Usability Testing for Technical Writers
Download as PPTX, PDF0 likes69 views
The document provides insights into usability testing specifically for technical writers, highlighting its importance and various methodologies, such as formative and exploratory testing. It lists practical tips for conducting effective usability tests, including focusing on user behavior, selecting appropriate participants, and avoiding biases. Real-world examples demonstrate how usability testing can improve documentation by identifying user difficulties and preferences.
Usability Testing for Technical Writers
- 1. Usability Testing for Technical Writers dimiter simov
- 2. agenda 1. about usability and usability testing 2. tips for better testing experience and results 3. examples from testing documentation with users
- 3. usable that can be used convenient or easy to use
- 4. usability a quality of a thing that shows how fit, convenient, or easy to learn and use that thing is for particular people in a particular context for a particular task
- 5. usability testing a way to evaluate, with the help of users, how usable the thing is
- 6. types of usability testing • formative • exploratory • A/B • remote • 5-seconds • walkthrough • card-sorting essence = users
- 7. why • users will test it anyway; better fix the problems in advance • gives you confidence that you are doing the right thing • speeds you up • makes you rich (check out The $300 Million Button)
- 8. TIPS
- 9. tip 1: solve opinion wars with tests decide based on test data, not opinions
- 10. tip 2: what they say is not what they do “I know how to do it, others would not“ tell them, they were chosen for their qualifications, you want to see them using the product “I want images in the documentation” in fact, they need better documentation; more likely a more usable product
- 11. tip 3: ask someone else to test your baby you are biased - familiarity breeds content
- 12. tip 4: look for patterns base conclusions on behavior observed in several users not just one user
- 13. tip 5: focus on red-route tasks basic and critical tasks first In Outlook: reply to an email or send email new, fun, vanity stuff – if time and resources permit In Outlook: set stationary or signature
- 14. tip 6: give scenarios and goals we do not want to give tasks to users Use the tutorial to learn how to make a birthday-wish video… we’d rather tell a story Your niece has a birthday. Her mother told you she would enjoy a birthday video very much. You want to give it a try.
- 15. tip 7: select representative participants no point testing: IDE with students of English studies online banking with yuppies and skipping house wives phones for grandmas with their grandchildren define user groups/personas/profiles include experienced and novices
- 16. tip 8: start with proper introduction do not scare them; avoid words such as "test"; "we are not testing you but the product" is scary use a script tell them what will happen; give them a few minutes to play around; start with an easy first task
- 17. tip 9: leave them on their own don’t help (unless necessary) watch your tongue and body language
- 18. tip 10: learn what they think participants often need reassurance and ask questions if we answer directly: we don’t learn much we’d rather answer with a question “What makes you think this is …?” “How do you think…?”
- 19. EXAMPLES 1: Testing a tutorial 2: Testing a product use case 3: Card sorting for new IA
- 20. example 1: Testing a tutorial We explicitly tested a piece of documentation
- 21. ex1: Goals 1. Discover how they find info on a scenario 2. Observe users as they use a tutorial Build a Business Application Using Node.js 3. Check how experience matters for: Novices - no or little experience with product Builders - no scenario experience but other experience with the product Extenders - masters of the scenario
- 22. ex1: Flow 1. Let them explore We watched where and what they searched and what they tried 2. Give them the tutorial after that
- 23. ex1: Some points of interest 1. Search 2. Terminology 3. Follow the tutorial 4. Enough info 5. Skip steps 6. Navigate
- 24. ex1: Findings: search • 7 sources of info (no, they do not start at the documentation) • 41 different expressions
- 25. ex1: Findings: terminology Our observations are not conclusive We are kind of OK A few terms caused trouble
- 26. ex1: Findings: follow tutorial • Much harder for inexperienced users • Difficult for confident users – often skip steps • A number of points for improvement • Success rate Novices Builders Extenders 0.5 of 3 2.5 of 6 4 of 4
- 27. ex1: Findings: enough info In general, OK. Some requests/issues for details: • Code samples: descriptions Participants read the code to understand the step • Concepts: links to definitions and details
- 28. ex1: Findings: skip steps Going through the tutorial was 'bumpy'; they skipped texts, phrases, entire sections Interaction pattern: scroll up and down many times get to a step, scroll down to scan the whole step scroll back up, and start following the instructions Reasons: formatting and use of screen real estate
- 29. ex1: Findings: navigate All ended with many tabs and windows Some did not remember or understand how they got from one page to another All tried to use multiple tabs consciously: right-click to open in a new tabs; but it did not always work
- 30. ex1: Findings: navigate within tutorial Top navigation Nobody used it Right navigation: Nobody used it Done buttons: 3 out of 13 used them
- 31. example 2: Testing a product use case We tested a specific scenario with the product and observed how and whether participants used the documentation
- 32. ex2: clip 1 - Find the help
- 33. ex2: clip 1 findings The contextual help is easy to find did not help has no link to the documentation has no link to dive deeper on accounts The user needs to open the documentation
- 34. ex2: clip 2 - Create new item
- 35. ex2: clip 2 findings The concept info is OK (user scanned the topic) No adequate link to creating subaccounts (user scrolled up and down hunting for a link) Search does not return an adequate topic on creating subaccounts (user searched explicitly) Speed matters (user did not wait page to load)
- 36. example 3: Card sorting for new IA We restructured the documentation based on how users order the information
- 37. 2. give to users to arrange in groups and add labels 1. collect objects/concepts/tasks 3. combine into a unified model and build structure ex3: About card sorting plantsred things plants flowers red things flowers …
- 38. ex3: Results Period Dec-13 Apr-14 Jul-14 Aug-14 Dec-14 Apr-15 Jun-15 Mar-17 Visits 21 100 13 700 26 800 21 500 34 900 52 000 48 900 136 000 Users 11 400 6 700 6 400 10 000 18 600 27 100 25 800 67 300 Pages per visit 19.9 12.4 6.12 5.8 4.6 4.3 4.5 3.42 Bounce rate 2% 19% 33% 30% 39% 40% 39% 48% Avg visit duration 05:39 06:41 06:18 06:51 05:39 05:33 05:42 04:50 Avg time on page 00:35 01:14 01:34 01:40 01:38
- 39. ex3: Results
- 40. In summary 1. work with users to solve opinion wars 2. what users say is not what they do, so watch 3. test early and often 4. look for patterns, then draw conclusions 5. red-route tasks first 6. give scenarios and goals; not instructions 7. select representative participants 8. start with proper introduction 9. leave them on their own 10. learn what they think
- 41. YOUR TURN NOW
Editor's Notes
- #2: What is usability testing? Traditionally – people sit and do things while we watch them Extended – any research involving users interacting or thinking about a product (docu) Various methods: interviews, card sorting, tree testing, … Examples Testing a tutorial – checking whether users can follow the tutorial successfully end to end. Testing a scenario in SCP – testing the product explicitly and observing the documentation as well. Card sorting – re-building the information architecture of SCP docu. Take home Some tips