# How to Innovate
Contextual Inquiry | A vs. B studies |
Contextual Analysis (Design) | Questionnaires |
Paper prototypes | Surveys |
Think-aloud protocols | Interaction Relabeling |
Heuristic Evaluation | Log analysis |
Story Boarding | Focus groups |
Affinity diagrams (WAAD) | Card sorting |
Personas | Diary studies |
Wizard of Oz | Improvisation |
Task analysis | Use cases |
Story Boarding | Scenarios |
Cognitive Walkthrough | Cognitive Dimensions |
KLM and GOMS (CogTool) | "Speed Dating" |
Video prototyping | |
Body storming | |
Expert interviews |
# Contextual Inquiry 情境访谈
- Definition:
- A kind of "ethnographic" or "participatory design" method
一种人种学方法或参与式设计
Combines aspects of other methods: 结合其他几种方法- Interviewing, think-aloud protocols, participant/observer in the context of the work
面试,思考型协议,工作背景下的参与者 / 观察者 - User is given control of interview where the interviewer learns from the interviewee in an open ended way
给予用户访问控制权 - Possible to derive objective conclusions
可能得出客观的结论
- Interviewing, think-aloud protocols, participant/observer in the context of the work
Interpretive field research method 解释性现场研究方法
Depends on conversations with users in the context of their work
取决于与他们工作的背景Used to define requirements, plans and designs.
用于定义需求,计划和设计。
Discover the real requirements of the work
发现工作的真正需求Drives the creative process: 推动创作过程
- In original design 原创设计
- In considering new features or functionality 在考虑新功能时
# Context 情景
- Definition:
- The interrelated conditions within which something occurs or exists
sth 发生或存在相关的条件
Understand work in its natural environment 真实环境中的工作
- Go to the user 基于用户
- Observe real work 观察实际工作
- Use real examples and artifacts 使用真实的例子和产品
"Artifact": An object created by human workmanship 由人工制作的对象 - Interview while she/he is working 工作期间的面谈
- More reliable than asking them 比提问更有效
Context exists even when not a "work" activity
- Use "work" here just to mean "doing something"
- Can be home, entertainment, etc.
# Context Elements 情景元素
- User's work space 用户的工作空间
- User's work 用户的工作内容
- User's workarounds 用户的解决方法
- User's work intentions 用户的工作意图
- User's words (language used) 用户的文字(使用的语言)
- Tools used 使用的工具
- How people work together 人们如何一起工作
- Business goals 业务目标
- Organizational and cultural structure 组织文化结构
# Why Context
Design complete work process 设计完整的工作流程
- Fits into "fabric" of entire operations
适合整个运营的 "结构" - Not just "point solutions" to specific problems
不仅仅是针对特定问题的 "解决方案"
- Fits into "fabric" of entire operations
Integration! 整合
- Consistency, effectiveness, efficiency, coherent
一致性,有效性,效率,连贯性
- Consistency, effectiveness, efficiency, coherent
Design from data 根据数据进行设计
- Not just opinions, negotiation
不只是意见,谈判 - Not just a list of features
不只是功能列表
- Not just opinions, negotiation
# CI Difference
Interviews, Surveys, Focus Groups | Contextual Inquiry |
Summary data & abstractions | Ongoing experience & concrete data |
What customers say | What users do |
Subjective | Objective |
Limited by reliability of human memory | Spontaneous, as it happens |
What customers think they want | What users actually need |
# CI Quick Review Video
Contextual inquiry can be considered as three parts.
情境访谈可以分为三个部分。
The first is identifying the users.
首先是识别用户。Trust the identifying users' your goal should be to visit enough of the right people together, quality data without being swamped.
你的目标应该是访问足够多的合适的人,高质量的数据不被淹没。
Every project is different.
But my rule of thumb is to visit around six people and at least two from each significantly different user group.
我的经验是访问大约六个人,每个明显不同的用户组至少有两个人。
You also need to consider budget as the rule of thumb. Visiting six people will probably take two or three days of planning and organizing, three days for the visits and two or three days of analysis.
你还需要考虑预算。拜访六个人可能需要 2-3 天的计划和组织, 3 天的访问,和 2-3 天的分析。
It's time well spent, but it's a considerable undertaking. Squeezing your timeframe may be a false economy.
