A genuine Silicon Valley Web Flyer | One line of code at a time...

Learn SEO from a Google webmeister

Google is the first place many people go first when opening a web browser. It’s fast, reliable, and accurate. Search engine optimization could be the special sauce that makes or breaks your website. So let’s learn a little SEO from Google. Maile Ohye will be our driver.

Crawling

When you create a website, make it accessible without forms. Most often, search engine crawlers don’t index dropdowns.

Cookies

Content available without cookies

Most search engines don’t accept cookies. Allow non-cookie guest accounts to view basics on your site. Avoid the message, “You must enable cookies to use this site”.

Use static HTML links and textual based content. When you are making a search-friendly site, an all Flash interface won’t work.

Maile Ohye, Google Webmeister

Crawlable architecture

Consider progressive enhancement. Site structure, links, navigation in static HTML. It improves JS, forms, cookies, and Flash issues. It also reduces dilution of PageRank when sharing links between Flash and non-Flash version.

Add fancy bonuses like Ajax and Flash later. Link with descriptive anchor text. The most common anchor text is “click here”.

Guidelines for URLs

  • Organized
  • Sharable between users (each item referenceable among friends)
  • No orphan pages
  • Largely unique content per URL. Avoid serving different languages on the same URL

YouTube – one page per “thing”. Site navigation is in HTML; sepearate from rich media. Descriptive content along side rich media.

Consider sIFR for Flash. JavaScript can detect is Flash is installed, and show whichever version you want. The text matches content viewed by enabled users.

Ajax

Consider using Hijax. Format JS with a static URL as well as a JS function.

<a href="ajax.htm?foo=32" onClick="navigate('ajax.html#foo=32'); return false">Click here</a>

Images

Add descriptive alt-text. Use quality images. Include descriptive textual content near the image.

Nice:

<img src="malie-eating-bbq.jpg" alt="Malie eating BBQ chicken" />

Less nice:

<img src="image.jpg" />

Better to avoid:

<img src="DSC001.JPG" alt="Viagara, World of Warcraft, pigs" />

Google Webmaster Central

There is a lot of reference here, discussion groups, blogs, and tips. You can review crawl errors in webmaster tools. You can correct your crawl mishaps. Use Xenu for detecting broken links on your site.

Be sure that 500 errors are expected or normal. Correct URLs restricted by robots.txt, if necessary. There is a robots.txt generator and analyzer on the site.

Utilize web server options. Enable “if-modified-since” response header. Use gzip for utilizing compression.

Blog post: First date with the Googlebot

Eliminate soft 404s

Eliminate soft 404s

Correct your webserver if returning 200 rather than 404 (then often directs to homepage).

Drawbacks: it confuses users and causes duplicate content for search engines. Crawlers may not discover new pages or find modified pages as quickly.

Indexing

Serve Helpful 404s

Set the preferred domain. Influence canonical URLs, use 301s properly.

Dilution of PageRank with duplicate content

You want to have one version of your website running. Set preferred domain to www or non-www. 301 redirect one version to the other. PageRank will dilute your value if you have both running at once.

Reduce duplicate content

Do not duplicate content through URL parameters. This is common with product pages, sort order, and affiliate IDs. Keep URLs as clean as possible. Remove unnecessary parameters. Track visitor information with 301s to redirect URLs with paramaters to the canonical version. Use cookie data to set variables. Keep the site accessible without cookies though.

How else can you influence the canonical? For Google, Sitemaps can influence our understanding of the canonical URL. XML Sitemaps are great too (sitemaps.org)

Use Google Webmaster Tools to make your sitemap. If you have video content, you can submit a video sitemap. News, mobile, code search, etc.

If you use a sitemap, you get a lot more stat tracking.

Proper use of response codes

It signals search engine to transfer properties, like link popularity, to the target URL. If you have to modify a URL, use a 301.

“301s are your friends.” -Maile Ohye

Blog post: Best practices when moving your site

Anatomy of a search result

Anatomy of a search result

Title, snippet, URL, and sitelinks.

Titles are very informative. They act as an informative signal of the URL’s contents to search engines. “Untitled” is the most common title on the web. Not exactly informative. Webmaster Tools will help you out with this.

Help your snippets

Help your snippets. Snippets provide the user more context for each result. QUality of your snippet can impact click-through. Influence your snippet with a meta description. They can easily be utilized by Google in search results.

<meta name="description" content="This is what would show up as a snippet</meta>

Blog post: Improve snippets with a meta description makeover

Search within search

“Search within a site” is an experimental project right now.

Quick tips

Consider eliminating index.html or index.htm. Be consistent.

If you have a verified sub-domiain, you can use webmaster tools to target IPs by geography.

XML files are not indexed.

Official Google Webmaster Guidelines

  • What a great analogy! It's perfect - building a relationship with the search engines and therefore with the searchers. More and more it's all about relationship building, not just work.
  • socialmaker
    Online promotion with the aid of online marketing and seo are good ways to promote yourself and your business on the online environment. When it comes to seo services there are only a handful of companies who really know what they're doing. Only these know what content is all about, fixing site issues, URL canoniclization and rewrites and getting you a lot of links. If you want SEO for your site i recommend you search for a good company. Don't let a noname mess up your site !
  • I may be slightly biased, but this is the neatest recap ever. :)

    One thing: while Google doesn't often return xml in search results, they are indexed an can appear. [filetype:xml yourqueryterm] may surface them.

    Related info on RSS/Atom feeds removed from search results:
    http://googlewebmastercentral.blogspot.com/2007...

    Andrew, thanks for attending google i/o and hope to ttys!
  • me
    The Google Webmaster Tools blog is probably the best reference for SEO stuff.
  • looks pretty good, i'm impressed
  • Great recap post. Even though I know all this I love your presentation. You rock man...
  • That was actually an awesome read, I learned a lot. Mostly stuff I knew before but it reinforced all that.
blog comments powered by Disqus