Appropriate SEO strategies for a website's own SERPs?

[DanSerpico]

Hello all,

What are good on-page SEO practices for the search result pages on our own sites? For instance, what page titles do you use? Do you include page numbers? Meta-descriptions? Headers? Keyword utilization?

This is a consideration for us as we link to some popular search results on our sites.

Thanks!

[mjtaylor]

I would not be concerned with optimizing for internal search. You want to optimize for Google, Bing and Yahoo. The Learn SEO files here on SEOMoz should give you the basics, and you might find it very helpful to use the on page analysis tool to go over the basic onsite optimization factors.

Not sure I have ever used a page number on a site, though I do see sites that use them when they have a string of say, affiliate pages ... a way to answer this question is to ask: how will it help the visitor to my site?

[DanSerpico]

Thanks, M.-J.!

Yes, I am not concerned about optimizing my content pages for my search engine. Rather, I am concerned with optimizing my search engine result pages for Google/Bing/Yahoo.

For instance, suppose the page title was:

"kitten - Search Results | MyCompany.com"

What should page 2 of these results be?

"kitten - Search Results | Page 2 | MyCompany.com"

I ask because it seems that linking to popular searches on my own search engine seems like a reasonable strategy to help Google crawl a site and find content. What are the best on-page strategies?

I have consistently used page numbers on a site when I had multi-page articles and either my CMS or authors were unwilling/unable to create an appropriate page title for each subsequent page. (e.g, "My article title | MyCompany.com" and "My article title | Page 2 | MyCompany.com")

Thoughts?

[kadesmith]

I would ask the question, "What makes the most sense to the end user?"  The reason I would ask this question is because I have a strong feeling that this is the same question that search engines are asking and overall trying to achieve.

I suggest following Google, Bing, and Yahoo as an example, they have the same title for each page.  If I search "Christmas Decorations" in each of those engines, the title of the page is just "Christmas Decorations" no matter how deep I go. At the top of each page they let me know which page or which set of results I am looking at.

[ResslerMotors]

The best practice is to not let your internal search get indexed.  Use no-index,follow so that Google can follow any links on the page, but doesn't place your search pages in the index.

Reason why: duplicate content

How different is a search for "kitten" going to be from the search "kittens" on your website?  Probably pretty exact.  However, they will possess two different url's.  Ta-da!  Duplicate content!  Not to mention:

Kittie, Kitty, Kitties, cute kittens, funny kittens, black kittens, grey kittens, cat, cats, etc.

For proof, go check out your google analytics, how many different ways are people finding you right now?

[DanSerpico]

That's a really great point, and thanks for the "Good Answer!" I particularly appreciate the duplicate content point.

What are your thoughts on the matter if the popular searches are not created automatically (e.g., could include "kitten" and "kittens" because those are perfectly reasonable search terms) but rather created manually? They reflect broad categories to help provide search engines a way of seeing important items, along with a site map. (All this is, regretfully, in lieu of a proper browse functionality that will require a bit more work given how the data is structured). Search pages would then reflect only orthogonal terms, such as "kitten" "puppy" "guppy" etc, so they are (at least as far as I can think) unlikely to produce duplicate result pages.

That's only my uninformed, amateur strategy. What are your thoughts?

Thanks!

[KeriMorgret]

Google's Webmaster Guidelines at http://support.google.com/webmasters/bin/answer.py?answer=35769 states "Use robots.txt to prevent crawling of search results pages or other auto-generated pages that don't add much value for users coming from search engines."

So, they frown on having search result pages in the index in the first place.

[mjtaylor]

Gotcha! I say those pages should be excluded via robots.txt. Build static landing pages for any keyword term you wish to rank for and optimize those. Or rather than build a new page, take the top result of a search for keyword site:yourdomain.com, which will tell you what page already ranks highest for that target phrase, and and use that for your landing page for a term; optimize on page and build internal and external links to the page for the term.