Weekly SEO Quiz Question

[dyn4mik3]

I posted this before in the thread about Rand's SEO Quiz but I was hoping to get some more response and answers, so I'm bring it up again.

Inspired by Rand’s SEOmoz Quiz, I started a new series over at my blog called SEO Quiz. The goal is to ask some practical questions about SEO and generate some good discussion. These questions will deal with preferred methods for optimizing (on-page and off-page) - similar to Rand's title question.

Hopefully the questions and answers will become useful information for SEOs.

Every week I'll post a new question and at the end of the week I'll pull out the best comments and explain the answers.

I encourage everyone to drop by and give their answers. Add one or two lines about why you think your answer is correct.

SEO Quiz (http://www.socialpatterns.com/category/seo-quiz/)

This week's question deals with proper URLs.

Q: When linking to the homepage from an internal page, which of the following URLs is the best choice?

* A: /
* B: http://www.mydomain.com
* C: http://www.mydomain.com/
* D: http://www.mydomain.com/index.html

Either answer here or answer on my blog (I'll try and recopy answers here over at the blog so there is a comprehensive list) - I'll have the final summary/answer over at my blog and repost it here.

(Question for mods - is this the proper place to do this? Or is this breaking the TOS)

[dyn4mik3]

Well this week's quiz went well. Not as many responses as I'd like but good enough for a start.

Here's the repost from:
SEO Quiz - URLs: Conclusion (http://www.socialpatterns.com/seo-quiz/seo-quiz-urls-conclusion/)

What seemed like a simple and straightforward question, gradually generated an informative discussion. So far, the quiz looks like a success - with quite a number of people choosing different answers.

So what did we learn from SEO Quiz - Part 1: URLs? I’ll start off by pulling out some comments and summarizing what people thought were the best practices. I’ll finish off with my own explanation.

Most people said B or C was acceptable, having no real preference for either.


randfish:

My vote is for B or C. I don’t think it would make too much of a difference between the two.

Out of all the choices, most agreed that C was the best URL choice. Some noted they had learned of the correct URL structure in the past, but forgot why.

hink:

This is one of those things you learn, you have no idea where you learned it, and you are glad you happened to learn the right way.

There are a couple reasons why C is the better choice.

PhilC:

B is not a good choice because Apache servers need the trailing slash, and the way they get round it is to ask the requestor to request again but with the trailing slash added - a redirect to the proper URL. It may not be universal among Apache servers, and it may even be out of date, but as long as some servers do it that way, B isn’t the best choice.

It’s interesting that Google’s brand new Sitemap system needs the trailing slash in the URLs, so the redirect may be very common.

I’ll go for C on its own.

Lets take a look at http://www.mydomain.com

The browser communicates to the server, requesting ‘mydomain.com’. First the server looks for a file called ‘mydomain.com’ and can’t find anything. It then adds a trailing slash and tries to find a directory called ‘mydomain.com’ - bingo. Then the server returns the default file for the directory.

So if a browser requests http://www.mydomain.com instead of http://www.mydomain.com/, the visitor is actually being redirected to http://www.mydomain.com/.

Answer B forces the web server to work harder than it needs. By giving the server a hint as to what it needs to look for, you are saving your visitors the tiny wait.

Answer D is not a good choice because there is no room for expandability. Suppose you want to change your web site technology, lets say from pure html to php. You would need to update all your URLs in order to make the change. With answer C, you wouldn’t have to.

Another reason you want to use C is that you may not want to expose your web site’s underlying technology. By using a trailing slash directory structure, visitors will have a harder time figuring out what programming language your web site uses.

Answer A leaves the URL structure up to the server configuration. When Google goes to index the link, depending on the server configuration Google may index www.mydomain.com or mydomain.com - it is much easier and clearer to use absolute links.

