Michael Siebel has some great video discussing this exact topic. One thing he reiterates often is how most of these startups take too long to ship an MVP. They spend months, years in a vacuum burning cash and building something they think is needed, without ever actually testing their assumptions.

I notice this happens when teams lack someone with product management skills. It’s not enough to be a great engineer or a Steve Jobs level marketer. You need someone who tests assumptions in the field with real users and maps out what the market really wants.

Isn't this what YC is geared to help with? Get early-stage startups on the rails?

Yes but YC cannot force a team to “launch”

Many teams will stay in a perpetual loop of product development because they feel it’s not ready. YC has no control over this and are outspoken about this death spiral.

In one video Siebel reminisces about how many startups will reach out for help when they’re 30 days from running out of cash, instead of 6 months or a year earlier. At this point with no traction it’s too late to course correct.

https://youtu.be/C27RVio2rOs

That's very interesting. I did not know that was the case.

I've thought about how the reasons VC is so important can be ranked:

1. Validation is antidote to founder burnout 2. Collaboration improves iteration. 3. Money covers founder and startup expenses.

The money isn't the most important thing in my experience.

I think there's a big opportunity to shift the validation/collaboration to early stages and for more founders and the result will be more and better and bolder startups.

I'm pretty sure there's a giant unlock connecting founder vision, community, and VC.

Of course and they do a great job. But what is YC's acceptance vs. application rate. I imagine something more selective than Ivy League.

What about those not selected? What about show selected that fail?

Seems to me that the biggest problem we face today is centralized power and that the antidote to that is hyper-decentralization via tech entrepreneurship. We need like 1000 Elons and Bezos.

Most will keep failing, the trick is to fail sooner with less money burned.