I'm a U.S. site. My U.S, metrics are Ok-good.
My metrics from the Philippines are amazing. I never knew this until I used GA about 9 months ago.
Since then my traffic from the Philippines has increased every month. Nothing crazy, but definitely 10% each month. So I would say that there must be some correlation.
Hmm that would indicate a correlation between user metrics for a country and the countries SERPs - the intersting thing would be then if you saw feedback into your US results.
I'm finding that user metrics do play a large role but so far, none of my data indicates it being country-based.
Blocking users from a country from accessing your pages seems extremely foolish and would have a negative effect on user metrics.
A user from Russia finds your site in Google and sees "Your region has been restricted from viewing our content".
Do you think this would improve your overall bounce rate and time on site?
I'm finding good value in using the new Analytics feature, Adjusted Bounce Rate [analytics.blogspot.com]
This custom setup requires just one more line in your tracking script. It lets you set a threshold to separate those long time-on-page bounces from short time-on-page bounces. It generates a very helpful metric for separating "real" problems from pages that are not truly a problem.
Even if Google isn't directly using bounce rate and time-on-page in organic rankings, no doubt in my mind that they are aiming at something in the area of user engagement (and so is Bing, for that matter.) So addressing areas like this in your metrics is directly improving your user engagement and in turn that most likely has a secondary effect on whatever Google is measuring. Besides that, I love working directly with user experience rather than indirect and merely technical SEO.
[edited by: tedster at 2:39 am (utc) on Sep 14, 2012]
Thanks so much for the link, tedster! This is going to help out in so many ways.
Andem - it wouldn't be a case of them getting to the site and getting a response as you suggest, the firewall would be set to block the country.
In the short term it may have a small negative effect but the longer term gain would be worth it if user metrics are not country specific.
I don't think there's enough evidence that the bounce rate is used in this way to justify blocking entire countries. Matt Cutts stated that bounce rate is not used as a factor (at least not as measured by GA).
I think it is worth focussing on engagement as much as you can and bounce rate improvements overall are an indicator that you're doing something right. But just looking at GA bounce rate numbers and then blocking off whole countries as a result seems wrong.
Having read the Adjusted bounce rate link - I'm not sure how it would be of use.
Someone whois is price matching products is going to visit your site for only a short time and bounce if your price isn't good enough - if they find exactly what they want (i.e. the product with the price) on the page they land on - visitor has bounced but they found exactly what they wanted, page was useful to the visitor.
If they don't find the product and the price on the page they land on - but maybe find a list of product variants they are then going to click to the next page to find the exact product and price. In this case you have more user engagement, but the landing page was actually less useful for the visitor.
Someone who stays on your page a long time and bounces may still only have been price matching the products, but did it by opening all the SERPs in new tabs and then going through them to find the prices (that is what I often do when looking for the best deal and may keep some of them open while I make my purchase just in case there are catches like high delivery charges, too many personal details requested in the checkout and such like)
I'm not sure long bounce, short bounce or number of pages visited is actually an indicator, for an ecommerce site, of good user engagement at all.
You can put together a similar argument for some informational sites too, espcially if thy are providing specific answers - if the user has to click through to find what they want from the landing page then the landing page is wrong, the user engagement created by the user having to click through to a secondary page to find what they wanted is actually wasting the users time whereas for the user who finds what they want in a second on the landing page and then closes the page the landing page was spot on and the user got what they wanted with the minimum of fuss.
This is what I find really difficult with User Metrics of every kind - it is difficult to decide what is good and if you make your decision as to what is good and it doesn't match Google's you're not going to be in good from a traffic point of view.
@IanTurner: I think it would be nice to have the default bounce rate and the adjusted bounce rate. The latter helps me to determine whether users are satisfied at all with the page they have landed on and more importantly, whether they represent the people who immediately hit the back button when coming from search results (thus seeing the "Block this site" link on the SERPs).
Trying to draw a distinction between soft bounces (spends at least a few seconds on page) and hard bounces (gone within 1s) is useful for deciding what to focus your efforts on.
If they are mostly hard bounces, you can bet it's an instant decision on the part of the user; adding more content won't fix this, but improving your design, ad positioning or use of popups, etc might.
If there are a lot of soft bounces, then you can look at maybe improving the content or other links on page to try and keep them on site.
If they are spending a lot of time on the page and then bouncing, the content is probably good and the only thing to focus on would be additional calls-to-action to get them to become repeat visitors.