SEO OPERATIONS MANUAL Part 2: REPAIR

Posted by khoiron h On Sabtu, 12 November 2011 0 komentar
Part 2: REPAIR

Overall Site Assessment

Search engines seem to care less about less about images, layout and site aesthetics. But the problem is, we also want to appeal to humans who actually visit these sites - not just Google's algorithm and robots.
So a general assessment of the site needs to include the basic and hopefully obvious tactics of good navigation, clear content attractive layout an up-to-date look that is appropriate for the niche. Almost all of this is objective opinion and mush of it should be provided by the client because of that.

However, the structure of the site will matter to search engines and that structure will effect site design. For example, sites need to be CSS based and have few or no tables within their structure.  This affects ranking AND visual layout. Further, page load speed is a huge factor in search engine ranking and also a major consideration for the designer. Sites designed with cascading style sheets will fare better in this area.
The bottom line is that older, passé sites need to be updated to match the current trends in structure and layout.  Some of these decisions are subject to the logistics of the code, others are objective in nature.

Fast Indexing (for new sites/domains)
New sites can be indexed in approximately 24 hours by submitting the domain to statistic monitoring sites.  Also a submission to Digg.com can start the process rather quickly.  Submitting to as many as possible is the safest way to insure indexing.
These sites can simply be visited with the domain in question appending the URL.  This can generate enough data to create the equivalent of a back-link.  Since the monitoring sites are indexed more often, the sites submitted to them are as well.
For  example, visit:  http://www.statbrain.com/www.yourdomain.com (replace with actual domain) to generate this data.
Here is a list of such services and the proper syntax:
http://www.statbrain.com/www.yourdomain.com
http://www.websiteoutlook.com/www.yourdomain.com



http://www.builtwith.com/?yourdomain.com
http://snapshot.compete.com/yourdomain.com
http://searchanalytics.compete.com/site_referrals/yourdomain.com
http://aboutus.org/yourdomain.com
http://quantcast.com/yourdomain.com
http://cubestat.com/www.yourdomain.com
http://alexa.com/siteinfo/yourdomain.com
http://alexa.com/data/details/?url=yourdomain.com
http://siteadvisor.cn/sites/yourdomain.com/summary
http://aboutdomain.org/backlinks/yourdomain.com


               
Site Code Checklist
The following is a checklist of most attributes that need to be checked for repair, update addition to or deletion from the site's code.
Title Tags
Search engines rely on spiders to crawl websites and index pages appropriately.  When a spider (crawler) lands on your website, the first thing they take notice of is your domain name and your website's title tag.
A title tag should include a description of what your website is about, rather than just including your website's URL, you should always incorporate your primary keyword phrase into the title tags of each webpage you own
You want to make sure that you use different title tags on every single web page, that way you are able to rank for different terms rather than just one.Title tag appears at the top bar of your web browser. Based on the keywords in this tag¸ search engines list the site relevant to a search made in their engines. It is one of the important aspects of getting rankings in search engines. Since this title also appears in search engines’ result pages¸ appropriate use of keywords in promotional language may result in more click-throughs.

H1 thru H6 Heading Tags
These tags are important to Google and  in the priority of their number. In other words, the H1 tag is looked at by Google after the Title tag as an important attribute for search results.  The H2 tag is looked at next, then H3 etc.  Obviously, these tags need to be keyword based or keyword rich.  Not all 6 are needed, but a minimum of H1 is strongly suggested.

Bold/Strong Tags
Google looks for these tags to find emphasized phrases on a site to see how it relates to the content.  With this attribute, an overuse could be detrimental.  But 1 or 2 appropriate strong or bold tags is recommended.
Basically, we never bold a page or a paragraph. Whether using b or strong it’s important ONLY to bold keywords or links with anchor text. 
Since everything on web is related to semantics, to increase chances of ranking better we should use <strong> rather <b> since the rendering in all browsers will be the same for both tags.  So, search engines may give more importance to <strong> over  <b>.
Some say an italicized keyword has some positive effect, but not as as much as bold/strong and therefore should also be used wisely (i.e. not overused).
Bold Tags = <b>   </b>
Strong Tags = <strong>  </strong>

Image File Names
A simple and  effective fix is to change image names to those that include or are  exclusively keywords.  Images called header.jpg can be easily converted to keyword_header.jpg or even keyword.jpg when appropriate.  (As long as the original file name is also changed to match.)
The Alt tags for each image should contain short descriptions that are keyword rich as well.  Null alt tags are considered an error by W3c validation tests.

Hyperlink/ Anchor Tags
Hyperlinks provide another keyword placement opportunity.  Of course, the associated file must have a matching file name.  Again, these need to be sensible to not appear as keyword "spam." 
The same is true for the anchor text associated with these links.  The anchor phrase "Click Here for More Info" can be changed to "Click Here for More Info About [keyword]" as long as it appears to be an appropriate use of the site's language.


