Can htaccess redirects fool AdSense to not see the traffic?

[trader]

We recently purchased 3 domain names which have traffic going to the old websites, most of which comes from a number of outside links and goes to various htm pages, not the index page.

Domain-A according to both weblizer and another counter gets 600 visits/day but get only 20/day impressions according to the adsense url channel.

Domain-G according to both weblizer and another counter gets 550 visits/day but get only 30/day impressions according to the adsense url channel.

That means an incredible 95% of the actual traffic is NOT being counted by Google Adsense!

Domain-S according to both weblizer and another counter gets 700 visits/day but get only 200/day impressions according to the adsense url channel.

That's 1,850/day but only 270 adsense impressions are reported, with very few clicks or income.

We have written Adsense support several times asking how this is possible? They reply with several possible common reasons but none seem applicable and all look somewhat far fetched.

The only unusual thing with these 3 domains (and which they all have in common) is the vast majority of their traffic is coming from links which go to old htm files which are not in the folder (I only puchased the domain, not the files, don't even know the old htm file names). All I have setup is one page (index.html) for each site.

Only a fraction of traffic goes directly from the links to the index file. Therefore, this htaccess file is installed:

ErrorDocument 401 /index.html
ErrorDocument 403 [domain.com...]
ErrorDocument 400 [domain.com...]
ErrorDocument 404 [domain.com...]
ErrorDocument 502 [domain.com...]
ErrorDocument 501 [domain.com...]

Is there some way the htaccess could be causing this major discrepancy? Also, if in fact that is the case does G perhaps not count both impressions and clicks due to this issue? Any solution? Thanks.

[jdMorgan]

I can't comment on how AdSense behaves, but I do want to warn you that your "ErrorDocument" directives are malformed, and will return a 302-Found (Moved Temporarily) server status instead of the desired server error response code.

The argument for an ErrorDocument directive must be a local URL-path, not a canonical URL. As a result, only your first ErrorDocument directive (for 401) will work as expected. The others will all return 302 status, and this may be contributing to your problem.

See the Apache ErrorDocument documentation [httpd.apache.org] for the specific warning about this issue.

Jim

[trader]

I prefer to not send the visitor to an error code page (though it may be more proper or correct), but instead simply auto redirect them to the home-page so I always get the traffic.

ErrorDocument 401 /index.html
ErrorDocument 403 [domain.com...]
ErrorDocument 400 [domain.com...]
ErrorDocument 404 [domain.com...]
ErrorDocument 502 [domain.com...]
ErrorDocument 501 [domain.com...]

Can you or anyone else please tell me what is wrong with the above as I did not understand your reply or the info at the link you gave. Thanks.

[jdMorgan]

Yes, you will totally confuse any robot that hits one of those errors -- 404 being the main concern.

Instead of this:
ErrorDocument 400 http://domain.com/index.html (using a canonical URL is incorrect)

you should use this:
ErrorDocument 400 /index.html (using a local URL-path is correct)

If you wish to capture the traffic without confusing the robots (including mediabot for AdSense) and risking duplicate-content problems, then do a proper custom error page, and include a text link to your home page. An HTML meta-refresh to "/" or "index.html" with a timer of more than 5 seconds would also be acceptable, but make sure the 'bots have time to find the text link first.

While we all have "preferences" we also have to abide by the HTTP/1.x protocols, specifically the meanings of HTTP server response codes [w3.org]. This is the technical foundation of the Web. It's just not optional.

Jim