SlideShare a Scribd company logo
1 of 9
Shane Lewin
• Duplicate results for
the same product
• Poor categorization
• Poor customer
experience
• Poor success rates
and time-on-site
metrics relative to
curators
• Limited to linear UI
• Limited potential for
personalization
Example: Commerce Search With a
Key Value Index
Commerce Search With a
Contextual Index
• Easy to collapse and
group by categories
• Rich user
experiences are
possible
• Ranking can be
highly personalized
• Enables highly
targeted, rich
advertisements
• Remember the semantic web?
• The amount of training data you need to train
existing algorithms grows exponentially with the
number of nodes
• Can cost millions of dollars to train a 5,000 node graph
• However, lots of recent advances
• Ranking technology is less mature
Problem: Scale
• Limit the total number of nodes in a vertical by building
custom vocabularies
• Don’t try to span the entire space early on: a 1000 node
graph will span 95% for most verticals
• Metrics and ranking models need to be tuned to
requirements of the vertical – don’t build one-size-fits-all
applications
Solution: Custom Ontologies
Solution: Self-Optimizing
Indexes
Yes: Well defined ontologies
- Commerce search, job search, music search
Maybe: Undefined, open ontologies, but an ontology exists that
would span enough to be useful
- Health care search, code queries, app search
- Important: First find the spanning set, then build the index and ranker
No: Unbounded ontologies
- Web search, natural language search, etc.
- Google, IBM, and Microsoft have failed
- Exception: when your customers map the data for you (Facebook,
LinkedIn)
Final Considerations: not a good
option for every solution!
Contextual Search Indexes at Scale

More Related Content

What's hot

How to reduce your cost per lead slidedeck
How to reduce your cost per lead slidedeckHow to reduce your cost per lead slidedeck
How to reduce your cost per lead slidedeckSteve Teneriello
 
PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016
PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016
PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016Danny Mac
 
How to Do PPC with Joel
How to Do PPC with JoelHow to Do PPC with Joel
How to Do PPC with JoelAnton Shulke
 
Building a Website - Art or Science
Building a Website - Art or ScienceBuilding a Website - Art or Science
Building a Website - Art or Sciencebizboost123
 
So you want a website?
So you want a website?So you want a website?
So you want a website?Cole Henley
 

What's hot (6)

How to reduce your cost per lead slidedeck
How to reduce your cost per lead slidedeckHow to reduce your cost per lead slidedeck
How to reduce your cost per lead slidedeck
 
PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016
PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016
PPC 360 Presentation - Smart China Sourcing Summit - Hong Kong 2016
 
Google AdWords - Best Practices
Google AdWords - Best PracticesGoogle AdWords - Best Practices
Google AdWords - Best Practices
 
How to Do PPC with Joel
How to Do PPC with JoelHow to Do PPC with Joel
How to Do PPC with Joel
 
Building a Website - Art or Science
Building a Website - Art or ScienceBuilding a Website - Art or Science
Building a Website - Art or Science
 
So you want a website?
So you want a website?So you want a website?
So you want a website?
 

Viewers also liked

Jane eyles gypsy traveller
Jane eyles gypsy travellerJane eyles gypsy traveller
Jane eyles gypsy travellertammyfurey
 
Residents sites presentation
Residents sites presentationResidents sites presentation
Residents sites presentationtammyfurey
 
Stable way residents association
Stable way residents associationStable way residents association
Stable way residents associationtammyfurey
 
Qntev tech talk - contextual indexes at scale
Qntev tech talk - contextual indexes at scaleQntev tech talk - contextual indexes at scale
Qntev tech talk - contextual indexes at scaleShane Lewin
 
DIGIPAK ANALYSIS
DIGIPAK ANALYSIS DIGIPAK ANALYSIS
DIGIPAK ANALYSIS rmackrory10
 
Eleanor street residents group presentation for 2nd feb
Eleanor street residents group presentation for 2nd febEleanor street residents group presentation for 2nd feb
Eleanor street residents group presentation for 2nd febtammyfurey
 
