SlideShare ist ein Scribd-Unternehmen logo
1 von 24
Downloaden Sie, um offline zu lesen
Parvez Ahammad
Principal Engineer / Head of DS&ML Team
Instart Logic Inc.
July 2016
Perceptual Speed Index (PSI)
for Measuring above-fold visual
Quality of Experience of webpages
Acknowledgments
• WPO-Foundation/visualmetrics · GitHub
• Pat Meenan (Google & WebPageTest)
• DS&ML team @ Instart Logic
– Clark Gao
– Matthew Mok
• Karan Kumar (Instart Logic)
Why we (@ InstartLogic) care about metrics for end-user QoE
● Towards a data-driven ‘Control System’ for improving web performance
● Accurate measurement of end-user experience is key!
Page
Load
Idle
Response
to Input
Animation
&
Scrolling
Reference: How Users Perceive the Speed of The Web (2015): Paul Irish / Google
Four phases of interaction: end-user’s perception
The space of end-to-end perf. metrics
• PA’s taxonomy of web performance measurements
A. Milestone timings (of various kinds)
B. Byte size counts (relative to various milestones)
C. Request counts
D. Metrics on above-fold visual QoE
E. ...
Quick Refresher on Speed Index
● Popular public metric on above-fold visual QoE
● Patrick Meenan (WPT / Google Inc.)
● Aggregate function on quickness of visual completion above-fold
● Goes beyond milestone timer based approaches
Measurement of visual progress in Speed Index
● Frame-by-frame VC progress is computed from pixel-histogram comparisons
● Pixel-wise similarity (mean histogram difference / MHD) doesn’t capture visual
perception !
● Perception of Shape / Color / Object similarity
Pixel-wise similarity doesn’t capture shape similarity
Black / White :: 50 / 50
MHD (Mean Histogram Difference) = 0
Pixel-wise similarity doesn’t capture color similarity
WWW
Edward Adelson / MIT
Pixel-wise similarity doesn’t capture object similarity
1 2
3
4
5
6
7
Antonio Torralba / MIT
Proposal for a perceptually oriented visual QoE metric
● Update: Frame-by-frame VC progress computation using SSIM
Original, MHD = 0; SSIM = 1 MHD = 144, SSIM = 0.988 MHD = 144, SSIM = 0.913
MHD = 144, SSIM = 0.840 MHD = 144, SSIM = 0.694 MHD = 142, SSIM = 0.662
Wang / Bovik / Simoncelli - 2004
pSI (perceptual Speed Index)
● Frame-by-frame VC progress computation using SSIM
perceptual Via SSIM
Experiment #1
● Key question: Does the MHD  SSIM change capture “real” website
behavior?
● Experiments with IR-100 websites (April 2015)
● Presented to Pat Meenan in May 2015
● Let’s look at a few illustrative examples
Tracking Perceptual Change: SI vs. pSI – RalphLauren (2015)
Green: % visual completion via SSIM (PSI)
Red: % visual completion via MHD (SI)
Tracking Perceptual Change: SI vs. pSI – ChicosFas (2015)
Green: % visual completion via SSIM (PSI)
Red: % visual completion via MHD (SI)
Tracking Perceptual Change: SI vs. pSI – Etsy (2015)
Green: % visual completion via SSIM (PSI)
Red: % visual completion via MHD (SI)
Tracking Perceptual Change: SI vs. pSI – Express (2015)
Green: % visual completion via SSIM (PSI)
Red: % visual completion via MHD (SI)
Recap: Experiment #1
● Does the “MHD  SSIM change” capture “real” website behavior?
● Experiments with IR-100 websites (April 2015)
● Presented to Pat Meenan in May 2015
● Preliminary conclusions
● SSIM based visual progress measurements match human
perception more closely than MHD
● Key observation: PSI appears higher when visual jitter exists
● Pop-up ads / large lay-out changes / etc.
● SSIM / MHD swap doesn’t affect websites without visual jitter
● ChicosFas / Etsy examples
Experiment #2
● How do SI/PSI compare on a larger sample of Alexa top-K websites?
● Experiments with Alexa top mobile websites (March/April 2016)
● Dataset of 500+ mobile above-fold videos
● Dataset provided by Pat Meenan
● All analysis complete (May 2016)
● Results sent to Pat
● What did we find out?
#1: SI and PSI are highly (linearly) correlated (ρ = 0.91)
#2: Median PSI (5020) is slightly higher than median SI (4857)
Experiment #2
● How do SI/PSI compare on a larger sample of Alexa top-K websites?
● Experiments with Alexa top mobile websites (March/April 2016)
● Dataset of 500+ mobile above-fold videos
● Dataset provided by Pat Meenan
● All analysis complete (May 2016)
● Results sent to Pat
● What did we find out?
● SI and PSI are linearly correlated
● Key observation: PSI is much higher than SI for webpages with
visual jitter / layout thrashing behavior
Summary (1)
● Metrics on above-fold visual QoE of web application delivery must
take human end-user perception into account.
● Speed Index
● Primarily focused on progress of above-fold loading
● Does not account for layout stability
● Perceptual Speed Index
● A perceptually oriented metric to measure above-fold visual QoE
● Designed to account for visual jitter (layout stability)
● Complementary to SI
● PSI implementation complete
● Open sourced  GitHub: WPO-Foundation/visualmetrics
Summary (2)
● For more details and explanations:
● http://www.parvez-ahammad.org/blog/perceptual-speed-index-
psi-for-measuring-above-fold-visual-performance-of-webpages

