Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Understanding SEO

Understanding SEO

nefarioustim

March 26, 2012
Tweet

More Decks by nefarioustim

Other Decks in Technology

Transcript

  1. Understanding  SEO   or   How  to  Help  Users  Find

      Your  Stuffs  on  the  Intertronz  
  2. Qualifica:on   •  Web  developer  for  15   years  

    •  Client-­‐side  specialist  for   6  years    
  3. Qualifica:on   •  Web  developer  for  15   years  

    •  Client-­‐side  specialist  for   6  years   •  Informa:on  architect    
  4. Qualifica:on   •  Web  developer  for  15   years  

    •  Client-­‐side  specialist  for   6  years   •  Informa:on  architect   •  Massive  Nerd   –  Quite  literally   –  SEO  as  a  hobby    
  5. Results   •  Google  “Web  Development  Lead”   – Top  result

      •  Google  “Web  Development  Manager”   – First  page  
  6. Results   •  Google  “Web  Development  Lead”   – Top  result

      •  Google  “Web  Development  Manager”   – First  page   •  Google  “Object  oriented  JavaScript”   – Top  result  
  7. Results   •  Google  “Web  Development  Lead”   – Top  result

      •  Google  “Web  Development  Manager”   – First  page   •  Google  “Object  oriented  JavaScript”   – Top  result   •  Google  “Tim”   – Third  result  (above  the  inventor  of  the  web)  
  8. Search  Engines   •  Google   •  Microso]  Bing  

    – Now  own  Yahoo!  Search   •  Many,  many  more…  
  9. Tree   Home   Products   Wotsits   Oojamaflips  

    Dinguses   About  Us   Contact   FAQ  
  10. How  search  engines  work   •  Army  of  robots  

    – Automated  spiders   •  Work  page  by  page  
  11. How  search  engines  work   •  Army  of  robots  

    – Automated  spiders   •  Work  page  by  page   •  Cache  and  analyse  content  
  12. How  search  engines  work   •  Army  of  robots  

    – Automated  spiders   •  Work  page  by  page   •  Cache  and  analyse  content   •  Index  keywords  
  13. How  search  engines  work   •  Army  of  robots  

    – Automated  spiders   •  Work  page  by  page   •  Cache  and  analyse  content   •  Index  keywords   •  Assign  an  arbitrary  rank  
  14. What  is  a  keyword?   •  A  word  or  phrase

      •  Contains  high  relevance  to  subject  
  15. Examples   •  Sports   •  Sport   •  Bejng

      •  Gambling   •  Bet   •  Gamble   •  Sports  Bejng   •  Sport  Bejng   •  Sport  Bet   •  Sports  Exchange  Bejng  
  16. But  wait…   •  World  Cup  2014   •  Wimbledon

      •  Bri:sh  Grand  Prix   •  Cheltenham   •  Spring  racing  
  17. Why  so  important?   •  Search  engines  store  references  to

     URIs  in   keyword  specific  indices  
  18. Why  so  important?   •  Search  engines  store  references  to

     URIs  in   keyword  specific  indices   •  Vast  array  of  smaller  databases  
  19. Why  so  important?   •  Search  engines  store  references  to

     URIs  in   keyword  specific  indices   •  Vast  array  of  smaller  databases   •  Data  denormalisa:on  
  20. Why  so  important?   •  Search  engines  store  references  to

     URIs  in   keyword  specific  indices   •  Vast  array  of  smaller  databases   •  Data  denormalisa:on   •  Makes  searching  a  whole  bunch  faster  
  21. Choose  wisely     •  Keyword  compe::on   – Low  compe::on

     can  mean  low  relevancy   – “Long  tail”  
  22. Choose  wisely     •  Keyword  compe::on   – Low  compe::on

     can  mean  low  relevancy   – “Long  tail”   •  Level  of  use   – More  commonly  used  search  terms  have  higher   compe::on  
  23. Choose  wisely     •  Keyword  compe::on   – Low  compe::on

     can  mean  low  relevancy   – “Long  tail”   •  Level  of  use   – More  commonly  used  search  terms  have  higher   compe::on   •  Keyword  tools  
  24. Where  you’ll  want  to  use  them   •  URIs  

    •  Link  text   •  Page  :tles  
  25. Where  you’ll  want  to  use  them   •  URIs  

    •  Link  text   •  Page  :tles   •  Headings  
  26. Where  you’ll  want  to  use  them   •  URIs  

    •  Link  text   •  Page  :tles   •  Headings   •  Emphasised  text  
  27. Where  you’ll  want  to  use  them   •  URIs  

    •  Link  text   •  Page  :tles   •  Headings   •  Emphasised  text   •  Within  your  content  
  28. Keyword  Targe:ng   •  Never  mind  keyword  density…   • 

    Think  about  keyword  relevancy   •  Density  no  rela:on  to   – Quality   – Seman:cs   – Content  relevancy  
  29. LINKS   The  internet  is  “a  series  of  tubes…”  (c.

