Google is a great engineering company, but it still doesn't "get" the enterprise.
For a different domain that I was using regular Google Apps on, I couldn't get an email response to a support question I sent several months prior.
Whatever their focus, it's not on customer service.
One example: design. I still haven't seen a single page from Google that looked stunning, or even good (to my eyes, obviously). Their entire design philosophy seems to be infinite-amount-of-monkeys-hacking-css and A/B-testing. They even manage to make their main landing page, with only a handful of items, look quite messy.
The method may be scientific and all, but it doesn't produce good looks.
The author of this article backs up his argument with little substance.
To me Google have excelled at both the technology and the marketing - which is why they are so successful.
Having said that, if you can't manage to be good at both technology and marketing, I suspect that you are better off being weak at technology than marketing. Plenty companies thrive with average/poor technology but an excellent understanding of what the market wants. I've known plenty companies (some from the inside) where the technology was good but the marketing was weak - things generally don't turn out too well.
And I do seem to recall an overwhelming success like Windows starting off as somewhat of a failure.
Maybe they are successful because they do more rights than wrongs. Or they were in the right place at the right time. Or because they lacked a better competitor. The history of IT is littered with wildly successful yet sub-optimal companies.
It's very easy to look at a company like Google and find examples of significant failures.
Though actually, I think the 'failures' he brings up are not quite the failures he thinks they are. They're not actually data-driven decisions. They're data-collecting experiments. Google is in no way wedded to any given device or approach. It's pivot, pivot, pivot, often in eight directions at once. Releasing a polished product with top-notch marketing is a good way to make a lot of money in the short term, but your customers get very attached to what you've given them, and it makes it a lot harder to innovate in the long term.
When interviewing prospective software engineer employees, a Google interviewer asks software engineering related questions, to test the interviewee's knowledge of software engineering. Questions like these:
http://jpaint.drizzlehosting.com/google.html
That's a nice illustration of the balance of creativity, lateral thinking, logical thinking, and technical competence Google expects from the engineers it interviews. Note that the interview questions aren't in the form of, "Where did you complete your expected MSc in Computer Science?" Instead, the interviewee is expected to demonstrate real achievement from their education; i.e., knowledge and thinking ability.
But Google hires non-engineer craftsmen, too. Catering staff, to steal the article author's example. Just a hunch: Google conducts more than one type of interview: catering staff are not expected to design class libraries or database tables as part of their interview process.
I think that strategy is unimpeachable. I've worked at plenty of non-engineering companies, and the inability to question assumptions seems rampant. If Google manages to avoid that, then whatever it costs will likely be worth it.