Checking for Googlebot with Yahoo Store

[babyclassroom]

My store is hosted with Yahoo Store. We don't have access to our logs. We have "click trails" that show the IP address of visitors. However, when Googlebot visits, it only shows the IP address with no click trail. So -- I'm unable to tell which pages she visited.

Is there a solution which exisits to give me this information without me having to have access to my logs?

Thanks

[netcommr]

It has to be done with javascript on the pages. You can use one of the free tracker systems or if your handy with javascript do it yourself.

It should be very easy to add to all your pages with the global footer in your store builder.

[babyclassroom]

I'm using a tracking program (which I guess I can't mention here). The logs do NOT even show that Googlebot visited; however, the click trails in Yahoo Store Manger do show the visit.

I've been in Google off and on for the last 2 weeks, so I KNOW she visited.

I'm not at all handy with JAVA script. Can you give me an example of the code?

Thanks

[andreasfriedrich]

Considering that GoogleBot does not execute JavaScripts I´m not convinced that a JavaScript solution will really help. Referencing a web bug won´t help either, since GoogleBot does not request images.

While I have no idea whatsoever what click trails are I´m afraid you would have to use a server side solution. If you can use PHP or some other server side scripting language with Yahoo stores then you could build your own logfiles or log visits to a database.

BTW despite the similarities in their name JavaScript and Java are not related. JavaScript is not a scripting version of the Java language.

Andreas

[mack]

heres a quick idea that may work. to be honest I have never used yahoo store so I am not sure. Is it possible to insert a small image (even a 1x1 pixel) and have that image hosted on one of your other servers?, if so you could track that image in your other servers logs. forgive me if I am being stupid but I am not aware how yahoo store works.

[andreasfriedrich]

The problem with this approach is that GoogleBot won´t request images referenced in the page. This web bug technique is fine to track real users using real browsers. It is not suitable to track spiders.

Andreas

[netcommr]

I am sorry. don't know what I was thinking there. I just had tracking and the Yahoo store in my head without thinking about bots.

The only thing I can really think of is to put a link to an external document with no other references to that external document. Google should request this external document some time after they hit your site (not each page). It will not be an accurate account, but give you some idea when it is around.

I have talked with Yahoo about this in the past and they would not budge. No logs and no 'click trail' file downloads period. You really have no good option since the only way to really see bots is servers side which you have no access to.

-------------

andreasfriedrich, thanks for stepping in before I frustrated someone. The click trails function is a utility in the Yahoo store manager which shows visitors by IP showing which pages where requested and at what time. This is a very good thing to have if you can see it all, but all they give you is a link to see one at a time. Very frustrating to use and worthless if you have many visitors. It would take you all day to get any idea of what is going on.

[andreasfriedrich]

Thanks for the explanation netcommr.

If you can´t use PHP then you might consider using a local script to request all the click trail info.

Andreas

[toadhall]

Andreas...
> GoogleBot does not execute JavaScripts

So this trick wouldn't work to track SE bots?

Add this to the head of the Yahoo page(s)...
<script language=Javascript src="path/to remote/server.php"> </script>

And server.php looks like this...
<?
$id = uniqid(ID);
echo "document.write('IP: $REMOTE_ADDR / Linked from: $HTTP_REFERER / UA: $HTTP_USER_AGENT / ID: $id');";
?>

Which would return something like this...
IP: 127.0.0.1 / Linked from: [localhost...] / UA: Mozilla/4.0 (compatible; MSIE 6.0; Windows 98; Q312461) / ID: ID3dd9225a2cd4f

Of course you wouldn't "echo" the results to the Yahoo page, but save them to a log on the remote server.

CAVEAT:
This depends on a reliable remote server. If it's down a lot, or slow, then your store visitors will suffer.

[ciml]

toadhall, I don't think that you'll be able to get Googlebot to fetch the destination of a SCRIPT src attribute, at leat not as the page is crawled.

[andreasfriedrich]

So this trick wouldn't work to track SE bots?

Robert, I´m with ciml on this. Since Google does not execute JavaScript there is no point in fetching it right when it indexes a page. It might ftech it sometime to check for spam, but this will be no reliable way to track SE robots.

Andreas

[Brett_Tabke]

As far as I know, there is no way to track spiders or other text browsers on Yahoo stores.