The left-hand side: Pretty good, and most would do well to follow this advice. Remember, however, this was written by a VC firm. This isn't a list of things good companies do, it's a list of things that Sequoia portfolio companies do when they make Sequoia money. To be fair, the overlap is notable. Just remember where the incentives are for a piece like this.
The right-hand side: Steve Blank aptly defines a startup as "an organization formed to search for a repeatable and scalable business model." So a business plan written by a startup is, by definition, all lies. They're a fine exercise as long as you're grounded by the fact that you'll end up with 20 slides of pure bullshit. (Read: Don't put real effort into writing business plans, build a company instead.)
You can't validate a business idea on paper, but you can invalidate one. If you can't make the numbers add up in your best case, then you can save yourself a lot of time building a product that's destined to flop. What a business plan shows is that you know how to invalidate structurally flawed ideas and that you've found a starting point that appears to at least be numerically plausible.
For example, if you work out that your best theory for acquiring customers costs you $20, but their best-case total lifetime value is $15, you have a business model that you can throw out.
You're absolutely right. Like I said, it's a good exercise. A business plan that works on paper is, roughly speaking, a necessary but insufficient condition for success. So as such, you shouldn't bring it to a meeting with an investor or whatever and present it as proof that your business will work. The only thing it proves is you can dream up some numbers that result in an up-and-to-the-right graph.
What proves a business works is traction and revenue. Walking into Sequoia with "I've thought really hard about my startup and I'm pretty sure it'll work," is far less compelling than "my business is currently working and I need your money to scale it."
The only way to invalidate an idea is to prove that people don't want the product. While I wouldn't always bet on a business that has a proven demand, I definitely would not bet against one.
A great example is Instagram: I wouldn't have bet on that product because it seemed destined to make no money, but since so many people liked it they somehow found an exit.
That's nonsense. So long as we're talking about the mechanics of the idea, which is implicit when talking about a business plan, then yes, you most certainly can invalidate an idea. Just because people would like to have something doesn't mean you can either produce that or make money selling it at the price point they'd like to buy it at.
I'd like a hand-cranked machine that takes my daily garbage and spits out Ferraris in under 10 minutes and I'd like to pay $5 for this machine. I'm sure I can demonstrate demand quite easily. Will you fund it?
I was operating under the assumption of reasonable execution risk. If you can't even get the product out the door then you'll definitely fail. My main point was that it's hard to disprove ideas that are feasible and are something people want. To pursue such an idea is a gamble; to not pursue is also a gamble. I've unfortunately been wrong in both ways ;)
And how would you arrive at that assumption? Generally speaking a simple business plan is what states those risks and explains these essentials of your execution plan.
This "no business plan" crap everyone parrots is getting tired for me. Situations change, good companies learn and of course plans change. Just because a bunch of hipsters decided business plans aren't cool doesn't change the reality that, in general, failing to plan == planning to fail.
I wouldn't bet on anyone who couldn't answer those questions on the right. We are not talking some 100 page MBA bullshit, we are talking understanding what it is you are setting out to do. This is very, very basic stuff right here.
The right-hand side: Steve Blank aptly defines a startup as "an organization formed to search for a repeatable and scalable business model." So a business plan written by a startup is, by definition, all lies. They're a fine exercise as long as you're grounded by the fact that you'll end up with 20 slides of pure bullshit. (Read: Don't put real effort into writing business plans, build a company instead.)