the other metrics are quite solid." Try as you might, I don't think it's possible to downplay that statistic. The overall numbers of signups need to be higher than they were in 2003, but anecdotally it seems like the percentage of the market you need to capture (and percentages of the market the startup will reasonably capture) haven't changed drastically. 1,000,000 users used to be almost the entire iOS App Store market - now it's a small sliver. Everpix was the rare scenario in which there's a solid product, solid team, and traction that seems like great traction, but casts too much doubt on its ability to grow to where it needs to to justify investment. After the exception (that will always exist), your seed round says, "Alright, I think this could have potential. Let's try an experiment." Your Series A says, "Experiment successful. Let's grow this thing." There are more experiments being tried, but if you take those same ratios into 2003 (user adoption would be much slower, and the numbers would be lower), and Everpix would still be a failed experiment. It's harder to hit x users than x/2 users, surely, but it's easier than ever to get x users. I would argue easier than it used to be to get x/2 users. The ratios of companies not getting a Series A is different, but that's largely because more experiments (seed rounds) are tried than ever. It seems like you can raise a seed round (<$50K) on almost nothing. Raising a Series A is still hard. The target is moving, but so is the consumer. Comparing numbers today vs. numbers 5 years ago isn't a testament to the investment scenario changing, it's a testament to the Internet and the world changing.