     Ted  Stevens,  Alaskan  Senator  1968-­‐2009)  
  30. Links  win   •  Links  trump  everything   •  Link

     quan:ty   •  Link  quality   •  Largest  single  effect  on  rankings  
  31. Link  Juice   •  PageRank   •  Based  on  links

     into  and  out  of  the  page   •  Used  to  determine  page  quality   •  Really  is  a  series  of  tubes  (and  sluice  gates)  
  32. The  juice  will  flow!   England  Team   Football  

    Interna:onals   World  Cup   Euro  2012  
  33. rel=“nofollow”   •  Tell  robots  to  alter  the  flow  of

     link  juice   •  Useful  for  user-­‐generated  content  
  34. rel=“nofollow”   •  Tell  robots  to  alter  the  flow  of

     link  juice   •  Useful  for  user-­‐generated  content   •  Designed  to  prevent  link  spam  
  35. rel=“nofollow”   •  Tell  robots  to  alter  the  flow  of

     link  juice   •  Useful  for  user-­‐generated  content   •  Designed  to  prevent  link  spam   •  Use  for  “sculp:ng”  is  fu:le  
  36. Improving  “crawlability”   •  Flat  site  architecture   •  Short

     link  depth  to  pages   •  Recommend  max  100  links  per  page   – Misleading  value  
  37. Improving  “crawlability”   •  Flat  site  architecture   •  Short

     link  depth  to  pages   •  Recommend  max  100  links  per  page   – Misleading  value   •  Easily  maintain  250-­‐300  links  per  page   – Next  level  approx.  200  links  and  so  on  
  38. Link  building   •  The  art  of  obtaining  links  back

     to  content   – High  quality  from  high  relevance  
  39. Link  building   •  The  art  of  obtaining  links  back

     to  content   – High  quality  from  high  relevance   •  Fine  line  between  link  building  and  link  spam  
  40. Link  building   •  The  art  of  obtaining  links  back

     to  content   – High  quality  from  high  relevance   •  Fine  line  between  link  building  and  link  spam   •  Full  :me  job  
  41. Link  building   •  The  art  of  obtaining  links  back

     to  content   – High  quality  from  high  relevance   •  Fine  line  between  link  building  and  link  spam   •  Full  :me  job   •  Link  sharing  
  42. Link  building   •  The  art  of  obtaining  links  back

     to  content   – High  quality  from  high  relevance   •  Fine  line  between  link  building  and  link  spam   •  Full  :me  job   •  Link  sharing   •  Social  networks  
  43. Third-­‐party  links   •  No  direct  value   •  Promotes

     link  sharing   •  New  wave  of  Social  Media  Experts  
  44. Third-­‐party  links   •  No  direct  value   •  Promotes

     link  sharing   •  New  wave  of  Social  Media  Experts   •  Exercise  in  marke:ng  
  45. Link  bait   •  High  likelihood  of  reposts   • 

    Viral   •  Usually  created  with  intent  
  46. Link  bait   •  High  likelihood  of  reposts   • 

    Viral   •  Usually  created  with  intent   •  One  man’s  link  bait  is  another  man’s  quality   content  
  47. URI  Design   •  Highly  important   •  Embed  keywords

     in  domain   •  Embed  keywords  in  path  
  48. URI  Design   •  Highly  important   •  Embed  keywords

     in  domain   •  Embed  keywords  in  path   •  TLDs  are  important  
  49. URI  Design   •  Highly  important   •  Embed  keywords

     in  domain   •  Embed  keywords  in  path   •  TLDs  are  important   •  Order  is  important  
  50. URI  Design   •  Think  about  archiving   •  Try

     and  avoid  “magic  numbers”  
  51. URI  Design   •  Think  about  archiving   •  Try

     and  avoid  “magic  numbers”   •  Use  hyphen  as  space  separator   – Google  doesn’t  recognise  underscores  
  52. URI  Depth   •  Try  to  keep  URIs  shallow  

    •  Subdomains  similar  to  first  level   – hsp://soccer.beuair.com/   – hsp://beuair.com/soccer/  
  53. Branches  and  leaves   •  Branch  URIs  end  with  a

     slash   – hsp://beuair.com/football/  
  54. Branches  and  leaves   •  Branch  URIs  end  with  a

     slash   – hsp://beuair.com/football/   •  Leaf  URIs  don’t   – hsp://beuair.com/football/spurs.html   – hsp://beuair.com/football/the-­‐yid  
  55. Branches  and  leaves   •  Branch  URIs  end  with  a

     slash   – hsp://beuair.com/football/   •  Leaf  URIs  don’t   – hsp://beuair.com/football/spurs.html   – hsp://beuair.com/football/the-­‐yid   •  Users  read  URIs  in  search  lis:ngs  
  56. Branches  and  leaves   •  Branch  URIs  end  with  a

     slash   – hsp://beuair.com/football/   •  Leaf  URIs  don’t   – hsp://beuair.com/football/spurs.html   – hsp://beuair.com/football/the-­‐yid   •  Users  read  URIs  in  search  lis:ngs   •  Sets  up  an  expecta:on  
  57. Page  vs.  Site   •  Bots  have  a  vague  understanding