Presentación1
Presentación1Presentación1
Presentación1losmichael
 
EVALUATION QUESTION 4
EVALUATION QUESTION 4EVALUATION QUESTION 4
EVALUATION QUESTION 4rmackrory10
 
Christopher r m6_a2-b6027 (autosaved)
Christopher r m6_a2-b6027 (autosaved)Christopher r m6_a2-b6027 (autosaved)
Christopher r m6_a2-b6027 (autosaved)robertchristopher36
 
Komunikabideak
KomunikabideakKomunikabideak
Komunikabideakexhumed
 
Boswash andrei pelmus
Boswash andrei pelmusBoswash andrei pelmus
Boswash andrei pelmusAndrei Pelmus
 
Komunikabideak
KomunikabideakKomunikabideak
Komunikabideakexhumed
 
Трансформация медиа
Трансформация медиаТрансформация медиа
Трансформация медиаVictor Saxon
 
Как ввести paywall и не облажаться
Как ввести paywall и не облажатьсяКак ввести paywall и не облажаться
Как ввести paywall и не облажатьсяVictor Saxon
 

Viewers also liked (18)

Jane eyles gypsy traveller
Jane eyles gypsy travellerJane eyles gypsy traveller
Jane eyles gypsy traveller
 
Nme
NmeNme
Nme
 
Residents sites presentation
Residents sites presentationResidents sites presentation
Residents sites presentation
 
Stable way residents association
Stable way residents associationStable way residents association
Stable way residents association
 
Qntev tech talk - contextual indexes at scale
Qntev tech talk - contextual indexes at scaleQntev tech talk - contextual indexes at scale
Qntev tech talk - contextual indexes at scale
 
DIGIPAK ANALYSIS
DIGIPAK ANALYSIS DIGIPAK ANALYSIS
DIGIPAK ANALYSIS
 
Eleanor street residents group presentation for 2nd feb
Eleanor street residents group presentation for 2nd febEleanor street residents group presentation for 2nd feb
Eleanor street residents group presentation for 2nd feb
 
Nme
NmeNme
Nme
 
Presentación1
Presentación1Presentación1
Presentación1
 
EVALUATION QUESTION 4
EVALUATION QUESTION 4EVALUATION QUESTION 4
EVALUATION QUESTION 4
 
Christopher r m6_a2-b6027 (autosaved)
Christopher r m6_a2-b6027 (autosaved)Christopher r m6_a2-b6027 (autosaved)
Christopher r m6_a2-b6027 (autosaved)
 
Andrei pelmus
Andrei pelmusAndrei pelmus
Andrei pelmus
 
Komunikabideak
KomunikabideakKomunikabideak
Komunikabideak
 
Boswash andrei pelmus
Boswash andrei pelmusBoswash andrei pelmus
Boswash andrei pelmus
 
Komunikabideak
KomunikabideakKomunikabideak
Komunikabideak
 
Трансформация медиа
Трансформация медиаТрансформация медиа
Трансформация медиа
 
LTEcloudSecurityIssuesTakeaways-GP
LTEcloudSecurityIssuesTakeaways-GPLTEcloudSecurityIssuesTakeaways-GP
LTEcloudSecurityIssuesTakeaways-GP
 
Как ввести paywall и не облажаться
Как ввести paywall и не облажатьсяКак ввести paywall и не облажаться
Как ввести paywall и не облажаться
 

Similar to Contextual Search Indexes at Scale

Friesens agile adoption
Friesens agile adoptionFriesens agile adoption
Friesens agile adoptionsdeconf
 
5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...
5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...
5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...Kanban Solutions
 
Driving Digital Transformation with Machine Learning in Oracle Analytics
Driving Digital Transformation with Machine Learning in Oracle AnalyticsDriving Digital Transformation with Machine Learning in Oracle Analytics
Driving Digital Transformation with Machine Learning in Oracle AnalyticsPerficient, Inc.
 
