SlideShare a Scribd company logo
What You Don’t Know
About 3rd Party Scripts Can
Hurt You
About the Speakers
Scott D. Lowe, vExpert
Partner and Co-Founder
ActualTech Media
Buddy Brewer
SVP, Products
SOASTA
Jason Trester
Senior Director of Site
Engineering
CBS Interactive
The Problem
 A typical web page today can contain requests with
upward of 75% happening via 3rd party scripts
 Do you really know what kind of experience your
visitors are having?
 How do you manage the risk and the ROI of the tools
provided by these parties?
The Needs
 Measure 3rd party scripts to make sure they’re not
driving your customers away
 Truly understand the impact that every script has on
your visitors
 Provide decision-makers with critical, simple, and
actionable metrics
SOASTA Customer Case Study
Jason Trester
Senior Director of Site Engineering
CBS Interactive
CBS Interactive Key Metrics
200 Million+
Page views per month
350-400 Requests
Per page load
50 Million+
Unique visitors per month
2 Seconds
Target page load time
CBS Interactive Key Metrics
Question
How does CBS Interactive optimize the
user experience and support massive
traffic patterns while adhering to
performance requirements?
CBS Interactive Key Metrics
Answers
It starts by knowing your numbers (Ensure
adherence to business objectives)
Leverage tools to enable actionable
intelligence
Develop data-driven optimization
processes
CBS Interactive Key Metrics
Actionable Intelligence
Leverage tools to enable actionable
intelligence
SOASTA mPulse enables CBS Interactive staff to put site
metrics in front of a non-technical audience so they can
easily see what their page load times look like
SOASTA SpeedCurve has also allowed CBS Interactive
to automate a lot of their manual web page tests
SOASTA mPulse – Actionable Metrics
SOASTA mPulse – Actionable Metrics
SOASTA SpeedCurve – Actionable Metrics
SOASTA SpeedCurve – Actionable Metrics
Actionable Intelligence
 Every third party must go through an A/B test to see
how they perform
 SOASTA enables these efforts
Actionable Intelligence
12/7 RUM Performance Results (Article Page)
3.65s Control
9.73s Sample Tool
267% Increase in Page Load Time
Note: Document Complete = time it takes for page to become usable (required CSS, JS, and Images)
Actionable Intelligence
12/7 Synthetic Performance Results
1. Delayed visual start render by 15%-31%
2. Increased our doc complete size by 49% and time by 113%-150%
3. Sample Tool makes ~31 requests
• 5 Sample Tool requests (30% time; 78% size)
• 27 third-party requests (70% time; 22% size)
• Bandwidth consumption does NOT include Ad creatives
4. Bandwidth consumption 3x > Visual Revenue, Qualtrics, and DW link
5. Bandwidth consumption 3x > CNET Site JS
Example resource comparison
Group Request Count
Time
(ms)
Size
(KB)
CNET JS 19 1,504 168
Sample Tool 33 4,528 210
Excludes ad/tracking services and video player
Actionable Intelligence
1/26 RUM Performance Results (Article Page)
3.01s Control
8.17s Sample Tool
271% Increase in Page Load Time
Note: Document Complete = time it takes for page to become usable (required CSS, JS, and Images)
Operationalizing Intelligence
Develop data-driven optimization
processes
Use the results of performance metrics to make
decisions based on performance cost of 3rd party scripts
Use the information to make critical decisions, such as
priority load order of requests
Key takeaway: It’s easy to glean relative importance of
scripts based on the order in which they load
Operationalizing Intelligence
Operationalizing Intelligence
The Details Matter!
 CBS Interactive spent 6 months making sure that
fonts load quickly
 They established certain requirements on the font vendor
 They try to keep all 3rd party scripts under 1000 ms
 SOASTA enables these efforts
The Tools Matter!
 SOASTA
 mPulse
 SpeedCurve
 Grafana, Sitespeed.io, PhantomJS
 Tealium
 Helps CBS Interactive control the order in which items load
on their sites
 RequireJS
GENERAL GUIDANCE
Figure Out How Fast You Need To Be
Use the Activity Impact Score
• Tells media companies which pages are most
important so they can prioritize their efforts
Define the ROI
• If you were going to make an investment to speed
things up, what would that mean for the business?
(add revenue at expense of performance?)
Try SOASTA for Yourself!
https://www.soasta.com/performance-monitorin
Plans start at just
$99/month!