Weitere ähnliche Inhalte

Ähnlich wie PSI_aboveFold_parvez_2016_07.pptx

Turning Analysis into Action with APIs - Superweek2017
Turning Analysis into Action with APIs - Superweek2017Turning Analysis into Action with APIs - Superweek2017
Turning Analysis into Action with APIs - Superweek2017Mark Edmondson
 
Turning Analysis into Action with APIs - Superweek 2017
Turning Analysis into Action with APIs - Superweek 2017Turning Analysis into Action with APIs - Superweek 2017
Turning Analysis into Action with APIs - Superweek 2017Peter Meyer
 
Measuring web performance with user-centric metrics
Measuring web performance with user-centric metricsMeasuring web performance with user-centric metrics
Measuring web performance with user-centric metricsGiorgos Bamparopoulos
 
Data Driven Design: Using Web Analytics to Improve Information Architectures
Data Driven Design: Using Web Analytics to Improve Information ArchitecturesData Driven Design: Using Web Analytics to Improve Information Architectures
Data Driven Design: Using Web Analytics to Improve Information ArchitecturesAndrea Wiggins
 
Capturing speed of user experience using user timing api
Capturing speed of user experience using user timing apiCapturing speed of user experience using user timing api
Capturing speed of user experience using user timing apiSergeyChernyshev
 
Mozilla Foundation Metrics - presentation to engineers
Mozilla Foundation Metrics - presentation to engineersMozilla Foundation Metrics - presentation to engineers
Mozilla Foundation Metrics - presentation to engineersJohn Schneider
 
TripChain: A Peer-to-Peer Trip Generation Database
TripChain: A Peer-to-Peer Trip Generation DatabaseTripChain: A Peer-to-Peer Trip Generation Database
TripChain: A Peer-to-Peer Trip Generation DatabaseJon Kostyniuk
 
Agile Methods to Develop Tangible Products Quickly
Agile Methods to Develop Tangible Products QuicklyAgile Methods to Develop Tangible Products Quickly
Agile Methods to Develop Tangible Products QuicklyJohn Carter
 
ATS-16: Making Data Count, Krista Nordback
ATS-16: Making Data Count, Krista NordbackATS-16: Making Data Count, Krista Nordback
ATS-16: Making Data Count, Krista NordbackBTAOregon
 