这很花时间,但这是一项相当大的任务。压缩你的时间可能是一种错误的节约。The second is scheduling and conducting visits.
第二是安排和进行访问。Don't underestimate the effort and scheduling. Some people may be reluctant to commit too much time, so make sure you clearly communicate the benefits.
不要低估努力和时间安排。有些人可能不愿意花太多时间,所以要确保你清楚地传达了好处。
Generally, when our visit is likely to be acceptable, particularly if people understand what you were doing and make their life. But always let users know what to expect, that is that you want to observe them carrying out their regular to absence, prepare materials such as consent forms, questionnaires, and scripts in advance, and make sure any recording equipment is working.
一般来说,我们的访问可能是可以接受的,特别是如果人们理解你在做什么,并改善他们的生活。但是要让用户知道期望什么,就是你要观察他们通常缺少什么,提前准备好同意书、问卷和脚本等材料,确保所有录音设备都正常工作。
Take a colleague with you on the visits to discuss your observations and put their implications. Or take someone who, again, value from visiting users, a business person, or develop perhaps,
带一位同事与你一起访问,讨论观察结果,记录他们的影响。或者带上那些重视访问用户、业务人员或开发人员的人。
Limit visits to two per day. This gives you time to write up your notes and assimilate, learn, take detailed notes and sketches, and write them up as soon as possible after each visit
将访问限制为每天两次。这让你有时间写笔记,吸收,学习,做详细的笔记和草图,并在每次就诊后尽快写下来。
you know should contain what you saw, what was said, what was done, what artifacts were used on interruptions occurred.
应该包含你所看到的,说了什么,做了什么,在发生中断时使用了什么工件。Finally, we come to analyzing the data.
最后,我们来分析数据。
For small projects, you might be able to use spreadsheet or do a card sport. That's my preferred approach.
对于小型项目,您可以使用电子表格或进行卡片运动。这是我首选的方法。
For large data sets. It's news programs such as NVIVO or dedoose, although these are the steep learning curve, optimal workshops, new tool Reframer, or maybe worth checking out.
对于大型数据,一些新的项目,如 NVIVO or dedoose,虽然这些都是陡峭的学习曲线,最佳研讨会,新的工具 Reframer,或许可能值得一看。
You may want to create persona. As for scenario to communicate what you found, you may need to create a service man for service blueprint if you work in more effort projects. The main thing is to ensure that the data you gathered his mind for your immediate needs communicates well to stakeholders and has retained for future reuse. However you end up with the key is to get out and spend time with users.
您可能想要创建角色。如果您在更多的项目中工作,为了传达所发现的内容方案,您可能需要为服务蓝图创建一个服务人员。主要是确保您为您的即时需求而收集的数据,能够很好地与利益相关者沟通,并保留下来,供将来重用。然而,你最终得到的关键是走出去,花时间与用户在一起。
If you do this with an inquiring and flexible mind, you'll reap many benefits. And your designs will be all the better for your effort.
如果你带着一颗好奇和灵活的心去做这件事,你会得到很多好处。你的努力会让你的设计变得更好。
# Who?
# Users
- Between 6–20
6-20 个用户 - Representative of different roles
代表不同的角色
Note: may not be people who will be doing the purchasing of the system
E.g., if for an enterprise; public kiosk
# Interviewers
"Cross-functional" team
采访者:跨职能团队
- Designers 设计师
- UI specialists UI 专家
- Product managers 产品经理
- Marketing 销售
- Technical people 技术人员
# Partnership 合作关系
The whole approach to this study is that establishing of course sort of like a partnership.
这项研究的整个方法是建立一种伙伴关系。
- Definition:
- A relationship characterized by close cooperation 紧密合作的关系
- Build an equitable relationship with the user 与用户建立公平的关系
- Suspend your assumptions and beliefs 暂停你的假设和信念
- Invite the user into the inquiry process 邀请用户参加调查流程
# Why are Partnerships Important?
Information is obtained through a dialog
通过 对话 获取信息The user is the expert.
