Thriving In A Multi-Browser World – 4 Tips for Optimizing Website Performance across Web Browsers
2 likes992 views
The document discusses the importance of delivering great user experiences in a multi-browser environment, highlighting how uneven browser performance can negatively affect customer engagement and conversion rates. It emphasizes the need for realistic web testing and understanding users' perspectives to optimize performance across various browsers. The document outlines four key tips for improving user experience, including monitoring metrics that matter, evolving optimization techniques, and ensuring website functionality across different browsers.
Thriving In A Multi-Browser World – 4 Tips for Optimizing Website Performance across Web Browsers
- 1. Mike Gualtieri ‐ Senior Analyst, Forrester Research Imad Mouline ‐ CTO, Gomez division, Compuware
- 3. Ensure Great User Experiences In A Multi- Browser World Mike Gualtieri Senior Analyst Forrester Research December 8, 2009
- 4. Uneven browser performance can foil great user experiences.
- 5. Agenda • Great user experiences pay • M lti b Multi-browser l love i a challenge is h ll • Perform realistic Web testing 5 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 6. Your customers will respond to a first- rate user experience t i • Increase purchases – 14.4% more customers will be willing to purchase • Retain more customers – 15.8% more customers will resist doing business with competitors. • Attract more customers – 16 6% more customers will recommend you 16.6% 6 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 7. Leaving user experience to chance hurts h t companiesi • Hurt conversion rates – A well-designed site can have up to a 200% higher visit-to-order conversion rate than a poorly designed site. • Alienate customers – Well-designed sites have page abandonment rates up to 41% lower than their inferior cousins. • Run up development costs – Greatly reduce the need for extensive redesign and redevelopment t fi a poor user experience d l t to fix i 7 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 8. Definition: User Experience –Users’ perceptions of the usefulness, usability, usability and desirability of a Web application based upon the sum of all their direct and indirect interactions with it. Web it W b site performance 8 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 9. Great User Experiences Are Useful, Usable, A d D i bl U f l U bl And Desirable • Users can accomplish Useful their goals • Users can easily Usable perform tasks f t k • Users enjoy their Desirable experience 9 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 10. How would you describe your firm’s goal for f customer experience? t i ? Differentiate ourselves from all 8% firms across any industry Differentiate ourselves from 72% competitors in our industry Maintain parity with other leaders 13% in our industry Keep from falling too far behind 4% leaders in our industry Stay in the mainstream in our 0% industry i d t Stay slightly behind the 2% mainstream in our industry Base: 90 North American companies with $500 million or more in revenues Source: Forrester’s Q4 2008 Customer Experience Peer Research Panel Survey 10 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 11. Check the performance of your peers at http://benchmarks.gomez.com htt //b h k 11 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 12. Agenda • Great user experiences pay • M lti b Multi-browser love is a challenge l i h ll • Perform realistic web testing 12 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 13. Your UX must perform in the browsers used by your customers 13 Entire contents © 2009 Forrester Research, Inc. All rights reserved.
- 14. Web applications are increasingly assembled within the browser bl d ithi th b 14 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 15. Performance differences vary widely Source: Gomez sampled 160 million pages across several Websites over a 30 day period from broadband users in the United States 15 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 16. Browsers versions have different capabilities biliti Source: www.browserscope.com November 19, 2009 16 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 17. The browser wars have turned hot! 17 Entire contents © 2008 Forrester Research, Inc. All rights reserved. 2009 Forrester Research, Inc. All rights reserved.
- 18. “Realistic testing” means testing from your customer s customer’s point of view • Where they do it – LA, Edmonton, London, Bangkok, Des Moines, Miami, Cannes, Dubrovnik, Osaka, Pusan • Wh they d What h do – Key tasks, transactions, and business processes • How they do it – Browsers galore, mobile devices • When they do it (and how frequently) – Peak times, spikes p
- 19. The user’s distance from the content matters tt 19 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 20. Browsers render differently Source: Gomez (http://www.gomez.com) 20 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 21. Agenda • Great user experiences pay • M lti b Multi-browser l love i a challenge is h ll • Perform realistic Web testing 21 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 22. Realistic testing R li ti t ti means taking a t ki walk in your customers’ shoes. Source: Flickr (http://www.flickr.com/photos/megantroyer/175472390/)
- 23. What features should a realistic web performance testing platform have? It should: • Test from the end user’s point of view – Different locations – Different browsers – Different loads – Different tasks • Be easy for developers, testers and businesspeople to use. • Identify and resolve problems along the entire web application delivery chain. 23 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 24. Ensure great user experiences in a multi-browser world lti b ld Great G t UX pays Multi-browser love is i a challenge h ll Perform realistic web testing 24 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 25. Follow best practices in design, architecture, and performance hit t d f 25 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 26. Thank you Mike Gualtieri +1 617 613 6145 1 617.613.6145 mgualtieri@forrester.com Twitter: mgualtieri www.forrester.com 26 Entire contents © 2008 Forrester Research, Inc. All rights reserved.