Online Marketing Success Strategies for Ecommerce: AdTech 2012 Delhi
Online Marketing Success Strategies for Ecommerce: AdTech 2012 DelhiOnline Marketing Success Strategies for Ecommerce: AdTech 2012 Delhi
Online Marketing Success Strategies for Ecommerce: AdTech 2012 DelhiOhana Media
 
Driving Online Sales - Jill Culbertson, Developing your online marketing stra...
Driving Online Sales - Jill Culbertson, Developing your online marketing stra...Driving Online Sales - Jill Culbertson, Developing your online marketing stra...
Driving Online Sales - Jill Culbertson, Developing your online marketing stra...Invest Northern Ireland
 
Planning Your Web Build - The Blueprint for Digital Performance
Planning Your Web Build - The Blueprint for Digital PerformancePlanning Your Web Build - The Blueprint for Digital Performance
Planning Your Web Build - The Blueprint for Digital PerformanceGareth Cartman
 
Creating the blueprint for digital performance
Creating the blueprint for digital performanceCreating the blueprint for digital performance
Creating the blueprint for digital performancesemrush_webinars
 
This Old House: The Marketo Mansion Instance Audit
This Old House:  The Marketo Mansion Instance AuditThis Old House:  The Marketo Mansion Instance Audit
This Old House: The Marketo Mansion Instance AuditPerkuto
 
Advanced PPC Techniques
Advanced PPC TechniquesAdvanced PPC Techniques
Advanced PPC Techniquespointit
 
Why Customer-Driven Content is the Gold Standard for Documentation
Why Customer-Driven Content is the Gold Standard for DocumentationWhy Customer-Driven Content is the Gold Standard for Documentation
Why Customer-Driven Content is the Gold Standard for DocumentationLasselle-Ramsay
 
The Global Marketing Center: A Crafter CMS and Alfresco Case Study
The Global Marketing Center: A Crafter CMS and Alfresco Case StudyThe Global Marketing Center: A Crafter CMS and Alfresco Case Study
The Global Marketing Center: A Crafter CMS and Alfresco Case StudyCrafter Software
 
5 Ways an LMS can Grow your Training Business
5 Ways an LMS can Grow your Training Business5 Ways an LMS can Grow your Training Business
5 Ways an LMS can Grow your Training BusinessUpsideLMS
 
Technical guidance in SaaS Startups
Technical guidance in SaaS StartupsTechnical guidance in SaaS Startups
Technical guidance in SaaS StartupsMalinda Kapuruge
 
SEO for Online Startups - Small Business Festival Victoria 2015
SEO for Online Startups - Small Business Festival Victoria 2015SEO for Online Startups - Small Business Festival Victoria 2015
SEO for Online Startups - Small Business Festival Victoria 2015Optimising
 
This Old House: The Marketo Mansion Instance Audit
This Old House: The Marketo Mansion Instance AuditThis Old House: The Marketo Mansion Instance Audit
This Old House: The Marketo Mansion Instance AuditMarketo
 
Data analytics and SEO to grow your international business | John Caldwell | ...
Data analytics and SEO to grow your international business | John Caldwell | ...Data analytics and SEO to grow your international business | John Caldwell | ...
Data analytics and SEO to grow your international business | John Caldwell | ...Enterprise Ireland
 

Similar to Contextual Search Indexes at Scale (20)

Friesens agile adoption
Friesens agile adoptionFriesens agile adoption
Friesens agile adoption
 
5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...
5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...
5 Reasons Content Strategy & Content Engineering Go Together Like Milk and Or...
 
Driving Digital Transformation with Machine Learning in Oracle Analytics
Driving Digital Transformation with Machine Learning in Oracle AnalyticsDriving Digital Transformation with Machine Learning in Oracle Analytics
Driving Digital Transformation with Machine Learning in Oracle Analytics
 
Benefits of ppc audit
Benefits of ppc auditBenefits of ppc audit
Benefits of ppc audit
 
Online Marketing Success Strategies for Ecommerce: AdTech 2012 Delhi
Online Marketing Success Strategies for Ecommerce: AdTech 2012 DelhiOnline Marketing Success Strategies for Ecommerce: AdTech 2012 Delhi
Online Marketing Success Strategies for Ecommerce: AdTech 2012 Delhi
 
Driving Online Sales - Jill Culbertson, Developing your online marketing stra...
Driving Online Sales - Jill Culbertson, Developing your online marketing stra...Driving Online Sales - Jill Culbertson, Developing your online marketing stra...
Driving Online Sales - Jill Culbertson, Developing your online marketing stra...
 