More Related Content

PPTX
How to measure the business impact of web performance
PPTX
Performance: Key Elements to Consider in the Cloud - RightScale Compute 2013
PPTX
Tis The Season: Load Testing Tips and Checklist for Retail Seasonal Readiness
PPTX
Building a slack bot
PPTX
Using machine learning to determine drivers of bounce and conversion (part 2)
PPTX
Using machine learning to determine drivers of bounce and conversion
PPTX
Empathy in Monitoring
PPT
Understanding the value of agility in the cloud
How to measure the business impact of web performance
Performance: Key Elements to Consider in the Cloud - RightScale Compute 2013
Tis The Season: Load Testing Tips and Checklist for Retail Seasonal Readiness
Building a slack bot
Using machine learning to determine drivers of bounce and conversion (part 2)
Using machine learning to determine drivers of bounce and conversion
Empathy in Monitoring
Understanding the value of agility in the cloud

Similar to What You Don't Know About 3rd Party Scripts Can Hurt You! (20)

PPTX
Applying a Methodical Approach to Website Performance
PPTX
Show Me the Money: Connecting Performance Engineering to Real Business Results
PPTX
Load Speed PSI development of webcore vitals
PPTX
Tech Stack & Web App Development For Startups
PPTX
The Evolution of a Scrappy Startup to a Successful Web Service
PDF
Platform for Comprehensive Vendor Research & Analysis
PPTX
Module 1 introduction to web analytics
PPTX
Module 1 introduction to web analytics
PDF
Website Loading Time Statistics Crucial Data for Optimal Performance.pdf
PPT
Web analyticsandbigdata techweek2011
PDF
17 Web Performance Metrics You Should Care About
PPT
Data Driven Design: Using Web Analytics to Improve Information Architectures
PPT
Content Management Systems: Would You? Could You? Should You?
PPTX
2020 Chrome Dev Summit: Web Performance 101
PPTX
Cms Today: Knowing When You Need A CMS
PPTX
Starter Kit for Collaboration from Karuana @ Microsoft IT
PPTX
Neev Load Testing Services
PDF
10 Essential Steps to Successful Website Development Projects.pdf
PPT
Migrating Large Amounts of Web Content
PDF
The Advancement of Web Design and Development Eflot.pdf
Applying a Methodical Approach to Website Performance
Show Me the Money: Connecting Performance Engineering to Real Business Results
Load Speed PSI development of webcore vitals
Tech Stack & Web App Development For Startups
The Evolution of a Scrappy Startup to a Successful Web Service
Platform for Comprehensive Vendor Research & Analysis
Module 1 introduction to web analytics
Module 1 introduction to web analytics
Website Loading Time Statistics Crucial Data for Optimal Performance.pdf
Web analyticsandbigdata techweek2011
17 Web Performance Metrics You Should Care About
Data Driven Design: Using Web Analytics to Improve Information Architectures
Content Management Systems: Would You? Could You? Should You?
2020 Chrome Dev Summit: Web Performance 101
Cms Today: Knowing When You Need A CMS
Starter Kit for Collaboration from Karuana @ Microsoft IT
Neev Load Testing Services
10 Essential Steps to Successful Website Development Projects.pdf
Migrating Large Amounts of Web Content
The Advancement of Web Design and Development Eflot.pdf
Ad

More from Jennifer Finney (14)

PDF
SOASTA Office Depot Case Study
PPTX
Webinar: How to Load Test for Your Mobile Peak
PDF
Turning Digital Performance into Competitive Advantage
PPTX
Welcome to the Intelligent Testing Revolution!
PPTX
Using JMeter Scripts in CloudTest for Continuous Testing
PDF
Webinar: Proactive Strategies for Finding and Fixing Performance Issues
PPTX
Webinar: Are you ready for your peak season?
PPTX
Webinar: Version Control in SOASTA 57
PPTX
Network Emulation in SOASTA 57 Spring Release
PPTX
Integrating Cloud-based performance test in VSTS with SOASTA CloudTest
PPT
New Features in CloudTest & TouchTest
PDF
How To Introduce Cloud Based Load Testing to Your Jenkins Continuous Delivery...
PDF
DPM Overview Soasta Partners.pptx
PDF
Soasta New Features in CloudTest & TouchTest Webinar
SOASTA Office Depot Case Study
Webinar: How to Load Test for Your Mobile Peak
Turning Digital Performance into Competitive Advantage
Welcome to the Intelligent Testing Revolution!
Using JMeter Scripts in CloudTest for Continuous Testing
Webinar: Proactive Strategies for Finding and Fixing Performance Issues
Webinar: Are you ready for your peak season?
Webinar: Version Control in SOASTA 57
Network Emulation in SOASTA 57 Spring Release
Integrating Cloud-based performance test in VSTS with SOASTA CloudTest
New Features in CloudTest & TouchTest
How To Introduce Cloud Based Load Testing to Your Jenkins Continuous Delivery...
DPM Overview Soasta Partners.pptx
Soasta New Features in CloudTest & TouchTest Webinar
Ad