用户是 专家Not a conventional interview or consultant relationship
不是常规的采访或顾问关系Alternative way to view the relationship: Master/Apprentice
查看关系的另一种方法:师徒- The user is the "master craftsman" at his/her work
用户是其工作中的 "大师级工匠" - You are the apprentice trying to learn
你是尝试学习的学徒
- The user is the "master craftsman" at his/her work
# Establishing Partnership 建立合作关系
- Share control 分享控制
- Use open-ended questions that invite users to talk:
使用邀请用户交谈的开放式问题- "What are you doing?"
你在做什么 - "Is that what you expect?"
这是你期望的吗 - "Why are you doing...?"
你为什么要做...
- "What are you doing?"
- Let the user lead the conversation
让用户主持对话 - Listen!
倾听 - Pay attention to communication that is non-verbal
注意非语言的交流
# Recording and Noting 记录
# Do record interview
- Video recordings 录像
- Screen capture software with laptop microphone for user
带有麦克风的屏幕捕获软件
# When to take notes?
- Note taking can help you pay closer attention 帮助提高注意力
- Notes lead to faster turn-around 加快周转速度
- Do not let it interfere with interviewing 不要让记笔记干扰采访
Usually would use a second person 笔记通常由另一个人记录
# How to record?
- What the user says – in quotes 用户说的话,用引号引起来
- What the user does – plain text 用户的工作,纯文本
- Your interpretation –in parentheses 你的解释,放在括号中
- Write fast! 快速记录
# Task analysis 任务分析
# Defining Tasks 定义任务
In a real Contextual Inquiry, user decides the tasks
用户决定任务
Investigate real-world tasks, needs, context
调查实际任务,需求,情景But you still must decide the focus
仍然必须确定重点
What tasks you want to observe that are relevant to your product plan
你想要观察的任务,与你的产品计划有关
# Test Tasks 任务测试
Task design is difficult part of usability testing
任务设计是可用性测试中的困难部分Representative of "real" tasks
代表 "实际" 任务- Sufficiently realistic and compelling so users are motivated to finish
足够逼真和引人入胜,使用户有动力完成 - Can let users create their own tasks if relevant
如果相关,可以让用户创建自己的任务
- Sufficiently realistic and compelling so users are motivated to finish
Appropriate difficulty and coverage
适当的难度和覆盖范围- Should last about 2 min.
应该持续约 2 分钟。 - for expert, less than 30 min.
对于专家,少于 30 分钟。 - for novice Short enough to be finished, but not trivial.
对于新手,足够短,可以完成,但不琐碎
- Should last about 2 min.
Tasks nothumorous, frivolous, or offensive
任务不幽默,轻浮或令人反感Easy task first, progressively harder
轻松完成任务,逐步提高难度- But better if independent
但如果独立则更好
- But better if independent
Remember: No tasking their opinions.
没有要求他们发表意见。
# Personas 角色
- Popularized by Alan Cooper 由这个人推广的一种方法
- User archetype you can use to help guide design decisions
是一种用户原型,可以用来帮助指导设计决策 - Created after contextual inquiry or equivalent
在情境访谈后创建 - Summarizes properties of a group of users
汇总一组用户的属性 - Use: helps keep designers & implementers focused on user needs.
使用:帮助使设计师和实施者专注于用户需求。 - Include: behavior patterns, goals, skills, attitudes, and environment, with a few fictional personal details to bring the persona to life
包括:行为模式、目标、技能、态度和环境,以及一些虚构的个人详细信息,使人物形象栩栩如生 - Have a small number for each product
每个产品都有少量- One for each important group of users
每个重要用户组一个
- One for each important group of users
# User + Persona
- Description of user and what user wishes to do
用户说明以及用户希望做什么 - Be specific/detailed, even give names and picture
要具体 / 详细,甚至给出名字和图片 - Three personas for ATM usage follow
使用 ATM 的三个角色- Adapted from User Interface Design and Evaluation
改编自用户界面设计和评估 - The Open University
开放大学
- Adapted from User Interface Design and Evaluation
- Developed by Cooper (1999)
库珀开发(1999)
# Felix – Teenage ATM User
Felix is 13 and gets pocket money each week. He spends it with his friends, so doesn't make regular deposits. He does receive gifts for his birthday, Christmas, etc. and saves that money for special purchases, such as a computer games console or trendy clothes. He has an ATM card allowing him make withdrawals when needed for his purchases.
