SlideShare a Scribd company logo

Patrick Kajirian - Advanced Search Summit Napa 2019

•Download as PPTX, PDF•
•
Digital Marketers Organization
Digital Marketers Organization

The document discusses how artificial intelligence and machine learning are advancing Google's search capabilities and algorithms. It notes Google's investments in big data, cloud computing, and neural networks to power technologies like AlphaZero, Google Lens, Google Photos, and Google Duplex. The document then summarizes how technical SEO is evolving to adapt to these changes, such as sites needing to optimize for voice search and mobile indexing. It also recommends steps for future-proofing SEO, like leveraging AI tools and focusing on user experience.

1 of 21
The Future of Technical SEO 2019 Advanced Search Summit in Napa Patrick Kajirian
Where is Google Headed?
Google’s Big Play ● Big Data & Cloud Computing ○ More data storage ○ Scalable computational power ○ Robust data aggregation ○ Ubiquitous computing ● Machine Learning & Neural Networks ○ Natural language understanding ○ Virtual Assistants ○ Image recognition ○ Live transcription of video/audio ○ Structured data extraction
AI Applications Within Google’s Ecosystem ● AlphaZero, a single system, taught itself chess, shogi, and Go then defeated the world-champion computer program of each game. ● Game masters cite AlphaZero’s style of play as ground-breaking, highly dynamic, and “unconventional” by human standards.
AI Applications Within Google’s Ecosystem Google Lens Google Photos
AI Applications Within Google’s Ecosystem ● Google Duplex bridges the divide between Google’s systems and the real-world. ○ Used by the Google Assistant to book appointments with offline business. ○ Used to verify local business data within Google Maps at scale.
AI Applications Within Google’s Search Algorithm RankBrain ● Helps Google better relate pages to concepts and understand searcher intent. ● Considered the “third most important ranking signal”. ● Live since 2015, gets periodically re-trained. ● More accurate than human search engineers. Neural Matching ● Helps Google better relate words to searches. ● Impacts 30% of queries as of Sept 2018. “Machine learning generated algorithms are more obscure to humans, and hence require more validation and testing than human ones.” - Eric Enge
How is Technical SEO Evolving?
Evolution of Page Discovery & Index Management PAST PRESENT FUTURE Limited crawl budget Improved crawl budget Unlimited crawl budget Desktop only index Mobile first index, desktop index Mobile, desktop, voice index No JavaScript content got indexed Delayed indexing of JavaScript content Instant indexing of JavaScript content XML sitemaps were important XML sitemaps are important Product feeds are supported XML sitemaps will be less important Indexing API Pagination tags were important Pagination tags are depreciated Canonical tag was important Canonical tag is important Canonical tag will be depreciated
Evolution of Content Relevance & Data Extraction PAST PRESENT FUTURE Meta Keywords tag was important Meta Keywords tag is depreciated No natural language understanding Basic natural language understanding (RankBrain & Neural Matching) Perfected natural language understanding User-driven image captioning User-driven image captioning Basic AI image recognition Perfected AI image recognition Alt tags will be depreciated User-driven video/audio transcription User-driven video/audio transcription Basic AI video/audio transcription Perfected AI video/audio transcription Microdata/Schema Markup (video, recipes, products, ratings, nav, etc.) Microdata/Schema Markup & JSON-LD (news headlines for voice search) Structured Data will be depreciated
Evolution of Authority & Ranking Signals PAST PRESENT FUTURE All content mattered High quality content is an advantage Poor quality content will be penalized All backlinks mattered Quality backlinks are critical (trusted source, topically authoritative) Mentions matter (text) Quality backlinks will matter less Mentions will matter more (text or voice) Context didn’t matter Context matters (location, search history) Context will really matter (more personalization, peer affiliation) Site security didn’t matter HTTPS is a slight ranking boost Pages with malware are flagged Non HTTPS sites will be penalized Sites with malware will be penalized
