&id= Parameter

To change or not to change? Help!

[Whoa]

Googleguy recently wrote:

This was posted by a Google employee:

<i>I've been aching for a long time to mention somewhere official that sites shouldn't use "&id=" as a parameter if they want maximal Googlebot crawlage, for example. So many sites use "&id=" with session IDs that Googlebot usually avoids urls with that parameter.</i>

I've got a site with about 1,500 articles and all the URLs use "&id=9999" to identify the articles -- tons of them are in Google's index so I've never worried about the &id problem before. I had seen GoogleGuy post on it ages ago, and just let it slide because we were doing OK and because I figured eventually Google would move past that challenge, like all the other challenges they've nailed.

But in the last two big updates, including this recent one, we've seen traffic drop by 90%. In Allegra, our unique domain name was buried way deep below scrapers for a couple of months. In Bourbon, traffic is just way down and I haven't had time to figure out why.

Do you think I should change the id= parameter to article= or something like that? If so, is there any way to do it without completely losing all Google traffic for three months while Google figures out the change?

Any advice or thoughts would be appreciated. Thanks.

[Whoa]

I am bumping this up because nobody has replied to it.

Any help is appreciated. Thanks!

Again, my question boils down to this:

Does &id= only hurt your being crawled or do it potentially hurt your rankings and make you suffer for a month or two after every major update?

My site uses &id= and all the articles are in the index, so I'm wondering if it's worth a change.

[Lord Majestic]

I am speculating here but I think if value of ID is a simple small integer then its okay, or GG really meant "sid" rather than "id".

[inbound]

I guess that if your pages look too similar then Google will consider any parameter to be useless, if you create pages that are sufficiently unique then Google will see them as that.

The problem with using ID is that it gives Google another opportunity to add points to whatever similarity algo they use.

I'd say if your traffic is really low then it may be time to experiment, however I would not change pages that have good inbound links.

[vordmeister]

I suspect the use of id as a parameter is an issue. I built a site a couple of years ago using id=** (like they do in the books). It was initially spidered very well, but since the summer the id=** pages have been de-indexed.

Mod-rewrite is the ultimate answer, but I'd definately avoid building a site using?id= Better to use?page= or something. If it's an existing site using?id= and it's still nicely spidered then it's probably not a worry.

[Reid]

what I got from googleguy's comment was that &id= is often used for session id's (which will throw googlebot into an endless loop because it will treat each id as a seperate page) that goglebot will flag the string &id= and not index/follow it to avoid the loop problem.

You should not use &id= in your URL's

[sr123]

I actually posted a question related to this matter yesterday but have not had any responses yet.

My concern is that ANY parameter in the form of <some_thing>id=## may be getting nailed. For instance I use pages that have expertid=6 or answerid=12 or newsid=31, etc.

This is because I store much of my content in a database.

What I've noticed is that Google is not crawling some of my key pages that have such parameters in them. This is for the past 3 or 4 months (I believe). Also, the pages in this form that Google has indexed now just show up as URL's in the search results.

I don't think the problem is just limited specifically to?id=##.

It would be REALLY REALLY useful if Googleguy could clarify this one as the advice is not clear. In the worst case it almost makes me want to make all of these pages load as static html pages without any parameters. I don't want to have to do this unless I'm certain that this is the problem.

I have a bad feeling due to the way I see Googlebot avoiding pages like the ones I mentioned above.

It shouldn't be too difficult for Google to discern whether the parameter is a sessionid versus just an id that refers to an entry in a database driven site....

sr123

[Whoa]

I think I am just going to make the change to static URLs and see how that plays out.

Based on what I've read, I assume there's no downside to going to static addresses without any parameters being explicitly passed.

I just hope that the change does more good than harm, and that is doesn't take forever to get reindexed.

Thanks to all of you for the advice/comments.

[GoogleGuy]

If you're getting crawled, I wouldn't worry about it. I wouldn't be surprised if the "&id=" restrictions applied to longer parameters, not to 3-4 digits.

[sr123]

Thanks for the response Googleguy. Would it be possible for you to find out for certain what this applies to? Is it just?id= exactly that should be avoided? Or are things like?articleid= also effected?

Really appreciate your time and the energy you've taken to educate us here Googleguy!

sr123

[Reid]

What I've noticed is that Google is not crawling some of my key pages that have such parameters in them. This is for the past 3 or 4 months (I believe). Also, the pages in this form that Google has indexed now just show up as URL's in the search results.

it is very clear on the google webmaster guidelines that dynamic URL's take longer to crawl. This is because googlebot will limit itself to a specified # of pages on dynamic sites - to prevent overloading the server. URL only just means that the url is registered with google but not crawled yet.

[Johan007]

QueryStrings initially take longer to spider for new sites but are just as quick when the base page has been trusted.

Page.asp?ID=999999

is fine!

But thats a QueryString!

A session ID looks more like this:

Page.asp?sID=jj44t94k43k9gk34443wewq54