Search Engine Optimization Overview

Editor's Notes

• #2: Not a Ninja… but the areas I’m going to cover are brief intro to the world of SEO“Why am I not #1 when I type my name into Google?”
• #3: Defining a market – people searching for a specific productKeywords aggregate searchers into specific audiences – need to channel audience to the right content
• #4: Brand – company/product brand awareness, or building credibilityConversion – (conversion rates), selling products, email subscription, white paper downloads…
• #5: Own site – server records – referrals; GA – popular search termsOther sites – Google Adwords Keyword ToolRight terms: terms where your site ranks highest, but need to trade off amount of competition with level of traffic (subjective analysis)Paid tools: Google Adwords, SEO Consultants (he’s called Shimon BTW) etc
• #6: Could stop presentation! Good content will always be found, regardless of SEO
• #7: URLs – avoid parameterised ?id=12345 etc; use descriptive terms (keywords); hypens to join words (no spaces or underscores); use dates structures when appropriateMenu – puts current page into context, and allows search engines and users find related materialH1 for page title/heading only – ONE occurrence; use H2-H6 for section headings (do not apply CSS to body text to simulate headings). Turn off CSS for insight.If inserting PDFs or images, use a meaningful URI, eg, man-relaxing-beach-sun.jpgIt’s a webpage, not a document, so link stuff: PAGE RANK – Google Toolbar. Ranks revised 3-4 times/year. Link to pages with high RANK, then you get a bit of their juice… Even more juice if high ranking pages link to you.Tag all content correctly (ie, ALT, TITLE tags etc). Good for accessibility; great for SEOOh – might say this a few times – WRITE INTERESTING CONTENT… lay it out in visually appealing manner to attract users in the BLINK moment (Blink – Malcolm Gladwell). Google caches every 5-6 days - keep it up to date as your ranking will slip over time if not updated.
• #8: KEYWORDS – largely ignored by search engines due to abuse in 90sDESCRIPTION – used as text summary in SE listingCOPYRIGHT – should be on every pageGEO – add positional info (now appears on Google listings). ICBM = sad geek humour (drop bomb here)ROBOTS – tells search engines whether to index page or not, or tell them to ignore embedded URLs (useful if page includes public comments)CANONICAL URL – if content can be accessed by duplicated URLs (esp on parameterised URLs, or sites that use REWRITE rules), this lets the Search Engine now definitive reference and stops you being penalised for duplicate contentTITLE tag – place in HEAD use keywords – can be same as first H1… but most important/unique keywords first (ie, priority LEFT to RIGHT)REDIRECT – if a page’s URL moves and you cannot configure a server-side redirect, there are various methods to auto forward users to the new page. At very least put a link to new page.
• #9: Dublin Core: 15 elements Title Creator SubjectDescriptionPublisherContributorDateTypeFormatIdentifierSourceLanguageRelation CoverageRightsRDF: Food for Robots. Essentially embedding XML type descriptors around information to define relationships and meaning. Types of RDF = FOAF and RSS. Really useful is linked to external taxonomies , eg, Thompson-Reuters Open Calais for tagging with named entities (people, organizations, geographies, books, albums etc), facts and events. Web 3.0 stuff.(Triples: subject-predicate-object expressions: “The sky has the color blue”: a subject denoting "the sky", a predicate denoting "has the color", and an object denoting "blue".)PICS: AOL usersShort URLs (W3C approved?) – TinyURL was banned! Bit.lyGna.rl Get with Twitter generation. May die a death soon
• #10: Robots.txt – food for robots – tells search engine crawlers what content to look at, what to ignore and how often to crawl. Google will drop you site ranking (or even ignore) without such a file in the ROOT directorySitemaps – XML file useful for search engines to work out the information architecture of your site; text sitemaps can be useful to your human visitors tooPage Load – not only frustrating for human viewers, Google (as of March 2010) now factors page load time into Rank calculation. Lots of techniques for improving HTTP requests – Yslow is the only real tool for this.Redirects – assume old URLs never die. Server side tweaks – HTTP 301 redirect for permanent changes; 307 for temporary redirects. Helps users, and informs search engines so you don’t get penalized for duplicate contentError pages – make a custom error page for 404 errors (page no longer exists) with something helpful – will keep users on your site longer; same for (403) access deniedGoogle Analytics – Uses Javascript on each page to track all access into, within and out of your site. Great tool for site analysis – but Google now knows exactly how popular your site really is(… page ranking again!). Can analyze site server log (doesn’t need JS) but data not as rich
• #11: Google Account – passport top all their various tools. Toolbar gives page rankingGoogle Analytics – track your users – you’ll need to upload a verification file or embed a code in your site. Wait at least a few days to get some useful info (see other CWP Presentation)Google Webmaster Tools: List top search queries, keywords, crawler stats, 404 errors, broken links, malware check, site performance…Google Adwords: Type in keywords and see stats across whole web for related terms (competition, monthly traffic country/global)SEO Plugin has some useful features, but for ucla.edusubdomains (MOST OF US), most analysis is for the overall domain of ucla.eduW3C/508 validators… Firefox Web Developer Toolbar probably the most useful tool