     of  site   •  Based  on  URIs  and  link  map  
  58. Page  vs.  Site   •  Bots  have  a  vague  understanding

     of  site   •  Based  on  URIs  and  link  map   •  Domain-­‐level  ranking  factors  
  59. Page  vs.  Site   •  Bots  have  a  vague  understanding

     of  site   •  Based  on  URIs  and  link  map   •  Domain-­‐level  ranking  factors   •  Page-­‐level  ranking  factors  
  60. Page  vs.  Site   •  Bots  have  a  vague  understanding

     of  site   •  Based  on  URIs  and  link  map   •  Domain-­‐level  ranking  factors   •  Page-­‐level  ranking  factors   •  Construct  as  pages,  link  as  site  
  61. Duplica:on   •  Duplica:ng  content  on  two  URIs  is  bad

      •  Serving  the  same  page  on  different  URIs  is   duplica:on  
  62. Duplica:on   •  Duplica:ng  content  on  two  URIs  is  bad

      •  Serving  the  same  page  on  different  URIs  is   duplica:on   •  Try  and  avoid  syndicated  content  
  63. Dynamic  and  sta:c  content   •  Dynamic  content   – Blog

     posts   – News  stories   – Markets   •  Sta:c  content   – About  us   – Help   – Bejng  rules   •  Good  to  have  a  balanced  mix  
  64. Titles   •  Page  :tle   –  <:tle>Football  –  Beuair</:tle>

      •  H1   –  <h1>Football</h1>   •  Titles  should  repeat  in:   –  URI   –  Page  :tle   –  H1   –  Throughout  content   –  Links  to  the  page   •  This  one  is  important!  
  65. Meta  elements   •  Keywords   – Not  used  by  Google

      – Occasionally  used  by  older  search  engines   •  Descrip:on   – Not  used  for  rankings   – Important  for  users   •  noindex,  nofollow  
  66. Per:nent  mark-­‐up   •  Links!   – <a  href=“…”>Football</a>   – <link

     rel=“index”  href=“…”>   •  Emphasis   – <strong>,  <em>   – <b>   •  Headings    
  67. Link  element   •  Canonical:   – hsp://sportsnews.com/football/   – hsp://sportsnews.com/football/?ads=false  

    •  More  subtle:   – …/teams/tosenham-­‐hotspur/peter-­‐crouch   – …/teams/england/peter-­‐crouch   •  Others   – next,  prev,  first,  last,  index…  
  68. Google  is  Skynet   •  Knowledge-­‐based  system   •  Algorithms

     update  regularly   •  Features  added   •  Try  to  understand  trends  in  content  
  69. Breadcrumbs   •  Google  looks  for  them   •  Tries

     to  understand  naviga:on  system   •  Improves  crawlability   •  Helps  the  user  
  70. Rich  snippets   •  Microformats,  microdata,  RDFa   – Reviews  

    – People   – Products   – Businesses  and  organisa:ons   – Recipes   – Events  
  71. UI  Paradigms   •  S:ll  learning   •  Common  paserns

      – Tab  widgets   – Accordion  widgets   – Carousel  widgets   •  Widgets   – Common   – jQueryUI   – YUI  
  72. JavaScript   •  Headless  browser   •  Support  for  JavaScript

     is  limited   •  Evaluates  some  JavaScript   – Results  vary  
  73. JavaScript   •  Headless  browser   •  Support  for  JavaScript

     is  limited   •  Evaluates  some  JavaScript   – Results  vary   •  Trawls  embedded  JavaScript  for  links  
  74. Using  HTTP   •  Correct  status  codes   – 200  OK

      – 301  Moved  Permanently   – 302  Moved  Temporarily   – 404  Not  Found   – 410  Gone   – 500  Errors  
  75. robots.txt   •  hsp://www.robotstxt.org/   •  Automa:cally  looked  for  

    •  Stored  at  the  root   •  No  robots.txt  is  a  free  pass   •  Use  to  guide  bots,  not  control  
  76. Sitemap  Protocol   •  hsp://www.sitemaps.org/   •  Aids  discovery  of

     content   •  Increases  regularity  of  spidering   •  Doesn’t  have  to  include  every  page   •  Useful  for  dynamic  content  
  77. Sitemap  Page   •  Improves  crawlability   •  Helps  users

      •  Can  help  flasen  site  architecture  
  78. AdWords  Keyword  Tool   •  Analyse  poten:al  keywords   • 

    Shows  compe::on   •  Shows  search  traffic   •  Useful  as  a  research  tool  
  79. Webmaster  Tools   •  Track  search  queries   •  Discover

     crawl  errors   •  Manage  sitemaps   •  Shows  Google’s  keyword  assessment  
  80. Analy:cs   •  Track  search  terms   •  Adapt  user

     paths   •  Some  tools  allow  tracking  of  bot  traffic  
  81. Log  files   •  Track  bot  traffic   •  Spot

     errors   •  Use  gnuplot  to  visualise  
  82. Rinse  and  repeat   •  Ongoing  analysis   •  Keyword

     evalua:on   •  Link  farming   •  Keep  up  with  algorithm  updates