Budget Usability without a Usability Budget
•
1 j'aime•376 vues
Presented at the 2010 Michigan Library Association Academic Libraries Conference
Recommandé
Recommandé
Contenu connexe
Tendances
Tendances (12)
En vedette
En vedette (10)
Similaire à Budget Usability without a Usability Budget
Similaire à Budget Usability without a Usability Budget (20)
Dernier
Dernier (20)
Budget Usability without a Usability Budget
- 1. Budget Usability without a Usability Budget Suzanne Chapman User Experience Department Julie Piacentine Reference Department Image by flickr user alancleaver_2000
- 2. Who Are We?
- 3. “The purpose isn’t to What is Budget prove anything; it’s to get Usability? insights that enable you to improve what you’re building” aka "discount” or “informal” or "do it yourself" – Steve Krug It’s not just about the money… It's also about the time and effort. It’s qualitative, informal, and unscientific. Our definition: Anything that you can do with low overhead that involves users interacting with a site. Image by flickr user sarabc
- 4. Image by flickr user electrofantastic Why Use Budget Methods? • Quick answers to simple questions • Faster • Easier • Cheaper • Targeted • More staff participation
- 5. When to Use Budget Techniques? When you just want a quick reaction to something. • Link label • Placement of something • Findability of some piece of content • Attitude towards a design • Do users get it? Image by flickr user s2photo
- 6. How to Use Budget Techniques? • Early and often • Alongside usage statistics & user feedback • In conjunction (or in preparation for) larger evaluations • With a grain of salt Image by flickr user sergesegal
- 7. Participants • Anywhere from 6-100+ • Where & how to find participants: o "in the wild" & on-the-fly! o links from website o emails sent to departments via Subject Specialist Librarians • Incentives: candy, MLibrary gadgets, or a few "blue bucks" each by flickr user faultypixel Image
- 8. Lessons Learned & Tips • Test the test. Time spent piloting the test is time well spent. • Articulate your expectations but be flexible. o Just want general feedback? Ask an open question. o Want to solve a specific problem? Ask a direct question. • Iterate. Know when to admit that something didn't work well. Refine and repeat.
- 10. Participatory Description: Design X/O & Ideal Design Actively involve users in the design process. (inspired by Nancy Foster)
- 11. Participatory Design X/O Instructions: 1. Circle the things you find useful 2. Put an X through the things you don't find useful 3. Add a note for anything that's missing
- 12. Participatory Design Ideal Design Instructions: 1. Draw your ideal library website.
- 13. Participatory Description: Design X/O & Ideal Design 36 Participants: • 15 Undergrads • 5 Grad Students • 2 Faculty • 15 Library Staff Materials Cost: $0 Incentives Cost: $75+ Actively involve users in Set up time: ~1hr the design process. Test time: ~4hrs (inspired by Nancy Foster) Analysis: >12hrs
- 20. Description: Card Sorting Did a combination of sessions with individual participants and groups. 158 Participants: • 18 Undergrads & Grads • 140 Library Staff Materials Cost: $0 / $125 for online tool. Incentives Cost: $90 Ask users to sort a series of Set up time: ~3hrs cards, each labeled with a piece Test time: ~2hrs of content, into groups that make sense to them. Analysis: >10hrs
- 21. Card Sorting - Services/Departments/Libraries Group paper card sort
- 22. Card Sorting - Services/Departments/Libraries OptimalSort online card sort
- 26. Description: • Print out web page Guerrilla Testing • Approach someone “in the wild” & ask if they can spare 5 min. • Ask 1-2 short questions Participants: • 20 undergrad/grad Materials Cost: $0 Incentives Cost: $0 Set up time: ~2hrs Quick and short answers to quick and short questions. Five minutes Test time: ~2hrs is our goal! Analysis: ~4hrs
- 27. Guerrilla Testing Contents: • Removed/added links Labels: • “Quick Links” is good • Some link labels revised Location: • Not good! Needs to be more prominent
- 28. Description: Online Guerrilla “Survey” distributed via Testing Subject Specialist Librarians, news items, and directly from access system interface. Participants: • In progress Materials Cost: $0* Incentives Cost: $0 Set up time: ~1hr Automated version of paper guerrilla test to reach a larger Test time: 0 audience. Analysis: ~1hrs
- 29. Where would you click to find more information about the 1st item in8. Where would you click to go the list? directly to an article?