Keyword Titled Pages
This technique is directly connected to the previous regarding hyperlinks and anchor tags. Obviously, a keyword hyperlink would need to link to an actual page of the same name (keyword based.)
But, technically, it would still make sense to create and name several  pages on a site with major keywords.  Even if there was little internal linking to those pages.  Google would still index these pages.  But not linking to them would be impractical and might ultimately look weird to Google if they study the site's structure. 
So these techniques work in tandem for good reason.


Home Link Canonicalization
We add Canonical Link Element in source code to avoid duplicate content issues. This tag or link element is added between head tags.  All internal links to the index page on the site need to be in the following format: 
http://www.domain.com/
Note the deliberate inclusion of the trailing slash.
Canonicalization also applies to other site pages.  Internal links should all be written in the same consistent format so that the search engine will recognize what an appropriate URL is for the site's navigation.
In other words, the link to "page 2" should not vary from page to page. It can't be "http://www.domain.com/page2.html/"  on one page, and "http://domain.com/page2.html" on another.  It needs to stay consistent site wide and best in this format:
"http://www.domain.com/page2.html/"            


Robots.txt
With the Robots Exclusion Protocol, we add or edit the robot.txt file to instruct search engines on what to index. The robots.txt allows us to hide files or directories we don’t wish the search engine spiders to find. 
 
Robots.txt syntax examples
 
To disallow the entire website:
User-agent: *
Disallow: /
 
To disallow Google from indexing the entire website:
User-agent: googlebot
Disallow: /
 
To disallow one specific file in a specific folder:
User-agent: *
Disallow: /folder/file.html
 
To disallow one specific folder and its contents AND any file or folder
that begins with the characters "folder":
User-agent: *
Disallow: /folder
 
To disallows all files beginning with subfolder in the folder directory :
User-agent: * 

Disallow: /folder/subfolder 
 
 
To disallow multiple files and directories:



User-agent: * 

Disallow: /folder/post- 

Disallow: /folder/posting 

Disallow: /folder/search 

Disallow: /forums/login 

Disallow: /forums/memberlist 





Additionally, search engines comply with the no follow attribute when used in the page code for specific links.
The rel="nofollow" attribute value was created for this purpose. This gives us more specific control: instead of telling search engines and bots not to follow ANY links on a page, it lets us easily instruct robots not to crawl a specific link.
For example:
 <a href="login.php" rel="nofollow">Log in Here</a>


Sitemap
Google likes sitemaps and can accept them in a number of formats, but they recommend creating a Sitemap based on the Sitemap protocol because the same file can be submitted to the other search engines, such as Bing and Yahoo!, that are members of sitemaps.org.
Here’s an example of a basic Sitemap with a single entry for a URL that includes an image and a video (for convenience, only a subset of available video information is shown).
<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
        xmlns:image="http://www.sitemaps.org/schemas/sitemap-image/1.1"
        xmlns:video="http://www.sitemaps.org/schemas/sitemap-video/1.1">
  <url> 
    <loc>http://www.example.com/foo.html</loc> 
    <image:image>
       <image:loc>http://example.com/image.jpg</image:loc> 
    </image:image>
    <video:video>     
      <video:content_loc>http://www.example.com/video123.flv</video:content_loc>
      <video:player_loc allow_embed="yes" autoplay="ap=1">http://www.example.com/videoplayer.swf?video=123</video:player_loc>
      <video:thumbnail_loc>http://www.example.com/thumbs/123.jpg</video:thumbnail_loc>
      <video:title>Grilling steaks for summer</video:title>  
      <video:description>Get perfectly done steaks every time</video:description>
    </video:video>
  </url>
</urlset>
Once we’ve created our Sitemap, we submit it to Google using Webmaster Tools. (Must have added the site to our Webmaster Tools account first.)
Google also accepts the following as Sitemaps:
RSS, mRSS, and Atom 1.0 Sitemaps
Google accepts RSS 2.0 and Atom 1.0 feeds. Most blog software creates a feed for us. Although that the feed may only provide information on recent URLs.
Text file Sitemap:
We can also provide Google with a simple text file that contains one URL per line. For example:
    http://www.example.com/file1.txt
    http://www.example.com/file2.txt
   
For best results, we need to ....
...specify URLs, as Google attempts to crawl them exactly as provided.
...use UTF-8 encoding.
...include nothing but the list of URLs.
...name the text file anything we want with the file a .txt extension (for instance, sitemap.txt).
Sitemap files should be submitted to Yahoo/Bing as well - according to their own guidelines.

