Comments on: Why Google Allows Target.com to Spam Results

By: Cor

Cor — Tue, 26 Jan 2010 03:34:45 +0000

I can understand that larger retail organisations can receive a number of positions in the directory. It is all based on the turnover and interest surrounding the products.

By: Shiju Alex

Shiju Alex — Fri, 08 Jan 2010 09:24:05 +0000

That is an interesting discussion here. But I could follow it up only today.

I think that @Josh Driver @Jbaker @Jeff @KorbenDallas and @Justin have good points on how these pages get indexed.
Another way might be from the logs/reports of some server based analytics apps, which are displayed in html.

@James what I said was in the spirit of what @Sjan Evardsson mentioned. It is a page for something not found. Difference is that it was not found in the database (server). The found ones were also not on the web server but found in a db

I just made a fresh query on google for ‘Exercise Bike Clearance’ – http://www.google.com/search?q=Exercise+Bike+Clearance – and was not surprised. Blogs discussing this very issue has taken over the SERPs. Will a genuine product searcher prefer this scenario? Or would they have liked a ‘not found’ page from a shopping site?

Another interesting thing: the domain exercisebikeclearance.com was registered on 2009-12-23 and is 301ed to pedobearplush.com
Ever watchful webmasters!!!!

By: IrishWonder

IrishWonder — Sun, 03 Jan 2010 15:59:05 +0000

I have seen lesser known brands do the same on their site – and seems like there is no way to fight it for as long as Google doesn’t do anything about it itself. Google’s webmaster guidelines, for example, clearly state that having search result pages of a site indexable is a no-no – yet if you see a site doing that and complain about it to G they don’t do anything (in my experience, they may even index yet more of such pages – not going to out anyone but tried this personally with one especially nasty site and got 0 result)

By: Joe Devon

Joe Devon — Sun, 27 Dec 2009 03:28:24 +0000

I can understand why Target uses GET requests. If they use Akamai or a similar service, the GET allows them to cache the results.

Is it a 404?: “The server has not found anything matching the Request-URI”

Seems debatable to me. They are reaching the request URI which happens to be a search page which is informing people that the terms don’t CURRENTLY match any pages on the site. Maybe tomorrow those terms will match…

If the concern is Search Engines/Google indexing search pages aggressively, the solution is not a 404. It’s a meta tag: http://www.robotstxt.org/meta.html when there are no results.

In my opinion, search result pages shouldn’t appear in a search engine’s index at all. The resulting page should. Therefore there should be a robots.txt requesting search engines to not index ANY search pages… And the SE should reduce the weight of these results regardless…

By: Sjan Evardsson

Sjan Evardsson — Sat, 26 Dec 2009 19:28:26 +0000

I think the error here is in how Target is treating its search results. Because they are relying on GET searches (which result in unique URLs) they could (should) close the loop by being RESTful in answer by returning an actual 404 (with a useful search page as the result body.) This way, the experience looks the same to the end user, unless that end user is a crawler, which will see the 404 header and know that it should not be indexed.

This is my opinion, yours may vary.

By: Robert

Robert — Thu, 24 Dec 2009 16:32:25 +0000

These are not ERROR pages. That’s like calling this an error page too: http://www.google.com/search?hl=en&q=TEST_1222211&aq=f&oq=&aqi= and then saying that is duplicate content because the structure of the page is the same as this one http://www.google.com/search?hl=en&q=PIZZA_25612&aq=f&oq=&aqi=

Most website search engines will display a “product not found” message when you search for something that isn’t found.

This isn’t black hat seo, or spamming the search engines.

What Target is doing wouldn’t even come close to hurting someone’s SEO/SEM campaign. A lot more goes into ranking than just creating a bunch of pages.

The reason Target ranks for the “emery shallow” phrase is because it is LINKED TO on this website: http://www.purevolume.com/emery

It counts as a backlink, you’ll notice that this website also links over to BestBuy and a few other places where you can purchase the CD. It appears that Target’s website no longer has the CD available, thus produces the search result saying it is not available.

By: Outtanames999

Outtanames999 — Thu, 24 Dec 2009 11:59:46 +0000

Those of you mentioning favoritism to Amazon will not be surprised by the fine print at the bottom of Target’s pages that says POWERED BY AMAZON.COM.

By: Bruce

Bruce — Thu, 24 Dec 2009 11:10:40 +0000

You can be sure that Target did not “spam” a search like “Jon Payne is so hot” into Google.

What looks like is happening is that Target’s product catalog isn’t searchable from the web except through Target’s own search, probably in order to enforce site security on the database. In order to get their results out in Google search and get customers, they must have a deal where Google will pass search terms to the target.com internal search engine and then format the result. The trick on Target’s side is that rather than returning an error page, they are returning a “user friendly” empty search results page.

This is a Google error interpreting the results they get back from Target, not Target spam. Now if Google starts to filter out results from target.com containing the string “We could not find matches…”, and Target changes their empty search page to use different text, then I’ll go along with the spam allegation.

By: Justin

Justin — Wed, 23 Dec 2009 21:48:37 +0000

KorbenDallas makes sense. Others may be responsible for the links getting added to Google’s crawler. However, why is Google not smart enough to remove the pages as duplicated content?

By: Justin

Justin — Wed, 23 Dec 2009 21:46:13 +0000

Correct, those are not 404 pages. How ever, where and how are those search pages getting added to googles crawler? Somewhere, Target is leaving links to those empty search result pages. My guess is that recent searches are saved/stored someplace on Targets site and possibly even shown to users under a “Recent searches” area. I think and would hope Google would be smart enough to see that all but a few words differ on each of those pages and eliminate those pages as duplicated content pages.