You are expected to deliver. This doesn't mean 80-hour work weeks (at least in my end of the company), but you are expected to make goals and hold yourself to them -- and ideally deliver something that beats the rest of the industry while you're at it. Do not come to Google expecting a research environment; there's plenty of research going on, but if it's not in your 20% time you'd better have something to ship. I list this as a con only because I suspect this bit some of the people who rated Google negatively.
The company has grown really fast. This has caused some strain, both in terms of infrastructure and in communication. Many of the nooglers are fresh out of college, and the naivety cuts both ways: they're simultaneously over-eager to re-engineer everything in sight, and bitter when perks change or they don't get promoted fast (nevermind that it's still better than the rest of the industry -- they've never been there). With rapid growth comes a lack of focus, as well, with weird projects with no monetization story. (Caveat: I'm in Ads, where money matters.)
I get the impression that we've got our share of bad managers. My manager (and the two immediately above him) are all great -- technically sharp, caring, and happy to get their hands dirty. But I've heard horror stories from other teams, and you can read plenty of them on this site, I'm sure -- preference for low employee numbers, suckups and yes-men, etc. I suspect most of the thoroughly bitter people have gotten a bad manager and not requested a transfer out from under him/her.
On that note, Google does a bad job of handling underperformers. They try hard to help crappy engineers and managers improve, but at some point you've got to cut your losses -- a bad employee, particularly one with a bad attitude, is poison that eats away at morale. I've seen people go through two or more performance improvement plans before finally getting the much-needed axe.
Google also does a fair-to-middling job of handling *overperformers*. I've gotten a few spot bonuses for successful projects (which, to be fair, were tens of times larger than the bonuses I got at similarly-sized companies in the past), but the motivating effect of the Founder's Awards are gone. No matter how many millions of dollars you make the company, if you're not in LSE's pet segments, no Founder's Award for you -- it'll go to some guys in search or something. Google is done making millionaires -- for those of us who got who got here after the IPO, the only way we'll retire is a 401(k). Again, having worked in industry outside of Google, this makes me wistful but not angry. I don't expect other people to make me a millionare.
And, finally, your family will ask you questions about everything Google related. Be prepared to have to explain dozens of times that you don't actually work on Search/Maps/Writely/whatever and can't fix their pet problem. :-)