Keyword Meta  Tags
Keyword meta tags are the most commonly known html attribute that has an apparent connection with SEO.  Unfortunately, this is one of the reason that Google doesn't pay as much attention to them.  Because they are easy to edit, they are easily abused.  rather than spend a lot of strategy on filling meta tags with keywords, most experts agree that it is more important to make sure this attribute has FEW values.  It would be more important to ELIMINATE a site's meta tag listing that looks too full than it would be to add a meta tag value to a site that has none.
But the consensus seems top be that it doesn't hurt to have approximately 5 very appropriate keywords in the meta tags listing, although it may help very little. And actually cause harm if over-used.
               
               
W3c Validation


The Markup Validation Service by the World Wide Web Consortium (W3C) allows us  to check HTML documents for conformance to HTML or XHTML and is also a quick method to check for errors in code. We validate sites as per W3C guidelines.
 
This is a good test to run after completing as much onsite SEO as possible as a "double-checking" process for other on page errors we may have missed or would not otherwise know about.
 
The validation can be run from here:
 
http://validator.w3.org/

Redirects
For sites that must redirect visitors to another site, there are strict requirements for appealing top Google.  Using NO redirection is best, but when that is not possible, it is important to follow these guidelines.
Because we have control of the mod_rewrite extension in teh Apache build of our servers, we can use it to dynamically change URL's using arguments on the fly - this is NOT a 301 redirect, but rather it's related behavior.
For example, if we wanted to redirect .htm files from an old server to their equivalent .php files on a new one using a 301 redirect, we would use a combination of mod_rewrite and the redirect directive to do the redirection + URL change.
We can do it on a file by file basis by making a really long list of possible redirects in the .htaccess file by hand without mod_rewrite, but that might  be a problem on a server with a lot of files, or a completely dynamic system. Therefore these 2 functions are often used together.
The syntax for the redirect directive is:
Redirect /yourdirectory http://www.newdomain.com/newdirectory
If the client requests http://myserver/service/foo.txt, it will be told to access http://www.yourdomain.com/service/foo.txt instead.
Note: Redirect directives take precedence over Alias and ScriptAlias directives, irrespective of their ordering in the configuration file. Also, URL-path must be a fully qualified URL, not a relative path, even when used with .htaccess files or inside of <Directory> sections.
If we use the redirect without the status argument, it will return a status code of 302 by default. This default behavior has given problems, so it's important to remember to use it, like this:
Redirect permanent /one http://www.newdomain.com/two
or
Redirect 301 /two http://www.newdomain.com/other
Both of which will return the 301 status code. If you wanted to return a 302 you could either not specify anything, or use "302" or "temp" as the status argument above.
You can also use 2 other directives - RedirectPermanent URL-path URL (returns a 301 and works the same as Redirect permanent /URL PathURL) and RedirectTemp URL-path URL (same, but for a 302 status).
For more global changes, we would use redirectMatch, with the same syntax:
RedirectMatch 301 ^(.*)$ http://www.newdomain.com
or
RedirectMatch permanent ^(.*)$ http://www.newdomain.com
These arguments will match any file requested at the old account, change the domain, and redirect it to the file of the same name at the new account.
You would use these directives in either the .htaccess file or the httpd file. It's most common to do it in the .htaccess file because it's the easiest and doesn't require a restart, but the httpd method has less overhead and works fine, as well.

This following scenario assumes we have a new domain (with no working pages under it) and want it to redirect properly to a main domain.
1. Ensure that we have 2 accounts - the old site and the new site (they do not have to be on different IP's or different machines).
2. The main (proper or canonical) site should be pointed at the new site using DNS. All other domains should be pointed at the old site using DNS. Parking them there is fine at this point.
3. Edit the .htaccess file at the root of your old account with this code:
Redirect 301 / http://www.newdomain.com/

Public WhoIs Record
This technique has little to do with SEO and more to do with good marketing - especially in the corporate world.  We want to make sure that a client's domain record shows an up-to-date and legitimate physical, business address in its public record.  Some potential customers will research this data  (because it is easy to do)  in an effort to verify the validity of a company's website.

               
hCard/vCard Physical Address Microformat
Microformats are a way of adding simple markup to human-readable data items such as events, contact details or locations, on web pages, so that the information in them can be extracted by software and indexed, searched for, saved, cross-referenced or combined.

Microformats are a way to use (X)HTML for data and a logical next step in the evolution of web design and information architecture and a way of thinking about data which combines html, content and presentation .

From the SEO perspective, the most commonly used microformat is the integration of Hcard for the contact address on the website contact page.