Improving the reported use and impact of institutional repositories
Improving the reported use and impact of institutional repositoriesImproving the reported use and impact of institutional repositories
Improving the reported use and impact of institutional repositoriesKenning Arlitsch
 
Cracking web development
Cracking web developmentCracking web development
Cracking web developmentEyal Kenig
 
The Statistics of Web Performance Analysis
The Statistics of Web Performance AnalysisThe Statistics of Web Performance Analysis
The Statistics of Web Performance AnalysisPhilip Tellis
 
Analytics tool vendors comparison study 2017
Analytics tool vendors comparison study 2017Analytics tool vendors comparison study 2017
Analytics tool vendors comparison study 2017Metriplica
 
Integrating web analysis in the user experience design process
Integrating web analysis in the user experience design processIntegrating web analysis in the user experience design process
Integrating web analysis in the user experience design processinternetarchitects
 
Building for success on the capable web - t3imd 2020
Building for success on the capable web -  t3imd 2020Building for success on the capable web -  t3imd 2020
Building for success on the capable web - t3imd 2020Andrey Lipattsev
 
Web analyticsandbigdata techweek2011
Web analyticsandbigdata techweek2011Web analyticsandbigdata techweek2011
Web analyticsandbigdata techweek2011Raghu Kashyap
 
Intro to Data Analytics with Oscar's Director of Product
 Intro to Data Analytics with Oscar's Director of Product Intro to Data Analytics with Oscar's Director of Product
Intro to Data Analytics with Oscar's Director of ProductProduct School
 
Preparing for Holiday Performance, API edition
Preparing for Holiday Performance, API editionPreparing for Holiday Performance, API edition
Preparing for Holiday Performance, API editionSmartBear
 
Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...
Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...
Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...Sri Ambati
 
UX Analytics for Data-driven Product Development
UX Analytics for Data-driven Product DevelopmentUX Analytics for Data-driven Product Development
UX Analytics for Data-driven Product DevelopmentTrieu Nguyen
 

Ähnlich wie PSI_aboveFold_parvez_2016_07.pptx (20)

Turning Analysis into Action with APIs - Superweek2017
Turning Analysis into Action with APIs - Superweek2017Turning Analysis into Action with APIs - Superweek2017
Turning Analysis into Action with APIs - Superweek2017
 
Turning Analysis into Action with APIs - Superweek 2017
Turning Analysis into Action with APIs - Superweek 2017Turning Analysis into Action with APIs - Superweek 2017
Turning Analysis into Action with APIs - Superweek 2017
 
Measuring web performance with user-centric metrics
Measuring web performance with user-centric metricsMeasuring web performance with user-centric metrics
Measuring web performance with user-centric metrics
 
Data Driven Design: Using Web Analytics to Improve Information Architectures
Data Driven Design: Using Web Analytics to Improve Information ArchitecturesData Driven Design: Using Web Analytics to Improve Information Architectures
Data Driven Design: Using Web Analytics to Improve Information Architectures
 
Capturing speed of user experience using user timing api
Capturing speed of user experience using user timing apiCapturing speed of user experience using user timing api
Capturing speed of user experience using user timing api
 
Mozilla Foundation Metrics - presentation to engineers
Mozilla Foundation Metrics - presentation to engineersMozilla Foundation Metrics - presentation to engineers
Mozilla Foundation Metrics - presentation to engineers
 
TripChain: A Peer-to-Peer Trip Generation Database
TripChain: A Peer-to-Peer Trip Generation DatabaseTripChain: A Peer-to-Peer Trip Generation Database
TripChain: A Peer-to-Peer Trip Generation Database
 
Agile Methods to Develop Tangible Products Quickly
Agile Methods to Develop Tangible Products QuicklyAgile Methods to Develop Tangible Products Quickly
Agile Methods to Develop Tangible Products Quickly
 
