SEO Masterclass Webit 2010
Download as PPTX, PDF49 likes11,977 views
Rand Fishkin's SEO Masterclass presents insights on optimizing websites for search ranks, emphasizing the importance of link diversity, content relevancy, and analytics. Key discussions include the correlation between domain names and ranking, the impact of Google Instant on keyword demand, and strategies for creating valuable links amidst social media's influence. The session highlights actionable methods for utilizing analytics to enhance SEO performance and the significance of on-page optimization beyond mere keyword usage.
SEO Masterclass Webit 2010
- 1. SEO MasterclassRand Fishkin, CEO & Co-founder, SEOmozWebit, Sofia October 2010
- 2. Topics for the Masterclass Correlation analysis of search results
- 3. Changes from Google Instant
- 4. Information architecture & navigation structure
- 5. Overcoming Twitter’s cannibalization of the link graph
- 6. Making Analytics Actionable
- 7. New Research: Topic Modeling in the Search ResultsUse Statistical Analysis to Answer Important SEO Questions
- 8. Correlation ≠ CausationThe more I wear suits, the more I speak on panels.Therefore: wearing suits causes me to speak on panels.
- 9. Understanding Correlation SignificanceNo CorrelationExact Match DomainPerfect CorrelationMost of our data for search rankings falls in this region(which we’d expect given algorithms w/ 200+ ranking factors)
- 10. Question #1:How to Best “Optimize” a Site for Search Engine Rankings
- 11. Methodology 11,351 SERPs via Google AdWords Suggest
- 12. 1st Page Only (usually ~10 results per page)
- 13. Correlations are w/ Higher Position on Page 1
- 14. Controlled for SERPs Where All (or None) of the Results Matched the MetricMethodologyLooking for elements that higher ranking pages have that lower ones do notNOT looking at raw counts of how many pages featured a given element
- 15. Contains All Query Terms in Domain NameExact Match Hyphenated DomainExact Match DomainHighest Stderr = 0.0241804
- 16. Our Interpretation Exact match domains remain powerful in both engines (anchor text could be a factor, too)
- 17. Hyphenated versions are less powerful, though more frequent in Bing (G: 271 vs. B: 890)
- 18. Just having keywords in the domain name has substantial positive correlationKWs in BodyKWs in Alt AttributeKWs in H1 TagKWs in URLKWs in TitleHighest Stderr = 0.00350211
- 19. Our Interpretation The Alt attribute of images is interesting
- 20. Putting KWs in URLs is likely a best practice
- 21. Everyone optimizes titles (G: 11,115 vs. B: 11,143). Differentiating here is hard.
- 22. (Simplistic) on-page optimization isn’t a huge factor.gov.edu.info.net.org.comHighest Stderr = 0.0269818
- 23. Our Interpretation More reasons to believe Google when they say .gov, .info and .edu are not special cased
- 24. The .org TLD extension is surprising – do they earn more links? Less spam? More non-commercial?
- 25. Don’t forget about branding/user behavior - .com is still probably a very good thing (at least own it)Content Length (tokes in body)URL Length(chars.)Domain Name Length (chars.)Highest Stderr = 0.0033353
- 26. Our Interpretation Shorter URLs are likely a good best practice (especially on Bing)
- 27. Long domains may not be ideal, but aren’t awful
- 28. Raw content length seems marginal in correlationQuestion #2:What Kind of Links Matter & How Should We Evaluate Links?
- 29. # of Linking Root Domains to URL# of Links to URLHighest Stderr = 0.00335677
- 30. Our Interpretation Links are likely still a major part of the algorithms
- 31. Bing may be slightly more naïve in their usage of link data than Google, but better than before
- 32. Diversity of link sources remains more important than raw link quantity# of Links w/ exact match anchor text# of linking root domains w/ exact match anchor textHighest Stderr = 0.00415058
- 33. Our Interpretation Many anchor text links from the same domain likely don’t add much value
- 34. Anchor text links from diverse domains, however, appears highly correlated
- 35. Bing and Google are relatively similar in evaluating these metricsCorrelation of Page-Level Link Valuation Metrics
- 36. Our InterpretationPageRank (and similar algorithms) are not particularly representative of rankings (but are somewhat correlated)
- 37. Linking domains are likely a better metric than raw links
- 38. Page Authority is reasonably good, but has a way to goCorrelation of Domain-Level Link Valuation Metrics
- 39. Our Interpretation No single domain valuation metric is especially well correlated with rankings
- 40. Rankings of individual pages may be more disparate we typical think re: “domain authority”
- 41. Overall, we’re still very naïve when it comes to understanding how links influence search rankingsQuestion #3:How Does Google Instant Change Keyword Demand / SEO?
