Tracking URLS and Dupe Content

[Receptional]

So - this question has been rumbling in my mind for some months now. I hope someone can shed a light for me without too much sniggering that the tracking and logging mod can't track :)

In New Orleans I heard two senior Google engineers give a Google "line" that either suggest they don't do that much coding themselves, or that I am not understanding server header responses correctly. Matt Cutts said it on the podium, and Paul Ahaare said it in the Engineers meeting. The question was:

"If people use "?source=googe" or "?source=whatever" on tracking URLS, how do you avoid Google indexing you for dupe content?"

The stock answer was: "When you do this, you should give a 301 message redirecting the user when going through the tracking link, so Google can figure it out"

Am I missing something here? But if you use a tracking URL, the page needs to start loading with the source= bit to be able actually TRACK. If you return a 301 message to a page without the tracking URL, you don't get to track the variable.

How do I return a 301 message redirecting a person to (say) a homepage, whilst still tracking that user's source?

Dixon.

[dcrombie]

If you issue a 301 then the tracking code will be recorded in the server logs. You could also set a cookie or session or call some other script before issuing the 301.

[Receptional]

You could also set a cookie or session or call some other script before issuing the 301.

Thanks DrCrombie. I can see that the server logs will probably pick it up, but when we tried it, we couldn't set a cookie or run a script before returning a server header message. Even when using PHP, we tried to run a script to work with Indextools, THEN return the server header, but the machine seemed to return the server header message first anyway.

Now - that was tried on a mchine using cpanel I think. We now have a plesk desk... but am I missing something or doing something wrong? I am admittedly not on the coal face on this, but intuitively, doing it using PHP was my plan, but it doesn't work. Do tyou think this is because I am trying to run a javascript? I need to do that to collect the user's environment variables and referrer info.

Am I just doing it wrong?

Dixon.

[plumsauce]

It's not that you are doing it wrong.

The problems is you have two goals which are not aligned.

First, you want to do the 301, second you want a .js to run.

It is entirely true that this can be done server side. But, your .js is not server side. This is not to say it cannot be done, but it *will* require some creativity. What does your vendor say?

p.

[Receptional]

Nothing yet, but I am out to confuse... I didn't call it the most urgent of support questions. yep - .js is not run server side. That's my basic problem, I can see now. Blindingly obvious of course.

I'll ponder the creatives... :)

Thanks.

[redzone]

seems to me the 301 would have to pass the referer to the redirect URL, so that Indextool's "js" file would pass it on to it's stats system.

I've worked quite a bit w/ indextools in the past, but almost exclusively on the paid search analytics side...

If you aren't getting a timely response from IT, sticky me the details, as we are a direct partner, and provide all the YSM/GAW bid cost data for their paid search analytics...

[Receptional]

No need Redzone we're a partner too. But thanks.

Dixon.

[plumsauce]

<<and provide all the YSM/GAW bid cost data for their paid search >>

how does this work?