Forum Moderators: DixonJones
Actually our bounce rate were 14% to 18% for more than a year, we haven’t done any thing except of one banner changes on the home page. Our bounce rate dramatically went to 50/60% on both direct and organic visitors ( even lots of visitors who come through searching our brand name has about 70/80% bounce rate which was less than 10% before).
Here is an overview
Direct visitor’s bounce rat = 34.77% currently while before it used to be 14 %
Google = 56.62% currently while before it used to be 14.37%
Yahoo = 66.86% currently while before it used to be 21.03%
etc
anyone has any experience with this issue, what could be the problem ?
many thanks in advance
no, because thats not a big changes .... just we have changed a banner
and let me tell also that the bounce rate increased after a sudden jump in direct visitors (around 1500 to 4000 etc). Before it we were getting around 1500 direct visitors while for 3 weeks we have been getting more than 3500 or like that.
but one more strange thing is that we are not just getting bounce rate on these direct visitors but also on organic ( Google + Yahoo etc), looks very very strange issue
any more helpful responses will be greatly appreciated !
thanks
Just wondering if it could be click fraud.
Yes we do, but Google analytics and Adwords admin shows normal visitors as of previous ….. actually only direct traffic has been increased and bounce rate has gone very high on both direct and Organic visitors.
We could say that the direct visitors are hitter etc, but even the bounce rate has been increased on organic visitors as well.
any more helpful replies would be greatly appric
and let me tell also that the bounce rate increased after a sudden jump in direct visitors (around 1500 to 4000 etc). Before it we were getting around 1500 direct visitors while for 3 weeks we have been getting more than 3500 or like that.
Sounds like some sort of automated attack maybe? To skew your CTR and other statistics?
Anytime I see referrers with high bounce rates, I'll tend to try and backtrack where they are coming from. In some instances, I find that there are a whole bunch of references to the target URI sitting out there in a bunch of junk directories, MFA sites, garbitrage sites, etc.
Same problems with time on site.
Something is wrong.
Do you have any changes in pattern for the visitors' networks (IP address, organization)? Could
Is the banner any different, technically? For example was the old banner in a frame? Was it Flash while the new one is not? and so on. It's possible that your visitors' experience has not changed, but the recorded data has changed.
The the bounce rate change on exactly the same day (or the day after) you changed the banner?
Does the banner contain any indexable text that would cause your home page to be indexed differently? For example, some new vocabulary that might cause your site to have good ranks on irrelevant searches?
Your best clues for this might be in your organic search terms. Look at the terms that have brought people to the site before and after. Any big differences?
Also you should look more closely at your referrers. A visit can look like a direct visit if the visitor used IE but if they used Firefox the real referrer is usually not masked. If your analytics program allows you to isolate Firefox users, then look at that group for any significant new referrers that might be inappropriately sending irrelevant traffic to your site.
Lots of other things to check. I really doubt it's your banner, whatever it might be.
sorry if that's a stupid question, but I only know about the email-related one.
I assumed it was "landing on the site and leaving without looking at anything beyond the landing page."
zahirshah, have you checked your log files for the user agent of these extra visitors ?
a) I don't care how they label my site
b) It gives them a taste of their own medicine ;o)
BUT still the problem not solved
we have noticed that one of “network location” which sends the false direct visitors mostly (the same network was sending around 1400 visitors while currently sends more then 2300), there are some more also, if a network location was sending 80 visitors before the same one is now sending 150 or like that.
now thing is that, the location which sends more visitors (1400 before while now 2300 one), there are more than half of visitors are real from that location, so we can’t block all of them. Or if we talk about to pick the particular IPs which has the bounce rate, that will take lots of time (and will not be an accurate way).
so any other solution how to detect it, how to know what would be the real problem, what if we change our hosting server etc ?
Any suggestion would be greatly appreciated guys
The cause is AVG LinkScanner if your "bouncing visitors" have this user-agent:
Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1;1813)
If they don't, then the cause is something else.
[webmasterworld.com...]
Let suppose the cause is "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1;1813)"
So how can I solve it, what should I do then?
OR is there any way to confirm if its the "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1;1813)"
How can I detect it, where should I check it ?
many many thanks for the help
How you deal with it will depend on what package you use.
[webmasterworld.com...]
We use Google analytics if you are talking about it …… So how can we check the logs? Any procedure would be greatly appreciated?
thanks a lot Samizdata
If you use Google Analytics, I think you would set this up as a "Visitor Browser Program" filter
Presumably if you set the AVG user-agent to be filtered out you will see if the stats are any different, but I don't use any of these packages myself and cannot help you further.
You might otherwise ask your webhost how to access the raw log files.
...
[edited by: Samizdata at 1:24 pm (utc) on June 12, 2008]
Exclude Pattern: This type of filter excludes log file lines (hits) that match the Filter Pattern. Matching lines are ignored in their entirety; for example, a filter that excludes Netscape will also exclude all other information in that log line, such as visitor, path, referral, and domain information.
A little research suggests creating a new profile:
Filter Type: Custom
Filter Field: Visitor Browser Program (possibly required, possibly not)
Filter Pattern: ;1813)
I haven't tried it myself but it should dispense with AVG LinkScanner.
At least until Grisoft is forced to change it.
...
[edited by: Samizdata at 5:49 pm (utc) on June 12, 2008]
i will let you know if the problem solved or not
Please do, as I suspect it will be helpful to many webmasters.
If Google Analytics can't filter out AVG LinkScanner the bounce stats will be nonsense.
...
I tried filtering
“Filter Type: Custom
Filter Field: Visitor Browser Program (possibly required, possibly not)
Filter Pattern: ;1813) and as well SV1
but still not solved :(
any other way to do it, our ranking and site performance really going down for a month now (I think its because of this rude bounce rate) …….. i heard that it has been happening with some other webmasters as well, any one here who is facing it, any one has any experience with it ?
many many thanks in advance
At any rate, it occurs to me that AVG doesn't fetch images, and no hosted analytics solution (i.e. page tagged i.e. uses a gif request) solution would be affected by it. That would include Google Analytics.
Or is somebody seeing evidence to the contrary?
I've never seen a User Agent report in Google Analytics, is there one?
