Tom Capper Mozcon 2021 - Core Web Vitals - The Fast & The Spurious
Download as PPTX, PDF0 likes434 views
The document discusses Core Web Vitals (CWV) and optimizations for them. It notes flaws in how CWV are defined and measured. It also summarizes that Google has delayed rolling out the CWV ranking update multiple times and provided clarifying details. The document suggests prioritizing high traffic pages for CWV work. While metrics can be optimized, improvements should not compromise page speed. Overall CWV may matter less for rankings than other factors like discoverability.
Tom Capper Mozcon 2021 - Core Web Vitals - The Fast & The Spurious
- 1. The Fast & The Spurious Core Web Vitals & SEO Tom Capper Search Scientist, Moz @THCapper
- 2. @THCapper
- 3. @THCapper
- 5. @THCapper You can’t afford to call…
- 6. @THCapper But you can cheat.
- 7. @THCapper
- 8. @THCapper Agenda: The flaws of CWV What’s taking so long? Does it matter? What to do next
- 9. @THCapper Agenda: The flaws of CWV What’s taking so long? Does it matter? What to do next
- 10. @THCapper The flaws of CWV
- 22. @THCapper
- 28. @THCapper
- 29. @THCapper The CrUX of the matter
- 30. @THCapper The CRuX of the matter … we can employ techniques like grouping pages that are similar and compute scores based on that aggregation … - https://support.google.com/webmasters/thread/104436075?hl=en
- 31. @THCapper Agenda: The flaws of CWV What’s taking so long? Does it matter? What to do next
- 32. @THCapper What's taking so long?
- 33. @THCapper May 2020: It’ll be in 2021 https://developers.google.com/search/blog/2020/05/evaluating-page-experience
- 34. @THCapper Nov 2020: It’ll be in May 2021 https://developers.google.com/search/blog/2020/11/timing-for-page-experience
- 35. @THCapper April 2021: It’ll be at least June https://developers.google.com/search/blog/2021/04/more-details-page-experience
- 36. @THCapper I too miss deadlines
- 38. @THCapper Original details If a page hits the recommended targets for all three metrics, it passes the web vitals assessment… https://support.google.com/webmasters/thread/104436075/core-web-vitals-page-experience- faqs-updated-march-2021 “ ”
- 39. @THCapper Original details If a page hits the recommended targets for all three metrics, it passes the web vitals assessment… The page experience ranking impact will be the same for all pages that are in the good range for all Core Web Vitals, irrespective of their individual Core Web Vitals scores. https://support.google.com/webmasters/thread/104436075/core-web-vitals-page-experience- faqs-updated-march-2021 “ ”
- 40. @THCapper May 2021: More clarifications
- 41. @THCapper May 2021: More clarifications It is not the case that unless you reach the good threshold for all of the Core Web Vitals metrics, that you have to reach the threshold to get a ranking boost… https://www.youtube.com/watch?v=HWm6WNkHs90&t=2180s “ ”
- 42. @THCapper May 2021: More clarifications It is not the case that unless you reach the good threshold for all of the Core Web Vitals metrics, that you have to reach the threshold to get a ranking boost… In fact, it's kind of the opposite. You will get a ranking boost for reaching the good threshold for all pages but beyond that point, you don't get additional boost for reaching it even better. https://www.youtube.com/watch?v=HWm6WNkHs90&t=2180s “ ”
- 43. @THCapper May 2021: More clarifications It is not the case that unless you reach the good threshold for all of the Core Web Vitals metrics, that you have to reach the threshold to get a ranking boost… In fact, it's kind of the opposite. You will get a ranking boost for reaching the good threshold for all pages but beyond that point, you don't get additional boost for reaching it even better. https://www.youtube.com/watch?v=HWm6WNkHs90&t=2180s “ ”
- 44. @THCapper May 2021: More clarifications
- 45. @THCapper May 2021: More clarifications
- 57. @THCapper The 7 Wonders of the Web Vitals https://developers.google.com/search/blog/2020/11/timing-for-page-experience
- 58. @THCapper The 7 Wonders of the Web Vitals What if I was already getting a boost for these https://developers.google.com/search/blog/2020/11/timing-for-page-experience
- 59. @THCapper The 7 Wonders of the Web Vitals But I didn’t fix this yet What if I was already getting a boost for these https://developers.google.com/search/blog/2020/11/timing-for-page-experience
- 62. 63,472 , 30% 149,165 , 70% CrUX data No CrUX data
- 63. 52.81% 47.96% 88.11% 63.76% 29.53% CLS LCP FID 2 of 3 All 3 % meeting threshold
- 64. 52.81% 47.96% 88.11% 63.76% 29.53% CLS LCP FID 2 of 3 All 3 % meeting threshold
- 65. 52.81% 47.96% 88.11% 63.76% 29.53% CLS LCP FID 2 of 3 All 3 % meeting threshold
- 66. 52.81% 47.96% 88.11% 63.76% 29.53% CLS LCP FID 2 of 3 All 3 % meeting threshold
- 67. 52.81% 47.96% 88.11% 63.76% 29.53% CLS LCP FID 2 of 3 All 3 % meeting threshold
- 70. @THCapper Agenda: The flaws of CWV What’s taking so long? Does it matter? What to do next
- 71. @THCapper Does it even matter?