- 42. Are Most Users Seeing/Using Google Instant?http://www.readwriteweb.com/archives/report_google_search_box_in_firefox_accounts_for_9.php
- 43. Methodology: Keyword Referral Search Data Look at keyword sending traffic via analytics
- 44. Distribute into groups by word-length
- 45. Analyze shifts in demand by keywords that brought visits to the site
- 46. Compare from period prior to Google Instant and directly afterVia MEC Manchester (UK)5 Sites, 4 Verticals, 10K+ Keywordshttp://www.mecmanchester.co.uk/blog/google-instant-data-after-12-days.html
- 47. Via Distilled Consulting (UK)11 Sites, Various Sizes (3.5K – 75K weekly visits), 75K+ Keywordshttp://www.distilled.co.uk/blog/seo/impact-of-google-instant/
- 48. Via ConductorMultiple sites, 880K visits, 10Ks of keywordshttp://blog.conductor.com/2010/09/what%E2%80%99s-been-the-impact-of-google-instant-on-searcher-behavior-so-far-not-much/
- 49. Interesting Takeaways Google Instant seems not to have shifted keyword demand by much (if at all)
- 50. Google “suggest” has been out for a long time already; users are likely accustomed to this feature
- 51. The “long tail” may get longer/shorter over time, but Instant seems less responsible than other factorsGoals of Successful Information Architecture
- 54. Maximize Usability of Navigation
- 55. Tips for Semantically Useful Navigation
- 56. Initially Design without Keyword Research
- 57. Add in Keyword Research Based Modifications
- 58. Validate Architecture/Path with Non-SEOs
- 59. Tips for Minimal Click-Depth
- 60. Imitate the Ideal Nav Pyramid
- 61. Broad Linking at Top Levels
- 62. Editorial Categorization > User-Defined
- 63. Editorial Categorization > User-Defined
- 64. HACK: Multi-Level HTML Sitemap
- 65. Tips for Usable Navigation
- 67. Naming Conventions that Match Intent
- 68. User & Usability Testing
- 69. Avoiding Common “Big Site” Problems
- 73. SEOmoz Web App
- 75. Employ Absolute URLsAbsolute:<a href=“http://www.seomoz.org/blog”> anchor </a>Relative: <a href=“../blog”> anchor </a>
- 76. C&D vs. Large, Credible Orgs that Scrape
- 77. Don’t Go Overboard w/ Bot Blocking
- 79. Track Referrals, not Site: Commands
- 80. Check Page “Types” that Don’t Receive Traffic
- 81. XML Sitemaps
- 83. RSS Feeds
- 85. “Search Results” in the SERPs
- 86. Create Category “Landing” Pages
- 87. Remove Obvious Traces of “Search” on Landing Pages
- 89. Bolster w/ UGC
- 90. Employ Scalable Content Production
- 91. Keep “Thin” Pages Out of the SERPs
- 93. Rel Canonical Can Help
- 94. Use AJAX to Reload Pages
- 95. Watch Out for Google Crawling Javascript
- 96. Offer Facets Only to Logged-In / Cookied UsersLogged-In = 345 / Googlebot = 141
- 97. Overcoming Twitter’s Cannibalization of the Link Graph
- 98. Way Back in 2007Interesting content, blog posts & linkbait earned LOTS of links
- 99. Fast Forward to 2010Not so many links (in comparison)
- 100. Fast Forward to 2010But tons of social sharing (and tweets)
- 101. Are Pages Linking Out Less?Via Linkscape’s web index
- 103. How Do We Earn Traditional Web Links (the kind search engines love)?
- 105. Infographics+1,085 links from 356 root domainshttp://royal.pingdom.com/2010/02/24/google-facts-and-figures-massive-infographic/
- 107. Value-Add Widgets
- 111. Citation-Worthy Explanationshttp://www.seomoz.org/blog/googles-algorithm-pretty-charts-math-stuff
- 113. Niches where content is low-supply/high-demandhttp://perfectmarket.com/vault_index_summer_2010_infographic
- 114. ID sites that already syndicate from someone else
- 115. Tactic #4:Stick to Niches w/o Twitter Adoption
- 116. Find sectors where traditional blogs/forums dominate conversationhttp://www.eng-tips.com/
- 117. Many of these “old-school” sites have followed external links (but don’t abuse these)Note the nofollow highlighting
- 118. Some “Web 2.0” Ones, Too Impressive domain and page level metrics
- 119. Tactic #5:Friends, Partners, Customers & Vendors
- 120. Friends + Family
- 121. Partners
- 122. Customershttp://www.seomoz.org/blog/headsmacking-tip-1-link-requests-in-order-confirmation-emails
- 124. Tactic #6:Twitter May Take Your Tweets, But They’ll Never Take Your Content!
- 125. Turn Your Tweets Into Contenthttp://twournal.com/home
- 126. Turn Tweets from industry leaders into content, too (this entices them to share)http://www.seomoz.org/blog/bing-vs-google-prominence-of-ranking-elements
- 127. Tactic #7:Can’t Beat ‘em? Join ‘em!
