1. First Things First: Fix the Tasks!Mike AtyeoJ Boye PresentationWednesday May 4, 2011 V1.0
2. Mike Atyeo Contacts Phone (613) 271-3001 Email info@neoinsight.com Web www.neoinsight.com Mail Suite 374 300 Earl Grey Drive Ottawa, Ontario Canada K2T 1C1 Strategic design(613) 271-3001 x101mike@neoinsight.com Gord Hopkins User Experience Specialist (613) 271-3001 x102gord@neoinsight.com Scott Smith Competitive strategy (613) 271-3001 x103scott@neoinsight.com Lisa Fast Experience Designer (613) 271-3001 x104lisa@neoinsight.com
3. Top task identification Not all tasks are equal - the top 4 tasks garnered the same number of votes as the bottom 43 1. Find a Person 2. Procedures, Policies... 3. Corporate News 4. Training, Learning 30,000 Intranet users... 4 Countries...
4. The Long Neck Target audiences’ votes on their top tasks Tasks 4
5. The Top Task Law Total website quality improvement is proportional to the frequency of use and the importance of the improved tasks to users. Corollary To get the biggest ROI on improvements to your website, concentrate on improving the most frequent and important user tasks.
9. Why Test? Why test behaviour? You are not your user Opinion-wars waste time and money Management responds to measurement Users’ expectations and technology are both changing too rapidly to rely on our past experience alone Small changes can make a big difference Even professionals are never certain of the results Web visitors are real people with things they want to do To obtain objective, continuous measures When people look at a page, they see things differently than when they use a page Users cannot explain their own behaviour
11. Why Test? Why test behaviour? You are not your user Opinion-wars waste time and money Management responds to measurement Users’ expectations and technology are both changing too rapidly to rely on our past experience alone Small changes can make a big difference Even professionals are never certain of the results Web visitors are real people with things they want to do To obtain objective, continuous measures When people look at a page, they see things differently than when they use a page Users cannot explain their own behaviour
13. Why Test? Why test behaviour? You are not your user Opinion-wars waste time and money Management responds to measurement Users’ expectations and technology are both changing too rapidly to rely on our past experience alone Small changes can make a big difference Even professionals are never certain of the results Web visitors are real people with things they want to do To obtain objective, continuous measures When people look at a page, they see things differently than when they use a page Users cannot explain their own behaviour
14. almost no fixations within advertisements users don't fixate within design elements that resemble ads Heatmaps from eyetracking studies: The areas where users looked the most are coloredred; the yellow areas indicate fewer views, followed by the least-viewed blue areas. Gray areas didn't attract any fixations. Green boxes were drawn on top of the images after the study to highlight the advertisements. http://www.useit.com/alertbox/banner-blindness.html
15. Why Test? Why test behaviour? You are not your user Opinion-wars waste time and money Management responds to measurement Users’ expectations and technology are both changing too rapidly to rely on our past experience alone Small changes can make a big difference Even professionals are never certain of the results Web visitors are real people with things they want to do To obtain objective, continuous measures When people look at a page, they see things differently than when they use a page Users cannot explain their own behaviour
19. Despite the fact that a high proportion of the gazes are to the right hand side.
20. And the layout of the page directs attention to the right.
21. Note that they didn’t glance at the image and ignore it – they didn’t even look at it.
22. So this happens subconsciously: people aren’t making a conscious decision to ignore the image.
23. And it happens very quickly – within milliseconds.Tullis, T., Siegel, M & Sun, E. (2009) Proceedings of the 27th International Conference on Human Factors in Computing Systems, Boston, MA.
24. Why Test? Why test behaviour? You are not your user Opinion-wars waste time and money Management responds to measurement Users’ expectations and technology are both changing too rapidly to rely on our past experience alone Small changes can make a big difference Even professionals are never certain of the results Web visitors are real people with things they want to do To obtain objective, continuous measures When people look at a page, they see things differently than when they use a page Users cannot explain their own behaviour
26. Additional navigation at the top of Version A helped convince 19% more visitors to purchase something.” “Clearly customers prefer more links to help them find their own desired path rather than being told what the promos of the day are... The deals damage sales!
27. Task Benchmarking: Metrics Completion rates: percent of successful task completions Disaster rates: percent of people who thought they’d completed the task successfully but had the wrong answer Completion times Target times: estimated or based on another site Frequency: estimated number of times a staff member completes the tasks during a given time Potential return: estimated money or time that can be saved by improving the efficiency of the tasks Confidence: self-rated – in disasters and in success Based on: http://www.customercarewords.com/task-performance-indicator.html
28. Priority: Increase success rates Reduce disasters Then, and only then: Decrease time Overall task performance breakdown
31. Usability Testing: Fast, Iterative Designing without testing is like designing blindfolded Testing one user is twice as good as not testing at all Testing one user early is better than testing 10 later Recruiting representative users is overrated Use realistic tasks, but swap them out when you’re not learning anything new Testing is not to prove/disprove but to inform decisions Testing is an iterative process to be repeated Test early – test often! Based on: Steve Krug, “Rocket surgery Made Easy”, 2010
32. Fix the basics first: Select only the top tasks Increase success rates Reduce disasters Then, and only then: Look at making the task quicker Move on to the next level tasks Tasks 25
33.
34. Mike Atyeo Contacts Phone (613) 271-3001 Email info@neoinsight.com Web www.neoinsight.com Mail Suite 374 300 Earl Grey Drive Ottawa, Ontario Canada K2T 1C1 Strategic design(613) 271-3001 x101mike@neoinsight.com Gord Hopkins User Experience Specialist (613) 271-3001 x102gord@neoinsight.com Scott Smith Competitive strategy (613) 271-3001 x103scott@neoinsight.com Lisa Fast Experience Designer (613) 271-3001 x104lisa@neoinsight.com