Recently uploaded (20)

PDF
Enhancing emotion recognition model for a student engagement use case through...
PDF
Microsoft Solutions Partner Drive Digital Transformation with D365.pdf
PDF
Hindi spoken digit analysis for native and non-native speakers
PPTX
Group 1 Presentation -Planning and Decision Making .pptx
PDF
Web App vs Mobile App What Should You Build First.pdf
PDF
DP Operators-handbook-extract for the Mautical Institute
PPTX
TLE Review Electricity (Electricity).pptx
PDF
NewMind AI Weekly Chronicles - August'25-Week II
PDF
Hybrid model detection and classification of lung cancer
PPTX
Programs and apps: productivity, graphics, security and other tools
PDF
Getting started with AI Agents and Multi-Agent Systems
PDF
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
PDF
STKI Israel Market Study 2025 version august
PDF
A comparative study of natural language inference in Swahili using monolingua...
PPTX
1. Introduction to Computer Programming.pptx
PDF
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
PDF
A contest of sentiment analysis: k-nearest neighbor versus neural network
PDF
NewMind AI Weekly Chronicles – August ’25 Week III
PDF
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
Enhancing emotion recognition model for a student engagement use case through...
Microsoft Solutions Partner Drive Digital Transformation with D365.pdf
Hindi spoken digit analysis for native and non-native speakers
Group 1 Presentation -Planning and Decision Making .pptx
Web App vs Mobile App What Should You Build First.pdf
DP Operators-handbook-extract for the Mautical Institute
TLE Review Electricity (Electricity).pptx
NewMind AI Weekly Chronicles - August'25-Week II
Hybrid model detection and classification of lung cancer
Programs and apps: productivity, graphics, security and other tools
Getting started with AI Agents and Multi-Agent Systems
Profit Center Accounting in SAP S/4HANA, S4F28 Col11
STKI Israel Market Study 2025 version august
A comparative study of natural language inference in Swahili using monolingua...
1. Introduction to Computer Programming.pptx
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
A contest of sentiment analysis: k-nearest neighbor versus neural network
NewMind AI Weekly Chronicles – August ’25 Week III
TrustArc Webinar - Click, Consent, Trust: Winning the Privacy Game
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf

What You Don't Know About 3rd Party Scripts Can Hurt You!

