Search was getting me nowhere so let's try the more direct approach... btw I'm running Google Analytics, which may or may not be useful information to you folks.
1. How is it possible for exit rate and bounce rate to be vastly different on a keyword analsyis when bounce is not measured by the 5-second rule? If the stats are telling you a user came to a page via "keyword-x" from a search engine, and then left the site, wouldn't the bounce and exit rate of that page via the keyword's traffic be nearly identical? Remember that Analytics goes by the visit-and-leave rule for bounce rate, not the 5-second rule (unless I've been wrong about that this whole time). I'm sure there are a couple of users who come to the site via keyword, go through a few pages, end up back where they started, and leave... that would be a keyword-initiated exit, and not a bounce, yes? But some of my bounce and exit rates are way too far off for that to be the reason... at least it seems that way.
2. How is it possible to have a 50% bounce rate on a page that got an odd number of total and unique visits? Whether a stats program measures bounce off total visits or uniques, it seems illogical that a page with 13 visits could have a 50% bounce rate. At higher volume, I can understand where maybe the percentage is rounded up.
3. How can a single-page site (like an SEO landing page, with nothing but external links) have a page view average of 1.5? The only thing I can think of is that stat programs like GA give a 2 or 3 second buffer for people leaving the site to still count as one visitor... so if you're on Page A, click through to external Site B, then bounce back out to Page A via your back button, the stat program says you're one visit to two pages instead of two visits to one page. I suppose stats could also count a refreshed page as a two-page visit, but I highly doubt that half my landing page traffic is hitting the refresh button.
4. Does anyone else find Google Analytics to be glitchy on the time-on-site stat? I just yesterday put up a new domain, and sent direct traffic to the index page 4 times to make sure the stats were operable. Today my measured results were 2 uniques, 4 total visits, avg. time on site 5:36, 50% bounce rate. In actuality I bounced out of all 4 visits (assuming again that GA doesn't go by the 5-second rule) and certainly didn't spend anywhere near 5 minutes total on the page. What gives?
Any and all replies appreciated. I've always just ignored these gaps in my knowledge, but I'm going to be presenting to the CFO next week and I'm sure she'll spot any statistic that doesn't inherently make sense, so I'd like to have an answer.
[edited by: Don_Hoagie at 1:47 pm (utc) on Sep. 1, 2007]