ATS-16: Making Data Count, Krista Nordback
ATS-16: Making Data Count, Krista NordbackATS-16: Making Data Count, Krista Nordback
ATS-16: Making Data Count, Krista Nordback
 
Improving the reported use and impact of institutional repositories
Improving the reported use and impact of institutional repositoriesImproving the reported use and impact of institutional repositories
Improving the reported use and impact of institutional repositories
 
Cracking web development
Cracking web developmentCracking web development
Cracking web development
 
The Statistics of Web Performance Analysis
The Statistics of Web Performance AnalysisThe Statistics of Web Performance Analysis
The Statistics of Web Performance Analysis
 
Analytics tool vendors comparison study 2017
Analytics tool vendors comparison study 2017Analytics tool vendors comparison study 2017
Analytics tool vendors comparison study 2017
 
Integrating web analysis in the user experience design process
Integrating web analysis in the user experience design processIntegrating web analysis in the user experience design process
Integrating web analysis in the user experience design process
 
Building for success on the capable web - t3imd 2020
Building for success on the capable web -  t3imd 2020Building for success on the capable web -  t3imd 2020
Building for success on the capable web - t3imd 2020
 
Web analyticsandbigdata techweek2011
Web analyticsandbigdata techweek2011Web analyticsandbigdata techweek2011
Web analyticsandbigdata techweek2011
 
Intro to Data Analytics with Oscar's Director of Product
 Intro to Data Analytics with Oscar's Director of Product Intro to Data Analytics with Oscar's Director of Product
Intro to Data Analytics with Oscar's Director of Product
 
Preparing for Holiday Performance, API edition
Preparing for Holiday Performance, API editionPreparing for Holiday Performance, API edition
Preparing for Holiday Performance, API edition
 
Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...
Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...
Prithvi Prabhu + Shivam Bansal, H2O.ai - Building Blocks for AI Applications ...
 
UX Analytics for Data-driven Product Development
UX Analytics for Data-driven Product DevelopmentUX Analytics for Data-driven Product Development
UX Analytics for Data-driven Product Development
 

