Google's Supplemental Results?

Or, Going Supplemental?

A friend noticed that half the pages in his web site had suddenly slipped into the "supplemental" results in Google, and asked me to take a look. The pages appeared normal in a browser, but when I checked them with the W3C validator, they failed validation miserably with numerous errors. He had been working on the site and made some global navigation changes, but without realizing it, he had introduced these formatting errors.

After correcting the problems (and waiting rather impatiently for a few weeks) the pages came back out of the supplemental index. Matt Cutts from Google has repeatedly stated that 40% of the web pages on the Internet do not pass W3C validation, and it would be absurd to omit these from their index.

However, that does not mean that those non-validating pages will be able to compete for the top spots in the SERPs, or that those page will not "go supplemental" at some point in time.

So, if you are serious about SEO and are vying for the top spot in a competitive category, I strongly recommend that your site's pages pass the W3C validation tests before you panic and ask for help in one of the online SEO forums.

Please help Marcela a.k.a. 2_Much

Originally posted on 7/27/2006:

I gave a very small amount of my existing PayPal money, and I urge you to do the same. I don't even know Marcela. I have never met her in person or in any the SEO forums, but my heart goes out to her and her unborn child, Nathan.

To paraphrase one poster, "If you met her in a bar, you would think nothing of spending a few dollars to buy her a drink" so why not spend a few dollars to help her out in her time of need? And, if you can't afford it, then add her to your blog, or prayers, or whatever. Every little bit will help, I promise. You just may be surprised at how far a little help will go...

grnidone has created a donation page here: http://www.greeneyewire.com/help-marcela-aka-2_much/

Congratulations on baby Nathan's birth!!!

Latest news from Marcela is here: http://www.prayfornathan.org/news.php

If searchers switch to a different search engine?

It seems that search engines like to have data handed to them quickly, and if they have to go looking for it, they put the page on the "back-burner" and get to it later. With more than 5 billion pages to index and update regularly, it makes sense for them to update pages that respond the fastest with the information they want, and push the other pages to the back of the queue.

What three things do the search engines want? Page Title, META description, and to a much lesser extent, META keywords. So, put them in that order in the head section of a page. You can argue all you want that the order of placement does not matter, but when everything else is equal, a page that delivers that data quickly is more likely to win the top spot than a page that makes it difficult to find that info.

The same is true for W3C validation - Matt Cutts from Google makes a great point that 40% of the web consists of pages that do not validate, so it would be absurd to exclude those from Google's index. However, that does not mean that those "invalid" pages get the top spots for a given search term.

When a page fails to pass in the W3C validator, you can't trust that it will display properly in all browsers at all monitor resolutions. And if you can't trust that it will display properly, would you really give it the number one spot and risk alienating all of your clients (the searchers).

Remember - If searchers switch to a different search engine, the advertisers follow them, and so does all of the revenue that comes from advertisers.

Curse of the Query String...

Once upon a time, all web pages were static HTML pages, SEO was a very simple process, and life was good.

Then, dynamic content appeared so webmasters could query a database, but SEO got much more complicated. At first, the search engines did not read query strings, and webmasters everywhere complained. So, then the search engines changed so they could read the query sting, and that's when things got incredibly complicated.

"?source=abc" is an innocent way to track incoming traffic to a site. It's great because you can use it to track your PPC campaigns, affiliate links, and even internal site traffic.

However, "default.asp?source=abc" is read as a completely different URL than "default.asp?source=xyz" so if they both go the same page, it creates duplicate content (and this is true in all of the search engines).

Without realizing what was happening, millions of websites around the planet started using query strings and created duplicate content on an unbelievably grand scale. Then, webmasters everywhere started wondering why their positions in the SERPs began to sink. They blamed it on updates, algorithm changes, and some even thought they were being punished by some person sitting on the other side of the screen.

Few people seemed to notice the plethora of patents on detecting duplicate content that have been granted to the search engines. It seems that even fewer people make the connection to sinking SERPs for sites that supposedly have been online and on top of the charts for years.

But if you show me a site that is sinking in the SERPs, I can guarantee that the same site has duplicate content that has been growing exponentially behind the scenes. The latest "updates" to the search engine algorithms may merely be adding new ways to detect duplicate content, or the duplicity may have reached a point that has pushed the site "over the edge" and started it sinking.

And now, the search engines seem to be following each other's query strings that appear in text Ads, so Yahoo will show a Google PPC tracking string in a URL, which causes duplicate content each time a new URL (with a unique tracking code) is added to the index.

I have fixed this on one site by putting 404 server header codes to delete the old pages, and by including the META tag (META NAME="robots" CONTENT="noindex") whenever a query string is used.

However, this is a real problem for dynamic sites that use query strings for other purposes. A dynamic site could have a page URL of: "/dept.asp?did=6&page=1" and the PPC tracking code will be added onto that string so it becomes: "/dept.asp?did=6&page=1?source=google_ad_code" so the best solution is to use a unique identifier that is appended to the string, and only put out a 404 and add the META tag to the page when that string is detected.

Did I mention that SEO used to be a very simple process?