The Fast, The Slow and The Unconverted - Emerce Conversion 2016
•
1 like•1,333 views
Slides from my talk at Emerce Conversion, Amsterdam on the importance of performance(page speed) for conversion. Explore some of the performance issues we face when relying on third-party CRO products / services
Recommended
More Related Content
What's hot
What's hot (20)
Viewers also liked
Viewers also liked (9)
Similar to The Fast, The Slow and The Unconverted - Emerce Conversion 2016
Similar to The Fast, The Slow and The Unconverted - Emerce Conversion 2016 (20)
More from Andy Davies
More from Andy Davies (18)
Recently uploaded
Recently uploaded (20)
The Fast, The Slow and The Unconverted - Emerce Conversion 2016
- 1. The Fast, The Slow, and the Unconverted https://www.flickr.com/photos/dvanzuijlekom/14579895325/ @AndyDavies, NCC Group
- 2. “Has it loaded yet?” http://www.flickr.com/photos/kindofindie/4099768084
- 3. Response Time in Man-computer Conversational Transactions Robert B. Miller, 1968 How we perceive response times 100ms 1s 10s IRUK Retail Top 100 : Average 1.8s delay before page starts showing content
- 4. “50% more concentration when using badly performing web sites” Foviance http://www.flickr.com/photos/yourdon/3366991042
- 5. https://www.flickr.com/photos/editor/57572350 “Stress levels comparable to watching a horror movie” Ericsson Mobility Report 2016
- 6. Abandonment rate for 200+ sites / 177+ million page views over 2 weeks – Measureworks / Gomez Abandonment increases with load time...
- 7. Slow sites = negative impact on brand perception ‘Normal Speed’ 0.5s added delay Mobile Web Stress: Understanding the Neurological ImpactofPoor Performance,Tammy Everts, Radware
- 8. Slow experiences can lose visitors forever!
- 9. Conversions decrease with load time Walmart 2012
- 10. Increased conversions by 10% Shaved 1 second off median home page time 6 seconds off 98th percentile http://www.slideshare.net/cliffcrocker/velocity-ny-how-to-measure-revenue-in-milliseconds
- 11. Improved load time from 1.2s to 0.5s +28% page views / session +21% time on site / visit +20% conversion rate http://blog.quanta-computing.com/etam-earns-20-of-conversion-by-optimising-its-online-store/
- 12. “We reduced latency by 0.3s across the funnel and customers spent 8m a year in additional revenue” Mark Holt, CTO, Trainline New Relic London Summit 2016
- 13. wpostats.com has many more examples
- 14. Good performance is fundamental to the success of our sites
- 15. And we know how to make sites fast…
- 16. https://www.flickr.com/photos/justinjovellanos/15340862812 But we have another challenge…
- 17. https://www.flickr.com/photos/jamescridland/613445810 Revenue attribution Analytics Personalisation A/B & MV testing Advertising Value added content Product images Critical content We depend on Third-Party services for:
- 18. Third-party = Infrastructure and code managed by someone else Guy Podjarny / Velocity Santa Clara 2014
- 19. https://discuss.httparchive.org/t/what-is-the-distribution-of-1st-party-vs-3rd-party-resources/100/7 and we’re using more and more of them 0 5 10 15 20 25 30 2011 2012 2013 2014 2015 Median Number of Third Parties
- 20. http://requestmap.webperf.tools/render/150519_VW_8c8fbf2ecdebe417724d645f9ac3c4cf But do we understand their impact?
- 21. Or the risks they bring? https://http://www.webperformancetoday.com/2011/07/14/fourth-party-calls-third-party-content/
- 22. https://www.flickr.com/photos/devos/163903 With great power… …comes great responsibility
- 23. https://www.flickr.com/photos/matt512/4065627169 What if a third-party includes something that’s blocked in China?
- 24. Or breaks something really important?
- 25. https://www.flickr.com/photos/derekbruff/9759290413/ What about their impact on speed?
- 27. SpeedIndex – Measure of how quickly screen is drawn (A lower is score better)
- 28. 0.00% 0.50% 1.00% 1.50% 2.00% 2.50% Speed Index Percentage Tests With AB Percentage Tests Without AB Pages that use AB or MV Testing are slower Data from httparchive.org
- 29. With Optimizely Optimizely Removed A Real-World Example
- 30. With Optimizely Optimizely Removed A Real-World Example
- 31. With Optimizely Optimizely Removed A Real-World Example
- 32. With Optimizely Optimizely Removed A Real-World Example
- 33. Just adding the testing service changes the visitor’s experience! https://www.flickr.com/photos/therontrowbridge/2654049085
- 34. https://www.flickr.com/photos/87975067@N06/8048048692 Will the value of the tests be larger than the impact on the visitor’s experience?
- 35. Are you brave enough to test the impact of AB testing?
- 36. This is was one example, other types of third-parties have different impacts
- 37. We need to balance third-party benefits with their impact on performance https://www.flickr.com/photos/synx508/5839490755/
- 38. https://www.flickr.com/photos/magickevin/3595741847 Find the third-parties we’re using
- 39. https://www.flickr.com/photos/jamescridland/613445810 Revenue attribution Analytics Personalisation A/B & MV testing Advertising Value added content Product images Critical content Establish the business value they bring
- 40. Remove the ones we don’t use anymore
- 42. https://www.flickr.com/photos/editor/57572350 Use Real User Monitoring to measure visitor’s experiences
- 43. 0.0% 0.5% 1.0% 1.5% 2.0% 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 %ofVisitors Page Load Time (s) Tablet Phone Desktop So we can see how we’re performing for them
- 44. And how it effects their behaviour https://www.flickr.com/photos/mujitra/5232270530
- 45. https://www.flickr.com/photos/borkazoid/221185817 To deliver a great experience we need to be deliberate