PSI_aboveFold_parvez_2016_07.pptx

  • 1. Parvez Ahammad Principal Engineer / Head of DS&ML Team Instart Logic Inc. July 2016 Perceptual Speed Index (PSI) for Measuring above-fold visual Quality of Experience of webpages
  • 2. Acknowledgments • WPO-Foundation/visualmetrics · GitHub • Pat Meenan (Google & WebPageTest) • DS&ML team @ Instart Logic – Clark Gao – Matthew Mok • Karan Kumar (Instart Logic)
  • 3. Why we (@ InstartLogic) care about metrics for end-user QoE ● Towards a data-driven ‘Control System’ for improving web performance ● Accurate measurement of end-user experience is key!
  • 4. Page Load Idle Response to Input Animation & Scrolling Reference: How Users Perceive the Speed of The Web (2015): Paul Irish / Google Four phases of interaction: end-user’s perception
  • 5. The space of end-to-end perf. metrics • PA’s taxonomy of web performance measurements A. Milestone timings (of various kinds) B. Byte size counts (relative to various milestones) C. Request counts D. Metrics on above-fold visual QoE E. ...
  • 6. Quick Refresher on Speed Index ● Popular public metric on above-fold visual QoE ● Patrick Meenan (WPT / Google Inc.) ● Aggregate function on quickness of visual completion above-fold ● Goes beyond milestone timer based approaches
  • 7. Measurement of visual progress in Speed Index ● Frame-by-frame VC progress is computed from pixel-histogram comparisons ● Pixel-wise similarity (mean histogram difference / MHD) doesn’t capture visual perception ! ● Perception of Shape / Color / Object similarity
  • 8. Pixel-wise similarity doesn’t capture shape similarity Black / White :: 50 / 50 MHD (Mean Histogram Difference) = 0
  • 9. Pixel-wise similarity doesn’t capture color similarity WWW Edward Adelson / MIT
  • 10. Pixel-wise similarity doesn’t capture object similarity 1 2 3 4 5 6 7 Antonio Torralba / MIT
  • 11. Proposal for a perceptually oriented visual QoE metric ● Update: Frame-by-frame VC progress computation using SSIM Original, MHD = 0; SSIM = 1 MHD = 144, SSIM = 0.988 MHD = 144, SSIM = 0.913 MHD = 144, SSIM = 0.840 MHD = 144, SSIM = 0.694 MHD = 142, SSIM = 0.662 Wang / Bovik / Simoncelli - 2004
  • 12. pSI (perceptual Speed Index) ● Frame-by-frame VC progress computation using SSIM perceptual Via SSIM
  • 13. Experiment #1 ● Key question: Does the MHD  SSIM change capture “real” website behavior? ● Experiments with IR-100 websites (April 2015) ● Presented to Pat Meenan in May 2015 ● Let’s look at a few illustrative examples
  • 14. Tracking Perceptual Change: SI vs. pSI – RalphLauren (2015) Green: % visual completion via SSIM (PSI) Red: % visual completion via MHD (SI)
  • 15. Tracking Perceptual Change: SI vs. pSI – ChicosFas (2015) Green: % visual completion via SSIM (PSI) Red: % visual completion via MHD (SI)
  • 16. Tracking Perceptual Change: SI vs. pSI – Etsy (2015) Green: % visual completion via SSIM (PSI) Red: % visual completion via MHD (SI)
  • 17. Tracking Perceptual Change: SI vs. pSI – Express (2015) Green: % visual completion via SSIM (PSI) Red: % visual completion via MHD (SI)
  • 18. Recap: Experiment #1 ● Does the “MHD  SSIM change” capture “real” website behavior? ● Experiments with IR-100 websites (April 2015) ● Presented to Pat Meenan in May 2015 ● Preliminary conclusions ● SSIM based visual progress measurements match human perception more closely than MHD ● Key observation: PSI appears higher when visual jitter exists ● Pop-up ads / large lay-out changes / etc. ● SSIM / MHD swap doesn’t affect websites without visual jitter ● ChicosFas / Etsy examples
  • 19. Experiment #2 ● How do SI/PSI compare on a larger sample of Alexa top-K websites? ● Experiments with Alexa top mobile websites (March/April 2016) ● Dataset of 500+ mobile above-fold videos ● Dataset provided by Pat Meenan ● All analysis complete (May 2016) ● Results sent to Pat ● What did we find out?
  • 20. #1: SI and PSI are highly (linearly) correlated (ρ = 0.91)
  • 21. #2: Median PSI (5020) is slightly higher than median SI (4857)
  • 22. Experiment #2 ● How do SI/PSI compare on a larger sample of Alexa top-K websites? ● Experiments with Alexa top mobile websites (March/April 2016) ● Dataset of 500+ mobile above-fold videos ● Dataset provided by Pat Meenan ● All analysis complete (May 2016) ● Results sent to Pat ● What did we find out? ● SI and PSI are linearly correlated ● Key observation: PSI is much higher than SI for webpages with visual jitter / layout thrashing behavior
  • 23. Summary (1) ● Metrics on above-fold visual QoE of web application delivery must take human end-user perception into account. ● Speed Index ● Primarily focused on progress of above-fold loading ● Does not account for layout stability ● Perceptual Speed Index ● A perceptually oriented metric to measure above-fold visual QoE ● Designed to account for visual jitter (layout stability) ● Complementary to SI ● PSI implementation complete ● Open sourced  GitHub: WPO-Foundation/visualmetrics
  • 24. Summary (2) ● For more details and explanations: ● http://www.parvez-ahammad.org/blog/perceptual-speed-index- psi-for-measuring-above-fold-visual-performance-of-webpages

Hinweis der Redaktion

  1. Add SI/pSI values on top