Plenty, and we see it every day, except they don't call it "failure by doing too much". They call it things like "not understanding the target market" or "focusing on features rather than experience" (more generally "not thinking enough")
I can't think of a startup, but I can think of people like that. The first people that come to mind are the Mythbuster people. Very handy, very good at cranking out their ideas into meat-space - but my impression is that there is very very little critical thinking. This is probably in large part due to the demands of TV (You need cool gadgets and experiments - who cares if the experiment is poorly conceived when it looks cool?)
But a lot of hobbyists also fall in to that category: doing projects that have already been done, or ones where the solution isn't particularly intellectually exciting/innovative/stimulating but just mostly takes a bunch of time (ex: spending your weekend programming a PLD to control a stop light).
It's kinda hard to accuse these people of "doing" too much.. b/c doing is better than just sitting watching TV. But say instead of programming a stoplight all weekend you for instance read some textbooks and learning some higher level math, or read some blogs to get some inspiration, or read some scifi to get in a more creative mood - It gets kinda harder to make an objective comparison of which is "better".
technical debt
Technical Debt = When you coded without thinking AND your project is successful and growing. It's debt because now you have to go back and fix/change the code.
Technical Grant = When you coded without thinking BUT your project is NOT successful and fails. It's a grant because it saved you time.
Most projects / startups fail so for most people "Just Do It" works better. Plus the product you envision is almost never the product you end up with. So all that thinking you invested can be a waste of time.
Improving skills can take the form of learning a new tool (css -> sass), a new programming pattern (like currying and monoids in functional programming), and more along those lines.
Granted, a huge problem with technical debt comes when your playing with a team, and if X people are going to be using your code, but it takes Y time to explain how that code works, then if takes Z time to refactor that code, and only W time to explain after the refactoring. Then, if XY > Z + XW, refactor every time!
But I would bet that if you let the market dictate too much too soon, it will be harder to truly innovate, which is usually what the original vision was about. I'm obviously making some pretty big assumptions here but people know only what they know, and sometimes it takes carefully timed and constructed execution to snap them out of it to create a new paradigm and take things to the next level.
http://www.codinghorror.com/blog/2009/02/paying-down-your-te...
It's almost a Maserati problem when you think about it.
Had the technical debt been paid up front, the iteration rate might have been much faster (e.g., good encapsulation and abstraction allows you to switch out things quickly to test different mechanics in a game), leading to a better end product that people will pay for.
In contrast, as you say, there are huge amounts of armchair ideas that are not worth very much even if they are correct. For example "software will eat everything" is something I believe is mostly right. It is also a business idea with direct implications for action. Despite this, it is only valuable to set the general direction (let's do software), but not any specific cases. It does not mean that every case of transforming a traditional business into a software one is going to be a grand success, or even profitable. In fact, these types of ideas have a tendency to cause one to forget the exceptions from the rule, and thereby being actively harmful.
Maybe it's because many of the ideas in the "ideas are nothing" category are twitbook/fourit/squaresnap/flickchat/... clones, rather than ideas like "let's build a product for market X using tech Y to reduce problem Z".
(Edit: In some ways we do have such a framework. For example, I am sure that venture capitalists have a lot of paradigms for doing just that. But this meme of "ideas do not matter" still persists and is oft repeated by venture capitalists themselves.)
That being said, it's very, very difficult to dismiss mistakes, as wastes of time, as everything is a learning opportunity in the end. Of course mistakes have consequences related to resources (time, money, etc.), so it depends on how long your game is (and therefore how large of a margin for error you have).
But the bill came.
http://investor.groupon.com/financials.cfm
Groupon had (in q3 2013, the latest I could find) 595M in revenue and 14M in operating income, an EPS of 0.02 excluding "excluding stock-based compensation and acquisition-related benefit", and 1.1B in the bank.
Not a smashing success, but is it a failure? The jury is still out on this, imho.
Given that success and failure are both relative terms, I would say Groupon is only a "failure" if you're measuring it by the perceived success it would have. So I would say that if you expected Groupon to be a $10bil company, they are a failure...but if you expected Groupon to be a normal company of a few million in revenue, then they are an amazing success.
Although Groupon may not have had the most lucrative business model in the world, I still feel like they've done a great job at "sticking to their guns" and not trying to enter all kinds of other markets (and failing miserably).
In less than one year
The ones that pivot endlessly and run out of cash?
I'm interested to read at least one post on the opposite advice -- what happens when you're too heavy on the "doing" end of the spectrum? Are there examples of startups failing because of too much emphasis of execution? At the very least, the "ideas are nothing, execution is everything" meme is clearly logically flawed. You can execute as fast as possible while creating zero customer value.