- 72. @THCapper 2.31 2.81 0.38 0.26 - 1.00 2.00 3.00 CrUX assessment pass CrUX data available CrUX assessment pass (control for availability or CrUX data) Top 50% quickest pages Average difference in rank
- 73. @THCapper 2.31 2.81 0.38 0.26 - 1.00 2.00 3.00 CrUX assessment pass CrUX data available CrUX assessment pass (control for availability or CrUX data) Top 50% quickest pages Average difference in rank
- 74. @THCapper 2.31 2.81 0.38 0.26 - 1.00 2.00 3.00 CrUX assessment pass CrUX data available CrUX assessment pass (control for availability or CrUX data) Top 50% quickest pages Average difference in rank
- 75. @THCapper 2.31 2.81 0.38 0.26 - 1.00 2.00 3.00 CrUX assessment pass CrUX data available CrUX assessment pass (control for availability or CrUX data) Top 50% quickest pages Average difference in rank
- 76. @THCapper https://developers.google.com/search/mobile-sites/mobile-seo/separate-urls
- 77. @THCapper There’s more to life than SEO(!?) https://www.portent.com/blog/analytics/rese arch-site-speed-hurting-everyones- revenue.htm
- 79. @THCapper Discover • Massive traffic opportunity that SEOs are sleeping on
- 80. @THCapper Discover • Massive traffic opportunity that SEOs are sleeping on • Far more responsive to speed than organic
- 83. @THCapper Agenda: The flaws of CWV What’s taking so long? Does it matter? What to do next
- 84. @THCapper What to do next
- 85. @THCapper Well, what have we learned?
- 86. @THCapper Well, what have we learned? 1. Prioritize high traffic pages … we can employ techniques like grouping pages that are similar and compute scores based on that aggregation … - https://support.google.com/webmasters/thre ad/104436075?hl=en
- 87. @THCapper Well, what have we learned? 1. Prioritize high traffic pages 2. Metrics can be gamed
- 88. @THCapper Well, what have we learned? 1. Prioritize high traffic pages 2. Metrics can be gamed 3. But not at the expense of speed
- 89. @THCapper 1.Prioritise high traffic pages
- 90. @THCapper
- 93. @THCapper Field: Representative As used by Google Lab: Artificial Not Google data
- 94. @THCapper Field: Representative As used by Google Apples to oranges Lab: Artificial Not Google data Scientific
- 95. @THCapper Field: • Representative • As used by Google • Apples to oranges • Only viable for existing, high traffic pages Lab: Artificial Not Google data Scientific Viable for staging sites, new pages, etc.
- 96. @THCapper Field: Representative As used by Google Apples to oranges Only viable for existing, high traffic pages Lab: Artificial Not Google data Scientific Viable for staging sites, new pages, etc.
- 100. @THCapper 2. Metrics can be gamed optimized
- 101. @THCapper 2. Metrics can be gamed optimized Some less obvious methods:
- 103. @THCapper Cumulative Layout Shift ● Use facades for late- loading content
- 104. @THCapper Cumulative Layout Shift ● Use facades for late- loading content
- 105. @THCapper Cumulative Layout Shift ● Use facades for late- loading content ● Move unstable elements below the fold
- 109. @THCapper Cumulative Layout Shift ● Use facades for late- loading content ● Move unstable elements below the fold ● Delay render until elements are ready (!)
- 111. @THCapper Largest Contentful Paint ● Identify which element is “largest”
- 112. @THCapper Largest Contentful Paint ● Identify which element is “largest” ● Reconsider hero images
- 113. @THCapper Largest Contentful Paint ● Identify which element is “largest” ● Reconsider hero images ● Drop big assets below fold
- 115. @THCapper The secret to FID:
- 116. @THCapper The secret to FID: Human manipulation.
- 117. @THCapper First Input Delay Do I need to click immediately?
- 121. @THCapper 3. But not at the expense of speed
- 123. @THCapper
- 124. @THCapper Thank You!