Evolution of Authority & Ranking Signals PAST PRESENT FUTURE Site speed didn’t matter Being fast is a competitive advantage Being slow will be a ranking penalty AMP will be deprecated UX didn’t matter UX matters (mobile-unfriendliness & intrusive interstitials yield slight penalties) UX will matter more (poor UX will be heavily penalized) Engagement metrics (bounce rates, time on site, etc.) didn’t matter Engagement metrics “don’t directly influence rankings” (data is too noisy) Engagement metrics will directly influence rankings (RankBrain) Social media signals (likes, shares, etc.) didn’t matter Social media signals “don’t directly influence rankings” (data is too noisy) Social media signals won’t matter because AI-powered social engineering will be rampant
How are Tools Adapting?
Tools are Adapting ● Generative adversarial network (GAN) systems ● Deepfake video systems thispersondoesnotexist.com McAfee DeepFake Video Demo
Tools are Adapting ● AI-powered content generators ○ OpenAI’s GPT2 text generator is so good it was deemed too risky to release publicly. SYSTEM PROMPT (HUMAN-WRITTEN) In a shocking finding, scientist discovered a herd of unicorns living in a remote, previously unexplored valley, in the Andes Mountains. Even more surprising to the researchers was the fact that the unicorns spoke perfect English. MODEL COMPLETION (MACHINE-WRITTEN, 10 TRIES) The scientist named the population, after their distinctive horn, Ovid’s Unicorn. These four-horned, silver-white unicorns were previously unknown to science. Now, after almost two centuries, the mystery of what sparked this odd phenomenon is finally solved. Dr. Jorge Pérez, an evolutionary biologist from the University of La Paz, and several companions, were exploring the Andes Mountains when they found a small valley, with no other animals or humans. Pérez noticed that the valley had what appeared to be a natural fountain, surrounded by two peaks of rock and silver snow. Pérez and the others then ventured further into the valley. “By the time we reached the top of one peak, the water looked blue, with some crystals on top,” said Pérez. Pérez and his friends were astonished to see the unicorn herd. These creatures could be seen from the air without having to move too much to see them – they were so close they could touch their horns. While examining these bizarre creatures the scientists discovered that the creatures also spoke some fairly regular English. Pérez stated, “We can see, for example, that they have a common ‘language,’ something like a dialect or dialectic.” Dr. Pérez believes that the unicorns may have originated in Argentina, where the animals were believed to be descendants of a lost race of people who lived there before the arrival of humans in those parts of South America. While their origins are still unclear, some believe that perhaps the creatures were created when a human and a unicorn met each other in a time before human civilization. According to Pérez, “In South America, such incidents seem to be quite common.” However, Pérez also pointed out that it is likely that the only way of knowing for sure if unicorns are indeed the descendants of a lost alien race is through DNA. “But they seem to be able to communicate in English quite well, which I believe is a sign of evolution, or at least a change in social organization,” said the scientist.
Tools are Adapting ● AI-powered content generators ○ Automated Insights has a natural language generation platform that uses data to “automate journalism”. ■ Associated Press transforms corporate earnings data into thousands of publishable stories. ■ Yahoo! automates fantasy football draft reports and match recaps.
Tools are Adapting ● Content writing tools improve topical relevance ○ E.g. Searchmetrics’ Content Experience or Ryte
Tools are Adapting ● Deeper datasets & robust visualization tools ○ Data aggregation via APIs ○ More insightful log analysis ● More sophisticated crawling tools ○ Improved browser emulation ○ Cloud, local client & hybrid spidering tools
How do I Future Proof My SEO?
Future Proof Your SEO ● Leverage AI tools to optimize for content relevance. ● Focus on user experience. ● Optimize for voice search where relevant. ● Tap into data to better inform SEO opportunities. ● Get ahead of Google’s research. a. Google AI Blog - ai.googleblog.com b. Monitor Google’s Patents & White Papers. i. Patents.google.com ii. Bill Slawski’s blog - www.seobythesea.com
Thank You! Patrick Kajirian LinkedIn - www.linkedin.com/in/patrickkajirian Twitter - @pkajirian