- 30. Online Guerrilla Testing Tools: • Qualtrics • Infomaki • Optimal Workshop’s Chalkmark • Usabilla • Many more!
- 32. Now, It’s Your Turn! Participatory Design: Circle 3 things you find useful. Put an X through 3 things you don’t find useful.
- 33. Now, It’s Your Turn! Guerrilla Test: Where would you click to find the The New York Times Bestsellers list? Draw a box around it.
- 35. Questions? All past reports: www.lib.umich.edu/usa bility Suzanne Chapman (suzchap@umich.edu) Julie Piacentine (juliepia@umich.edu) Image by flickr user krapow
Notes de l'éditeur
- Traditionally most usability work done through this committee but recently created UX department will now also be focusing on this type of research.Nice to mention that UTF members volunteer and rarely have prior experience so budget techniques are easy to learn and easy to take back to their departments to use for other purposes!
- This is OUR definition and it’s pretty loose.
- Ken – maybe you want the slide to just say “Faster” and fill in the bits in parens?Faster (less time investment for prep)Easier (less time designing evaluations, cut out time-consuming things like recruiting participants)Cheaper (don’t need fancy software or facilities)More targeted (have a question, answer it directly)More staff (with less expertise) can take it on. (and maybe just as reliable results)Doing a couple of budget tests is better than doing nothing. The ramp up to doing formal testing can be prohibitive to actually getting it done at all.
- [this seems better at end, but I like the idea of the last slide being the online guerrilla.]Good example would be to describe our first attempt at guerrilla:We were looking to relabel the link to our various delivery services… we asked 9 people what they’d call it and got 9 different answers. It was still interesting and useful but we had to redesign and redo the test to get a solid answer.
- Suz introask how many people have done “formal” vs. “informal”
- Search language
- talk about how we analyzed data (user groups, sep out areas) to identify trends
- Group Paper Card Sort w. Students 18 participants: undergrads, grad students (divided into 4 groups) Organized 84 cards representing half of this content Allowed us to see interaction among students, hear thought processes, and better understand confusing labels Individual Online Card Sort w. Staff Purchased license to OptimalSort allowing us to place in front of many individuals140 staff completed exercise Provided more data, but didn't expose the thought process
- Exploring the results can be tricky Task Force also came up with "unified" categories, based on the categories the participants created, as well as the comments they made during the card sort. Several similarities between categories surfaced across the various participant groups performing the card sort, whether performing a paper sort or using the online tool. Both the similar groupings across participant groups and the "unified" categories the Task Force came up with were suggested as bases for further tests. Implementing changes will be a large-scale change that would add significant complexities for users and staff.
- Exploring the results can be tricky Task Force also came up with "unified" categories, based on the categories the participants created, as well as the comments they made during the card sort. Several similarities between categories surfaced across the various participant groups performing the card sort, whether performing a paper sort or using the online tool. Both the similar groupings across participant groups and the "unified" categories the Task Force came up with were suggested as bases for further tests. Implementing changes will be a large-scale change that would add significant complexities for users and staff.
- Dark spot:#29 Scholarly Publishing Office + #30 UM PressLight spot:#Asia Library (15) & Area Programs (13): mediumSerials & Microforms Services (41) & Shapiro Undergraduate Library (20) & Askwith Media Library (16) = med dark
- Two Questions, One TestAdvantage: Made good use of participants’ timeDisadvantage: Spent more time analyzing results
- URL of survey is at http://umichlib.qualtrics.com/SE/?SID=SV_3rZvKvGPvIkS1msStill need to set up a TinyURL. And add URL to this slide.
- Set up that we want them to do a hands-on handoutsInstructionsAsk them to compare with neighborAsk to raise hands if they had identical marks as neighbor
- Now, part 2InstructionsRaise hands if they marked the “Books” link (with screenshot)Those who didn’t, where did they click?-say something interesting about % who got it rightBroader discussion about applying this methodSo, what’s something on your library website that you think users might have a hard time finding?
- Now, part 2InstructionsRaise hands if they marked the “Books” link (with screenshot)Those who didn’t, where did they click?-say something interesting about % who got it rightBroader discussion about applying this methodSo, what’s something on your library website that you think users might have a hard time finding?