Before Urchin (Google Analytics) had goal funnels, I did a review of the product back in 2004. I then asked for Goal Based Funnel Analysis to be added to the product. They have added it soon after with Urchin 6.0 (not Google Analytics). Let me show you a simple overview of how it works.
When people come to my forums, I would like them to become a registered member, if they are not one as of yet. To do that, they need to click on the "Register" link in the top black bar under "Search Engine Roundtable Forums" header. Then they have to go through several pages to become a member. Those pages include a terms of service agreement, a form, a confirmation form, and so on.
So I set up a funnel to analyze if I am losing any of those people along the way. Meaning, do I lose someone who is interested in becoming a member when they are at any stage of the registration process.
I picked a specific time frame that I want to look at and then look at the goal funnel report.

This graphic shows that of the 1,513 people who came to the forum in this time frame, 24 of them moved on to click on the register link. So that is about a 2% conversion rate on that goal.
From the register terms page, which is where the register link takes you, those 24 users then have to proceed to the registration form by accepting the terms of service on that page. How many of the 24 users did that?

As you can see, all of them.
From there, it asks them to confirm the information and press submit. Do people fall off from there and leave?

Not these 24 users.
From there, they must verify that they have registered. They do that by checking their email and clicking on a link to prove they are a real user. Did everyone do that?

Yup.
That page once again thanks those people for verifying, so all 24 should get those verifications...

As you can see they do.

Finally, this is the last step in the process. 27 people have gone to the thank you for registering page. Wait, 27 people? What happened to the 24? Where did the 3 extra people come from? Well, they must of clicked on a link from that email once again, which knew they were registered. So it added to the 24.
Overall, a conversion rate of 1.78%.
One more quick example.
I have an advertiser information page, where prospective advertisers can learn about placing ads on my blog.
In a specific time frame, 42 people visited the page. Of those people, only 10 people (24%) went to the contact us form to ask for more information.

So while only 10 went to the contact form, an other 10 exited the site, 5 of those people went to the home page<, 1 person viewed a specific article, 1 person went to the subscribe page and the last person went to the recent comments page.
Now, as a marketer, one would want to see how they can influence more of those people to click on the contact page. To note, they may have used the email address on that page to contact us, which I currently do not track.