A while back, GoogleGuy had this to say about relative URLs:

99.999% of the time a bot will be able to disambiguate relative urls, for example, but why leave that little bit of chance? I’m not saying that people should always use absolute urls–just that the simpler you make things for bots, the less likely you are to run into problems.

Summary

Always use a trailing slash URL for directory structures (including homepages). Why?

* Saves the web server a redirect
* Saves visitors some time
* Makes it easy for you and the crawler

[randfish]

Great work - You've cleared up an issue and put it neatly to bed, all with a quiz! Very nice.

[dyn4mik3]

Thanks Rand!

I'll be posting a new question tonight, so keep watch.

[dyn4mik3]

Same deal as last week - I encourage everyone to drop by and leave their answer here or on my site. This question may turn out to be less SEO specific, but it is an important issue and I'd love to get some expert opinions on it.

Drop by and leave your answer!

SEO Quiz - Part 2: Title Separators (http://www.socialpatterns.com/seo-quiz/seo-quiz-part-2-title-seperators/)

Q: Which separator character is best for a web page title?

* A: Keyword Phrase - Site Name
* B: Keyword Phrase | Site Name
* C: Keyword Phrase » Site Name
* D: Keyword Phrase > Site Name

[dyn4mik3]

This time around I asked you what separator you would use in your titles (http://www.socialpatterns.com/seo-quiz/seo-quiz-part-2-title-seperators/). Before doing research for this question, I never really gave a second thought about title separators. I would use either a hyphen or the pipe character and be done with it. After reading over the quiz’s comments, it’s obvious that many of you felt the same.

Who cares about title separators if it doesn’t influence our search engine marketing campaign?


Rand started off by explaining why he likes the pipe character.

Rand:

I like B the best, as I feel it makes a distinction between the two phrases. It’s the only character that looks appropriate to me when viewing two separate pieces of information about a site’s title.

Most people picked either A or B, but interestingly everyone had strong opinions as to their favorite.

Nick Fitzsimons:

Thanks for helping me to realize how opinionated I am about such apparently insignificant details!

Natasha Robinson:

I once spent the better part of an afternoon researching the use of “&” in titles… so don’t feel so alone - lol

But how exactly do the search engines feel about different title separators? Cary was the first to make a guess.

Cary:

I guess “A” because the other choices aren’t typical English grammar, and I’m guessing that someone like Google is looking for titles that make grammatical sense.

Not a bad guess, but Dr. Garcia dropped by and explained to us none of the title separators matter after tokenization. (As most of us expected, including me).

Orion:

Thus, from the relevancy and ranking standpoint, none of the candidate delimiters described in the quiz really make a difference.

So if I already knew that title separators would be filtered out after tokenization, why bring up the question at all? Because as search engine marketers, we ultimately are marketing our sites to humans. Obviously rankings are very important, but we also need to take the time and consider how our design choices affect the ultimate experience that we are conveying.

From an accessibility standpoint, answer C is horrible. In a screen reader, » is read as “right double angle bracket (http://www.standards-schmandards.com/index.php?2004/11/06/6-the-sound-of-the-accessible-title-tag-separator)“. Compare that to “dash” or “vertical bar”.

Answer D brings up a couple of potential issues. Bad coding can lead to very broken code when > is introduced. Tags can inadvertently be closed if there is incorrect html encoding. Sometimes > is escaped by the search engines and > is replaced by >

So that brings us to either A or B. Both are used very widely, but I’d like to make an argument for A over B. This is due to book marking in IE 6. In IE6, whenever someone bookmarks a page in favorites, the vertical pipe is not rendered in the favorites list. It just renders as a space. This can lead to confusing labels in your favorites.

Ultimately, title separators have very little impact on search engine optimization efforts. But these little separators could make an impression on your visitors and need to be considered when crafting titles.

Aside: Dr. Garcia brought up an important issue when searching for titles with separators. Often times separators mean different things to a search engine inside a query. A hyphen works as a localized EXACT operation, while a vertical bar works as an OR operation.