# Sandra – Young Adult
Sandra is 30, is married to Jason, has two children Todd(6) and Carly Carly (18 months). They live in a subdivision that is about three miles from the town center, where the bank and stores are located. Jason uses the car for work, and works long hours, leaving at 6:45 am and returning at 8:00 pm. Sandra does not drive, so has to use public transportation. She tries to run errands and shop while Todd is in school, so she only has to take Carly to town with her. She typically needs to make two trips to town each week to get everything done. She uses a stroller with Carly, and the bank is one flight up via escalator, so she prefers to use the ATM outside the first floor, even though there is no canopy to protect customers from bad weather.
# Marvin – Middle Ageto Senior
Marvin is 68 years old, and his social security is deposited into his bank account at the start of each month. He goes to the bank every week, withdrawing enough cash for the week for miscellaneous expenditure. Regular bills are paid by check.
He stands in line for a live teller, as he prefers the social interaction to using an ATM, even though his new artificial hip makes standing in line uncomfortable. He does not have an ATM card.
# Initial Questions 起始问题
First, assure the users that not testing them
确保用户未对其进行测试
- You are testing the product, and you want to find problems
您正在测试产品,并且想要查找问题
Find out the context through initial questions
通过最初的问题找出背景
When would you normally do this kind of task?
你通常在何时执行此类任务?Who would be involved in making the decisions?
谁将参与决策?What would influence any decisions?
什么会影响任何决定?How would you know what to do?
你如何知道该怎么办?- What information would you use to help decide?
您将使用哪些信息来帮助做出决定?
- What information would you use to help decide?
# Test Scripts 测试脚本
Useful to have a script
有用的脚本- Make sure say everything you want
确保说出所有想要的内容 - Make sure all users get same instructions
确保所有用户都获得相同的说明
- Make sure say everything you want
Should read instructions out loud
应该大声朗读说明- Ask if users have any questions
询问用户是否有任何问题
- Ask if users have any questions
Make sure instructions provide goals only in a general way, and doesn't give away information
确保说明仅通过一般的方式,不会泄露信息- Describe the resultand not the steps
描述结果而不是步骤 - Avoid product names and technical terms that appear on the web site
避免在网站上出现产品名称和技术术语- Don't give away the vocabulary
不要放弃词汇量
- Don't give away the vocabulary
- Describe the resultand not the steps
Example:
"The clock should have the right time";
时钟应有正确的时间
not: "Use the hours and minutes buttons to set the time"
使用小时和分钟按钮设置时间
# Alternative to CI CI 的替代方法
For intermittent tasks
对于间歇性任务- In-context cued recall
根据情景提示召回 - Activity logs
活动日志
- In-context cued recall
For uninterruptible tasks
对于不间断的任务- Post-observation inquiry
观察后查询
- Post-observation inquiry
For extremely long or multi-person tasks
对于极长或多人的任务- Artifact walkthrough
神器演练
- Artifact walkthrough
New technology within current work
当前工作中的新技术- Future Scenario
未来场景
- Future Scenario
Prototype or prior version exists
存在原型或先前版本- Prototype/Test drive
原型 / 试驾
- Prototype/Test drive
# CI Example from CMU
电子商务站点的示例会话
Issues to observe
要注意的问题
- Interview of work in progress, in "context"
在 “情景” 中采访正在进行的工作 - Actual session of doing a task
执行任务的实际时间- Not an interview asking about possible tasks, etc.
不要在采访中询问可能的任务,等。 - Note that focusing on expert behavior & breakdowns
请注意,关注专家 行为和故障
- Not an interview asking about possible tasks, etc.
- Questions to clarify about routine, motivations
要澄清有关日常活动,动机的问题- Why do certain actions: need intentfor actions
为什么要采取某些行动:需要采取行动的意图 - Notice problems ("breakdowns")
注意问题(“故障”)
- Why do certain actions: need intentfor actions
- Notice what happens that causes users to do something ("triggers")
注意会导致用户采取行动的情况(“触发器”)- E.g. appearance of error messages, other feedback, external events (phone ringing), etc.
例如错误消息的出现,其他反馈,外部事件(电话响铃)等
- E.g. appearance of error messages, other feedback, external events (phone ringing), etc.