If we write the contact address of my company withOUTthe using the hCard the HTML will be as follows:
<p> BlackWire Marketing, LLC<br>
220 E. 11th Ave - Suite 1<br>
Eugene, OR 97401</p>
<p>541-343-3653</p>

<div id="" class="vcard">
 <a class="url fn n" href="http://www.blackwiremarketing.com/"> </a>
<div class="org">BlackWire Marketing, LLC</div>
 <a class="email"
href="mailto:support@blackwiremarketing.com">support@blackwiremarketing.com</a>
<div class="adr">
 <div class="street-address">220 E 11th Ave Suite 1</div>
<span class="locality">Eugene</span>,
<span class="region">OR</span>,
 <span class="postal-code">97401</span>
<span class="country-name">USA</span>
</div>
<div class="tel">541-343-3653</div>

The above code produces this result:
BlackWire Marketing, LLC
220 E 11th Ave Suite 1
Eugene , OR , 97401 USA
541-343-3653

Note:
publishing the email address will get it spidered by mail harvesters. Use that data with caution.


hcard code can be generated here:
Note: It is advisable to put only one address on a page.  But if we have to put more than one, then change the  div id="" class="vcard" tag accordingly for every address.


Google Analytics
We add Google Analytics to all sites for monitoring SEO and traffic progress and for future research.
               
Usability and Conversion Tools
Usability testing is the measurement of the quality of a user's experience when interacting with the website. If a visitor finds it difficult or frustrating to use the site, they will decide to exit.
We install a usability tool (JavaScript) on the site to track user behavior of visitors on the website. This helps track strengths & weaknesses as we record the visitor experience. 



One such tool is at: http://www.clixby.com
Clixby records visitor movements (mouse clicks, etc) on the site and stores it in a database for researching user on site experience.

Favicon
Adding a favicon to the site adds some credibility- but mostly to the end user.  aside from looking better, when bookmarked, it is easier to recognized on the user's side.  Title Tabs will look better and make the site more "brand- able".
http://www.favicon.cc   can be used to generate favicon files.

Site Content Checklist:
Keyword Density (ratio)
There is no formula for the ratio of keywords that should appear on a site when measured against non-keyword words (including code).  Some claim that 2% is low and that 18% is high.  There seems to be a lot of variation but usually within those ranges.
The main point is to use good, relevant content that is more readable for humans that for search engine spiders.  Obviously, content with zero keywords is senseless, but probably hard to do anyway if the content is going to be naturally relevant. i.e.: It would be hard to write an article about magic tricks without using the word 'magic" at least a few times.
On the other hand, using a major keyword too often and in a way that is less than pleasant to read would be a red flag of keyword overuse.

Keyword  Proximity
We pay attention to keyword proximity and prominence - as well as the order of the content itself.
The prominence of
the keyword is based on thefirst instance of where it appears within your content.  A keyword phrase that is used at the end of the content will be considered less relevant than a keyword phrase that appears in the first portion of the content or article.
This means that we need to ensure that we implement primary keyword phrases into the first half of our content, so that it is given more weight when search engine spiders index the website.

Keyword proximity is also very important in terms of search engine rankings because it indicates to the search engines that these keywords are related to one another.
For example, if we search for the keyword phrase 'dog training tips' we will pull up listings of websites that include the keyword phrase 'dog training tips'.
Without quotations, all websites that feature the keyword dog, training and tips seperately will also appear in the search results.Since we want to rank for specific phrases, rather than just individual keywords, we need to do our best to focus on close keyword proximity, so that search engine spiders crawling the website index entire phrases rather than singular keywords.

300 Word Plus Content
As a general rule of thumb, articles and content (especially blog posts) need to be at least 300 words.  But again, good quality content is more powerful that quantity.  But it is a good starting point.  In an extreme case, if we found a blog with a thousand articles all of which were only a couple of sentences, we would be suspicious.  We are not sure if Google would be or not depending on the content, but it could be a risk.  The 300 word rule is not set in stone, but can apply in some cases.

General Content "Rules"
More important that density ratios or word count is content quality.  Google loves dynamic (changing) content that is readable by humans (not just appealing to spider formulas).  They hate patterns that look contrived.  Content that is actually helpful and valuable will always fair better.
They despise duplicate content (content that appears on other sites the same as ours) and will punish it severely.  This is why pre-purchased articles require massive re-writes when used as website and blog content.
                               
               
Mobile Enabled  Site
Websites can be built with a mobile friendly alternate version. Browser checking code is used on the index page to detect the appropriate version. In the case of a mobile browser application in use, the redirect could go to a subdomain set up as http://mobile. yourdomain.com
Mobile enabled websites can be ranked by Google separately when Google see them as "mobile enabled" because they can be searched for separately.   Google classifies a website as mobile enabled based on certain signals like good page layout, the markups used, good semantic structure (h1 before h2, etc),  the right doctype is being set. correct encoding is used.  A Mobile Enabled site has no iframes, tables or popups. Mobile websites get a boost if certain quality metrics hold.