- 27. Thriving In A Multi‐Browser World 4 Tips for Optimizing Performance across Browsers p p g Imad Mouline ‐ CTO, Gomez
- 28. Gomez: The Web Performance Division of Compuware We help organizations deliver quality experiences to their Web and mobile users Optimal performance and availability O ti l f d il bilit Quality operation across all users, browsers, devices, and geographies Rapid issue notification with actionable diagnostics Insight into how these issues affect your business (revenue, brand, cost) SaaS SaaS 2 500+ Customers 2,500+ Customers Solid Business Solid Business Recognized as Recognized as Platform Worldwide Performance Industry Leader • No installation • 1,250+ enterprise • Five consecutive • Forrester: • Rapid startup & customers years double digit “The leader in payback 1,250+ SMB • 1 250+ SMB g revenue growth p Web Experience customers • 44% revenue Management” • World’s most comprehensive • 12 of top 20 growth 2008 • Gartner: testing network US sites • Offices worldwide “A cool vendor” (100K+ testing locations) l ti )
- 29. Multi‐Browser World Tip # 1 Test from your customers’ perspective
- 30. Adopt an “Outside‐in” Customer Point of View The Web Application Delivery Chain 3rd Party/ Browsers Load Test & monitor your site the way your customers use it: Local ISP Cloud Services and devices Users Balancers • What they do (key pages and transactions) Web • Where they do it (geographic locations) Servers • How they do it (browsers and mobile devices) • When they do it (normal and peak usage) Mobile Components Determine the impact on their behavior and your business App pp Servers “Outside‐in” customer point of view Internet DB Major Servers ISP Storage g Mainframe Network Content Delivery Mobile Networks Carrier Traditional zone Zone of customer expectation Your zone of control with Gomez of control
- 31. Your Customers Use A Variety Of Browsers Browser Market Share Oct 08 to Nov 09 Browser Market Share Oct 08 to Nov 09 70 60 62.2 50 48.4 40 % 38.5 30 27.5 20 10 6.2 2.7 2.5 2.6 0.6 6.9 0 IE Firefox Safari Chrome Opera US Europe Look at your unique results 52.68 % of visitors use Internet Explorer 37.09 % of visitors use Firefox 7.93% of visitors use Safari 2.01 % of visitors use Chrome
- 32. Your Customers Have Different Experiences Based On The Browser They Use The satisfaction index ranges from 0 = frustrated to 1 = satisfied Internet Explorer 6 users have a 0.64 satisfaction index Firefox 3.5 users have a 0.93 satisfaction index Safari 4.0 users have a 0.99 satisfaction index
- 34. Don’t Optimize For Edge Cases Focus efforts on: Top browsers b Highest revenue transactions Hi h t t ti Most popular content Most complex RIA features
- 35. Performance Monitoring Needs To Be Browser Specific Homepage response times differ across Firefox and IE agents p g Internet Explorer Agent Firefox Agent Firefox Agent
- 36. Major Differences In IE And Firefox Agent Waterfall Charts IE Agent response time of 4.3 seconds IE Agent Firefox Agent response time of 2.2 seconds Firefox Agent Connection 3 Connection 1 Connection 2 Connection 2 Connection 1 Connection 6 Connection 5 Connection 4 Connection 2 Connection 7 Connection 8 Connection 9 Connection 3 Connection 10 Connection 10 Connection 5 Connection 4 Connection 4 Connection 12 Connection 6 Connection 11 Connection 7
- 37. Performance & Availability Issues Can Be Browser Specific Detailed data enables quick and easy troubleshooting and root‐cause analysis Gomez’s IE agent detects a performance issue impacting Internet Explorer 7 users Internet Explorer Agent Firefox Agent
- 38. Optimizing Performance Across Browsers Can Reduce Bandwidth And Hosting Costs Internet Explorer 8.x Safari 4.x Firefox 3.x Internet Explorer 7.x Internet Explorer 7.x Chrome Internet Explorer 8.x I E l 8 Safari 4.x Safari 4 x Chrome Firefox 3.x Firefox 3 x