More Related Content

Patrick Kajirian - Advanced Search Summit Napa 2019

  • 1. The Future of Technical SEO 2019 Advanced Search Summit in Napa Patrick Kajirian
  • 2. Where is Google Headed?
  • 3. Google’s Big Play ● Big Data & Cloud Computing ○ More data storage ○ Scalable computational power ○ Robust data aggregation ○ Ubiquitous computing ● Machine Learning & Neural Networks ○ Natural language understanding ○ Virtual Assistants ○ Image recognition ○ Live transcription of video/audio ○ Structured data extraction
  • 4. AI Applications Within Google’s Ecosystem ● AlphaZero, a single system, taught itself chess, shogi, and Go then defeated the world-champion computer program of each game. ● Game masters cite AlphaZero’s style of play as ground-breaking, highly dynamic, and “unconventional” by human standards.
  • 5. AI Applications Within Google’s Ecosystem Google Lens Google Photos
  • 6. AI Applications Within Google’s Ecosystem ● Google Duplex bridges the divide between Google’s systems and the real-world. ○ Used by the Google Assistant to book appointments with offline business. ○ Used to verify local business data within Google Maps at scale.
  • 7. AI Applications Within Google’s Search Algorithm RankBrain ● Helps Google better relate pages to concepts and understand searcher intent. ● Considered the “third most important ranking signal”. ● Live since 2015, gets periodically re-trained. ● More accurate than human search engineers. Neural Matching ● Helps Google better relate words to searches. ● Impacts 30% of queries as of Sept 2018. “Machine learning generated algorithms are more obscure to humans, and hence require more validation and testing than human ones.” - Eric Enge
  • 8. How is Technical SEO Evolving?
  • 9. Evolution of Page Discovery & Index Management PAST PRESENT FUTURE Limited crawl budget Improved crawl budget Unlimited crawl budget Desktop only index Mobile first index, desktop index Mobile, desktop, voice index No JavaScript content got indexed Delayed indexing of JavaScript content Instant indexing of JavaScript content XML sitemaps were important XML sitemaps are important Product feeds are supported XML sitemaps will be less important Indexing API Pagination tags were important Pagination tags are depreciated Canonical tag was important Canonical tag is important Canonical tag will be depreciated
  • 10. Evolution of Content Relevance & Data Extraction PAST PRESENT FUTURE Meta Keywords tag was important Meta Keywords tag is depreciated No natural language understanding Basic natural language understanding (RankBrain & Neural Matching) Perfected natural language understanding User-driven image captioning User-driven image captioning Basic AI image recognition Perfected AI image recognition Alt tags will be depreciated User-driven video/audio transcription User-driven video/audio transcription Basic AI video/audio transcription Perfected AI video/audio transcription Microdata/Schema Markup (video, recipes, products, ratings, nav, etc.) Microdata/Schema Markup & JSON-LD (news headlines for voice search) Structured Data will be depreciated
  • 11. Evolution of Authority & Ranking Signals PAST PRESENT FUTURE All content mattered High quality content is an advantage Poor quality content will be penalized All backlinks mattered Quality backlinks are critical (trusted source, topically authoritative) Mentions matter (text) Quality backlinks will matter less Mentions will matter more (text or voice) Context didn’t matter Context matters (location, search history) Context will really matter (more personalization, peer affiliation) Site security didn’t matter HTTPS is a slight ranking boost Pages with malware are flagged Non HTTPS sites will be penalized Sites with malware will be penalized
  • 12. Evolution of Authority & Ranking Signals PAST PRESENT FUTURE Site speed didn’t matter Being fast is a competitive advantage Being slow will be a ranking penalty AMP will be deprecated UX didn’t matter UX matters (mobile-unfriendliness & intrusive interstitials yield slight penalties) UX will matter more (poor UX will be heavily penalized) Engagement metrics (bounce rates, time on site, etc.) didn’t matter Engagement metrics “don’t directly influence rankings” (data is too noisy) Engagement metrics will directly influence rankings (RankBrain) Social media signals (likes, shares, etc.) didn’t matter Social media signals “don’t directly influence rankings” (data is too noisy) Social media signals won’t matter because AI-powered social engineering will be rampant