Traction Notes
Traction NotesTraction Notes
Traction Notes
 
Planning Your Web Build - The Blueprint for Digital Performance
Planning Your Web Build - The Blueprint for Digital PerformancePlanning Your Web Build - The Blueprint for Digital Performance
Planning Your Web Build - The Blueprint for Digital Performance
 
Creating the blueprint for digital performance
Creating the blueprint for digital performanceCreating the blueprint for digital performance
Creating the blueprint for digital performance
 
This Old House: The Marketo Mansion Instance Audit
This Old House:  The Marketo Mansion Instance AuditThis Old House:  The Marketo Mansion Instance Audit
This Old House: The Marketo Mansion Instance Audit
 
Writing for the Web
Writing for the WebWriting for the Web
Writing for the Web
 
Advanced PPC Techniques
Advanced PPC TechniquesAdvanced PPC Techniques
Advanced PPC Techniques
 
Avatech: Workshop Lean Canvas
Avatech: Workshop Lean CanvasAvatech: Workshop Lean Canvas
Avatech: Workshop Lean Canvas
 
Why Customer-Driven Content is the Gold Standard for Documentation
Why Customer-Driven Content is the Gold Standard for DocumentationWhy Customer-Driven Content is the Gold Standard for Documentation
Why Customer-Driven Content is the Gold Standard for Documentation
 
The Global Marketing Center: A Crafter CMS and Alfresco Case Study
The Global Marketing Center: A Crafter CMS and Alfresco Case StudyThe Global Marketing Center: A Crafter CMS and Alfresco Case Study
The Global Marketing Center: A Crafter CMS and Alfresco Case Study
 
5 Ways an LMS can Grow your Training Business
5 Ways an LMS can Grow your Training Business5 Ways an LMS can Grow your Training Business
5 Ways an LMS can Grow your Training Business
 
Technical guidance in SaaS Startups
Technical guidance in SaaS StartupsTechnical guidance in SaaS Startups
Technical guidance in SaaS Startups
 
SEO for Online Startups - Small Business Festival Victoria 2015
SEO for Online Startups - Small Business Festival Victoria 2015SEO for Online Startups - Small Business Festival Victoria 2015
SEO for Online Startups - Small Business Festival Victoria 2015
 
This Old House: The Marketo Mansion Instance Audit
This Old House: The Marketo Mansion Instance AuditThis Old House: The Marketo Mansion Instance Audit
This Old House: The Marketo Mansion Instance Audit
 
Data analytics and SEO to grow your international business | John Caldwell | ...
Data analytics and SEO to grow your international business | John Caldwell | ...Data analytics and SEO to grow your international business | John Caldwell | ...
Data analytics and SEO to grow your international business | John Caldwell | ...
 

Contextual Search Indexes at Scale

  • 2. • Duplicate results for the same product • Poor categorization • Poor customer experience • Poor success rates and time-on-site metrics relative to curators • Limited to linear UI • Limited potential for personalization Example: Commerce Search With a Key Value Index
  • 3. Commerce Search With a Contextual Index • Easy to collapse and group by categories • Rich user experiences are possible • Ranking can be highly personalized • Enables highly targeted, rich advertisements
  • 4.
  • 5. • Remember the semantic web? • The amount of training data you need to train existing algorithms grows exponentially with the number of nodes • Can cost millions of dollars to train a 5,000 node graph • However, lots of recent advances • Ranking technology is less mature Problem: Scale
  • 6. • Limit the total number of nodes in a vertical by building custom vocabularies • Don’t try to span the entire space early on: a 1000 node graph will span 95% for most verticals • Metrics and ranking models need to be tuned to requirements of the vertical – don’t build one-size-fits-all applications Solution: Custom Ontologies
  • 8. Yes: Well defined ontologies - Commerce search, job search, music search Maybe: Undefined, open ontologies, but an ontology exists that would span enough to be useful - Health care search, code queries, app search - Important: First find the spanning set, then build the index and ranker No: Unbounded ontologies - Web search, natural language search, etc. - Google, IBM, and Microsoft have failed - Exception: when your customers map the data for you (Facebook, LinkedIn) Final Considerations: not a good option for every solution!