- 39. Multi‐Browser World Tip # 3 Your optimization p techniques must evolve along with browsers Source http://shawnrider.com/wordpress/wp-content/uploads/2006/05/gamerEvolution.jpg
- 40. Do Common Performance Optimization Techniques Still Work? Common technique: Domain sharding Legacy browsers allow 2 connections per hostname Domain sharding increases parallelism to boost static object performance via pointing multiple hostnames to the same host l l h h h For older browsers, this could represent an easy 50% load time savings 2 connections 6 connections 6 connections
- 41. Performance Optimization Techniques Can Backfire In IE8 (& Firefox & Safari & Chrome) domain sharding leads to glut of simultaneous connections & a dramatic performance & infrastructure hit IE 8 Parallelized IE 8 Default
- 42. Browser Connection Profiles Change How You Need To Test Browsers with higher simultaneous connection support increase the strain on web servers and infrastructure Older browsers like IE6 support 2 connections per hostname Newer browsers like IE8 and Chrome support 6 connections per hostname Newer browsers like IE8 and Chrome support 6 connections per hostname
- 44. Tailor Web Applications To Browser Capabilities Not Browsers New browsers = new technologies and standards = new opportunities for optimization and associated benefits New, faster JavaScript engines & parallel JS download New, faster JavaScript engines & parallel JS download Structured client‐side storage HTML 5 & CSS 3 support Bespin determines browser HTML 5 Canvas support at runtime Which browsers provide adequate support? Yes: Firefox 3.5, Safari 4 Yes: Firefox 3 5 Safari 4 No: Firefox 2, Safari 3, Any IE No: Firefox 2 Safari 3 Any IE
- 45. Not Only Performance But Also Rendering Differs Across Browsers Functions correctly Does not function correctly = loss of user confidence in site and brand
- 46. Cross‐Browser Rendering Issues Cause Customer Frustration Functions correctly Does not function correctly
- 47. Cross‐Browser Rendering Issues Cause Brand Damage Functions correctly Functions correctly Does not function correctly
- 48. Recap ‐ 4 Tips To Thrive In A Multi‐Browser World 1. Test from your customers’ perspective 2. Monitor and optimize what matters what matters 3. Your optimization techniques must evolve techniques must evolve along with browsers 4. Ensure your site 4 Ensure your site functions in a multi‐ browser world
- 49. The Gomez Platform: Web Application Delivery Management Self‐Service SaaS Portal Self‐ Web Web Load Web Web Cross‐Browser and Performance Performance Performance Testing Testing Management Business Analysis • Compatibility High volume • High‐volume 20 Real user monitoring • Real‐user monitoring Web performance • Web performance • Functionality backbone load • Last Mile monitoring analytics • Performance • Real‐world • Backbone monitoring • Benchmarks Last Mile load • Internal monitoring • Business dashboard Web 2.0, RIA, Streaming, Mobile Dashboards, Recording & Alerting & APIs & Education & Metrics & Analytics Provisioning Diagnostics Data Feeds Best Practices World’s Most Comprehensive Testing Network 168+ countries 2,500+ ISPs 500+ 100+ 5,000+ 100,000+ combos of commercial nodes supported consumer‐grade browsers & O/S and data centers mobile devices desktops
- 50. Questions Gomez Customers Enjoy Measurable Benefits Increased conversions 10% Reduced homepage load time from 11.3 seconds to 3.4 seconds Improved page load times 23% Improved page load times Saved 50%+ in staff and fees Reduced seven‐step transaction time by 50% Reduced downtime 45% Achieved under 3 second Achieved under 3 second response time and 99%+ availability Validated decision to consolidate Validated decision to consolidate three data centers For more information visit Gomez.com or contact us at +1 781.778.2700