  • 13. How are Tools Adapting?
  • 14. Tools are Adapting ● Generative adversarial network (GAN) systems ● Deepfake video systems thispersondoesnotexist.com McAfee DeepFake Video Demo
  • 15. Tools are Adapting ● AI-powered content generators ○ OpenAI’s GPT2 text generator is so good it was deemed too risky to release publicly. SYSTEM PROMPT (HUMAN-WRITTEN) In a shocking finding, scientist discovered a herd of unicorns living in a remote, previously unexplored valley, in the Andes Mountains. Even more surprising to the researchers was the fact that the unicorns spoke perfect English. MODEL COMPLETION (MACHINE-WRITTEN, 10 TRIES) The scientist named the population, after their distinctive horn, Ovid’s Unicorn. These four-horned, silver-white unicorns were previously unknown to science. Now, after almost two centuries, the mystery of what sparked this odd phenomenon is finally solved. Dr. Jorge Pérez, an evolutionary biologist from the University of La Paz, and several companions, were exploring the Andes Mountains when they found a small valley, with no other animals or humans. Pérez noticed that the valley had what appeared to be a natural fountain, surrounded by two peaks of rock and silver snow. Pérez and the others then ventured further into the valley. “By the time we reached the top of one peak, the water looked blue, with some crystals on top,” said Pérez. Pérez and his friends were astonished to see the unicorn herd. These creatures could be seen from the air without having to move too much to see them – they were so close they could touch their horns. While examining these bizarre creatures the scientists discovered that the creatures also spoke some fairly regular English. Pérez stated, “We can see, for example, that they have a common ‘language,’ something like a dialect or dialectic.” Dr. Pérez believes that the unicorns may have originated in Argentina, where the animals were believed to be descendants of a lost race of people who lived there before the arrival of humans in those parts of South America. While their origins are still unclear, some believe that perhaps the creatures were created when a human and a unicorn met each other in a time before human civilization. According to Pérez, “In South America, such incidents seem to be quite common.” However, Pérez also pointed out that it is likely that the only way of knowing for sure if unicorns are indeed the descendants of a lost alien race is through DNA. “But they seem to be able to communicate in English quite well, which I believe is a sign of evolution, or at least a change in social organization,” said the scientist.
  • 16. Tools are Adapting ● AI-powered content generators ○ Automated Insights has a natural language generation platform that uses data to “automate journalism”. ■ Associated Press transforms corporate earnings data into thousands of publishable stories. ■ Yahoo! automates fantasy football draft reports and match recaps.
  • 17. Tools are Adapting ● Content writing tools improve topical relevance ○ E.g. Searchmetrics’ Content Experience or Ryte
  • 18. Tools are Adapting ● Deeper datasets & robust visualization tools ○ Data aggregation via APIs ○ More insightful log analysis ● More sophisticated crawling tools ○ Improved browser emulation ○ Cloud, local client & hybrid spidering tools
  • 19. How do I Future Proof My SEO?
  • 20. Future Proof Your SEO ● Leverage AI tools to optimize for content relevance. ● Focus on user experience. ● Optimize for voice search where relevant. ● Tap into data to better inform SEO opportunities. ● Get ahead of Google’s research. a. Google AI Blog - ai.googleblog.com b. Monitor Google’s Patents & White Papers. i. Patents.google.com ii. Bill Slawski’s blog - www.seobythesea.com
  • 21. Thank You! Patrick Kajirian LinkedIn - www.linkedin.com/in/patrickkajirian Twitter - @pkajirian

Editor's Notes

  1. Google AI Alpha Go Google Duo YouTube & audio transcription
  2. RankBrain resources: https://backlinko.com/google-rankbrain-seo by Brian Dean
  3. A generative adversarial network (GAN) is a class of machine learning systems. Two neural networks contest with each other in a zero-sum game framework. This technique can generate photographs that look at least superficially authentic to human observers, having many realistic characteristics. It is a form of unsupervised learning.
  4. Automated Journalism – AI Applications at New York Times, Reuters, Washington Post, Yahoo Sports
  5. Tools that build on TF*IDF principles