  • 1. What You Don’t Know About 3rd Party Scripts Can Hurt You
  • 2. About the Speakers Scott D. Lowe, vExpert Partner and Co-Founder ActualTech Media Buddy Brewer SVP, Products SOASTA Jason Trester Senior Director of Site Engineering CBS Interactive
  • 3. The Problem  A typical web page today can contain requests with upward of 75% happening via 3rd party scripts  Do you really know what kind of experience your visitors are having?  How do you manage the risk and the ROI of the tools provided by these parties?
  • 4. The Needs  Measure 3rd party scripts to make sure they’re not driving your customers away  Truly understand the impact that every script has on your visitors  Provide decision-makers with critical, simple, and actionable metrics
  • 5. SOASTA Customer Case Study Jason Trester Senior Director of Site Engineering CBS Interactive
  • 6. CBS Interactive Key Metrics 200 Million+ Page views per month 350-400 Requests Per page load 50 Million+ Unique visitors per month 2 Seconds Target page load time
  • 7. CBS Interactive Key Metrics Question How does CBS Interactive optimize the user experience and support massive traffic patterns while adhering to performance requirements?
  • 8. CBS Interactive Key Metrics Answers It starts by knowing your numbers (Ensure adherence to business objectives) Leverage tools to enable actionable intelligence Develop data-driven optimization processes
  • 10. Actionable Intelligence Leverage tools to enable actionable intelligence SOASTA mPulse enables CBS Interactive staff to put site metrics in front of a non-technical audience so they can easily see what their page load times look like SOASTA SpeedCurve has also allowed CBS Interactive to automate a lot of their manual web page tests
  • 11. SOASTA mPulse – Actionable Metrics
  • 12. SOASTA mPulse – Actionable Metrics
  • 13. SOASTA SpeedCurve – Actionable Metrics
  • 14. SOASTA SpeedCurve – Actionable Metrics
  • 15. Actionable Intelligence  Every third party must go through an A/B test to see how they perform  SOASTA enables these efforts
  • 16. Actionable Intelligence 12/7 RUM Performance Results (Article Page) 3.65s Control 9.73s Sample Tool 267% Increase in Page Load Time Note: Document Complete = time it takes for page to become usable (required CSS, JS, and Images)
  • 17. Actionable Intelligence 12/7 Synthetic Performance Results 1. Delayed visual start render by 15%-31% 2. Increased our doc complete size by 49% and time by 113%-150% 3. Sample Tool makes ~31 requests • 5 Sample Tool requests (30% time; 78% size) • 27 third-party requests (70% time; 22% size) • Bandwidth consumption does NOT include Ad creatives 4. Bandwidth consumption 3x > Visual Revenue, Qualtrics, and DW link 5. Bandwidth consumption 3x > CNET Site JS Example resource comparison Group Request Count Time (ms) Size (KB) CNET JS 19 1,504 168 Sample Tool 33 4,528 210 Excludes ad/tracking services and video player
  • 18. Actionable Intelligence 1/26 RUM Performance Results (Article Page) 3.01s Control 8.17s Sample Tool 271% Increase in Page Load Time Note: Document Complete = time it takes for page to become usable (required CSS, JS, and Images)
  • 19. Operationalizing Intelligence Develop data-driven optimization processes Use the results of performance metrics to make decisions based on performance cost of 3rd party scripts Use the information to make critical decisions, such as priority load order of requests Key takeaway: It’s easy to glean relative importance of scripts based on the order in which they load
  • 22. The Details Matter!  CBS Interactive spent 6 months making sure that fonts load quickly  They established certain requirements on the font vendor  They try to keep all 3rd party scripts under 1000 ms  SOASTA enables these efforts
  • 23. The Tools Matter!  SOASTA  mPulse  SpeedCurve  Grafana, Sitespeed.io, PhantomJS  Tealium  Helps CBS Interactive control the order in which items load on their sites  RequireJS
  • 25. Figure Out How Fast You Need To Be
  • 26. Use the Activity Impact Score • Tells media companies which pages are most important so they can prioritize their efforts
  • 27. Define the ROI • If you were going to make an investment to speed things up, what would that mean for the business? (add revenue at expense of performance?)
  • 28. Try SOASTA for Yourself! https://www.soasta.com/performance-monitorin Plans start at just $99/month!

Editor's Notes

  • #9: Create a culture where LOB people and decision makers are communicating
  • #14: Wants to show just the middle chart – Jason will send me a better image. Most of this data does not come from SOASTA. But it does speak to the percentage of CBSi stuff vs. other stuff. Move higher in the deck. Comes from sitespeed.io. Takeaway – they only control about 30% of their site.
  • #15: Jason will find a better graph here, too. Add Soasta example to deck
  • #17: Change to header bidding system (not BeanStock)
  • #18: Change to header bidding system (not BeanStock)
  • #26: Come back in and add some color Figure out how fast you need to be Activity Impact Score – tells media companies which pages are most important so they can prioritize their efforts Define the ROI – if you were going to make an investment to speed things up, what would that mean for the business? (add revenue at expense of performance?)
  • #27: Come back in and add some color Figure out how fast you need to be Activity Impact Score – tells media companies which pages are most important so they can prioritize their efforts Define the ROI – if you were going to make an investment to speed things up, what would that mean for the business? (add revenue at expense of performance?)
  • #28: Come back in and add some color Figure out how fast you need to be Activity Impact Score – tells media companies which pages are most important so they can prioritize their efforts Define the ROI – if you were going to make an investment to speed things up, what would that mean for the business? (add revenue at expense of performance?)