- 128. Twitter is (almost certainly) influencing (at least some) rankingsLots of tweets, virtually no links, but remarkable rankings
- 129. Twitter can send lots of direct traffic
- 130. You need to target the right Tweetershttp://twitaholic.com/top100/followers/
- 131. The Ones Who Send Real Traffichttp://mashable.com/2009/07/07/twitter-clickthrough-rate/
- 133. Not the “Grade”Use the “Rank”http://twittergrader.com
- 134. Takeaways # of Followers DEFINITELY isn’t everything
- 135. Tweeting heavily may not necessarily hurt the attention people pay to your tweets
- 136. Klout score probably isn’t great for predicting the reach of an individual’s tweets/messages
- 137. TwitterGrader Rank may be useful for predicting the traffic you’d get from a tweeter
- 138. Avg. CTR across 254 shared links = 1.17% (don’t feel bad if only 1/100 followers clicks a link)Making Analytics Actionable
- 139. To Make Analytics Actionable Always Ask:#1 - “Why am I measuring this?”#2 – “What would I do if results were different?”
- 140. # of Visits Per Search Engine Over Time
- 141. Action: Measure against search engine market shares & volume to determine whether you’re making positive strides
- 142. # Pages Getting Search Referrals Over TimeMeasure this number on a weekly/monthly basis
- 143. Action: Discover if indexation is an issue worth effortThis number sucks. Learn more about why at:www.seomoz.org/blog/indexation-for-seo-real-numbers-in-5-easy-steps
- 144. # of Keywords Sending Traffic from aSearch Engine over Time
- 145. Action: Determine if content additions are accretive and what drives growth/shrinkage in search trafficDid rankings fall? Or is demand down?
- 147. # of Visits per Keyword
- 148. Action: Analyze top traffic drivers from a value perspective, check rankings for potential easy wins & get answers if traffic dips“SEO Tools” is a big win and we could rank higher
- 149. First-Time vs. Returning Visits per KeywordThe keyword “SEO” leans toward first-time visits
- 150. Action: Determine value of reaching new visitors vs. converting branded users (focus efforts on the more valuable one)This metric speaks to business strategy about converting existing fans vs. reaching new customer segments
- 151. Distribution of Keyword Referrals
- 152. Action: Discover strengths vs. opportunities (60-70% of traffic is typically in the long tail and it converts better)
- 154. Action: Know if traffic spikes/dropoffs are from rankings, indexation or search demand shiftsRankings and Traffic both Dropped
- 155. Page Two RankingsReferrals from Page 2
- 156. Action: Identify low hanging fruit that can be optimized quicklyCould totally 301 this to www.opensiteexplorer.org
- 157. Engagement Analytics
- 158. Time on Site
- 159. Action: Compare to ROI metrics; if they correlate, improve on keywords/landing pages with low time on siteAverage “upgrade to PRO” visitor spent a whopping 44 minutes on SEOmoz!
- 160. # of Page Views
- 161. Action: Depending on your metrics, a “sweet spot” of pages browsed often dictates a conversion event – optimize towards itAverage “upgrade to PRO” visitor visits 12X the pages of an average visitor
- 162. Repeat Visit Ratio
- 163. Action: Find what content/activities/referrers send engaged traffic and copy those while improving subpar pages
- 164. Sharing/Linking Activity“Sharing Activity” Conversions
- 165. Action: Find patterns/sources that predict sharing activities (both content and CTAs) and make them testable conversion eventsGA allows you to set custom actions as “goals” then filter, monitor and improve on these metrics
- 168. Full Path Analysis
- 169. Initial Referrerwww.seomoz.org/blog/how-to-get-past-last-touch-attribution-with-google-analytics
- 170. ROI Analytics
- 172. Cost of Acquisition
- 173. Return on InvestmentNo. No. No.Yes. Yes. Yes.ROI = CLTV - CAC
- 174. Always Be Asking “What’s the ROI?”Get the ROI for every category (and subset)
- 175. New Research: Topic Modeling in the Search Results
- 181. Methodology: LDA (Latent Dirichlet Allocation) Build an LDA model based on the English language Wikipedia dataset (8mil+ pages)
- 182. Generate scores for top 10 rankings across several thousand search results
- 183. Look at correlation of search rankings with scores (in process)Simplified LDA FormulaChance of word is because of a topic= (Number of times the document already uses that topic a lot)X(Number of times that word has been in that topic)
- 184. Tool to Test it Outhttp://www.seomoz.org/labs/lda
- 185. Tool to Test it Outhttp://www.seomoz.org/labs/lda
- 186. Tool to Test it OutWe might need to work the “relevance” of our contenthttp://www.seomoz.org/labs/lda
- 187. Interesting Takeaways There may be more to “on-page” optimization then just using target keywords in the right places / ways
- 188. Search engines keep saying “make relevant content” – perhaps we can get more scientific and precise about what “relevant” means
- 189. Our LDA topic modeling work is still in its infancy. Expect more data, correlations, etc. in weeks to come.Q+ARand Fishkin, CEO & Co-Founder, SEOmoz Twitter: @randfish