Google Analytics: Average Time On Site

[pintofmilk]

Tweet

Looking at the Analytics info for one of our sites this evening I noticed that a massiv 70% of visitors from search engines had the following stats:


Visits Pages/Visit Avg. Time on Site % New Visits Bounce Rate
1 1.00 00:00:00 100.00% 100.00%


Does this mean any of the following or something completely different?
- the visitor left the site in less than 1 second
- the page did not load properly
- analytics javascript did not run properly

Is this something a lot of you are seeing in your stats?

Ben

[Fruit_and_Veg]

- the visitor left the site in less than 1 second

This one.

Someone correct me if I'm wrong, but the 100% bounce rate means someone clicked on your link, saw your site and didn't want to go any further so hit the back button. They didn't close the window or navigate anywhere else, they hit the back button.

You may find in your keyword stats that you ranking for words which aren't relevant to your site, eg. looking at your site in your sig, I could search for 'rugby hotels' land on your site and want a hotel in the town of Rugby not a hotel for a rugby match, therefore I hit the back button and continue searching elsewhere.

It's unavoidable really, you're always going to get a fair chunk of bounces although 70% does seem high. You can try to limit it by making your site more relevant to these terms, or setup new pages for these terms so they rank higher than the current page (eg. a Rugby hotels page which has hotels in Rugby), or simply remove irrelevant phrases as they are attracting useless traffic which you can't monetise.

[moredial]

The only way a script can know how long you have been on a page is if it can gather the landing time on the next page of your site.

All visitors who visit only one page on your site, even if they spent an hour reading it, will show as an immediate bounce.

Where you have relevant landing pages there is often no need for a visitor to look further into your site if all they are after is info. If you want them to take an action like buy or contact you, then a high bounce can be depressing - shows bad SEO if too many have to leave to find the info they really need.

The only way to solve the problem is by tracking each visitor's path through your site to see what search terms are bouncing and which are resulting in more pages being visited and a conversion being made. Improve the SEO for the conversion phrases and decrease for the bounces as they are just costing bandwidth.

Unfortunately the above can be very difficult if all you are doing is forwarding people through an affiliate link - create a redirect page between the link on your landing page and the affiliate redirect page to catch this traffic.

[pintofmilk]

Thanks for the replies.

The terms that people are finding us on are on target for the pages they are viewing so the user does not need to click elsewhere. I think that because pages use ajax then analytics doesn't pick this up this user/site interaction. Would this contribute to the stats we are seeing?

The other aspect I thought that might explain the stats is slow loading pages - maybe people get bored of waiting and hit back button. How quickly or slowly does this page load for you guys?
The Jungle hotels: hotels near The Jungle rugby stadium, England.

[moredial]

How quickly or slowly does this page load for you guys?
The Jungle hotels: hotels near The Jungle rugby stadium, England.

It loads within about 10-12 seconds on 2 Gb ADSL which is a bit slow if they are to do any reading within the 10 second window. The background map graphic is not loading - that may be because the browser security settings block any 3rd party content.

It sounds like you need to update your ajax to include some kind of tracking - no, I don't have a clue as I use php which does enable me to track visitors on dynamic pages through the POST call.

[babrees]

How quickly or slowly does this page load for you guys?
The Jungle hotels: hotels near The Jungle rugby stadium, England.

It doesn't! error....

You have an error in your SQL syntax; check the manual that corresponds to your MySQL server version for the right syntax to use near ')) * cos(radians(th.longitude)-radians()) + sin(pi()/2-radians(90-th.latitude)) ' at line 5 in SELECT th.*, "" as landmark_name, tr.region_name as city, ROUND(3958.78 * acos(cos(pi()/2-radians(90-th.latitude)) * cos(pi()/2-radians(90-)) * cos(radians(th.longitude)-radians()) + sin(pi()/2-radians(90-th.latitude)) * sin(pi()/2-radians(90-))) , 2) as distance FROM `tbl_hotel` as th LEFT JOIN `tbl_region` as tr ON th.region_id = tr.region_id WHERE th.`active` = '1' HAVING `distance` < '4' ORDER BY distance ASC LIMIT 0,50

[pintofmilk]

Thanks Babrees, that's fixed now

[pintofmilk]

moredial, we are implementing a solution to call tracking during AJAX calls.

Any ideas how to call Google analytics in AJAX calls? Need to avoid sending Analytics the same URL for the page being viewed as this will just look like a page refresh!

Thanks
Ben

[D-Mac]

try this:
Adam Howitt - Using Google Analytics to Monitor AJAX Applications

You could also log details of users' actions in your database, which might give you more useful info than analytics can, depending on how everything is configured. For example, you could log each search for country, stadium, star rating, distance, etc in each row of a database, together with a user/session ID, which cross references to how the person found your site. You could then pass the userID to the affiliate networks as a tracking ID, to see exactly where your sales are coming from.

ps. your site loaded pretty quickly for me. Certainly not slow.

[pintofmilk]

Thanks for the response D-Mac... we found that same URL (Adam Howitt's) an hour or so after posting the message!

Glad the site was quick for you - we have spent quite a bit of time improving the speed.

Ben