Back to Details

ux-researcher-designer

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

UX Researcher & Designer

UX研究员与设计师

Overview

概述

Apply systematic UX research and design methods to understand users, validate assumptions, and create evidence-based designs. This skill covers the full research-to-design pipeline: discovery research, persona creation, journey mapping, information architecture, usability testing, and heuristic evaluation.
运用系统化的UX研究与设计方法了解用户、验证假设，产出基于实证的设计方案。本技能覆盖从研究到设计的全流程：探索性研究、persona创建、旅程地图绘制、信息架构设计、可用性测试以及启发式评估。

Phase 1: Discovery Research

第一阶段：探索性研究

  1. Define research objectives and questions
  2. Select appropriate research methods
  3. Recruit participants (5-8 per segment for qualitative)
  4. Conduct research sessions
  5. Synthesize findings using affinity mapping
STOP — Present research plan with objectives and methods for user approval.
  1. 明确研究目标与问题
  2. 选择合适的研究方法
  3. 招募参与者（定性研究每个用户分层5-8人）
  4. 开展研究访谈/测试
  5. 使用亲和图法整理研究发现
暂停 — 先向用户展示包含目标与方法的研究计划，待确认后再推进。

Research Method Selection Decision Table

研究方法选择决策表

Generative (Discovery) Methods

生成式（探索性）研究方法

MethodWhen to UseParticipantsDurationCost
User InterviewsUnderstanding motivations, behaviors, pain points5-8 per segment45-60 min eachMedium
Contextual InquiryObserving users in their natural environment4-61-2 hours eachHigh
Diary StudiesLongitudinal behavior patterns10-151-4 weeksMedium
SurveysQuantitative validation of qualitative findings100+5-10 minLow
Focus GroupsExploring attitudes and preferences6-10 per group60-90 minMedium
方法适用场景参与者数量耗时成本
用户访谈了解用户动机、行为、痛点每个分层5-8人每人45-60分钟中等
上下文访谈在用户的真实使用环境中观察行为4-6人每人1-2小时高
日记研究追踪长期行为模式10-15人1-4周中等
问卷调研对定性研究结论做定量验证100人以上5-10分钟低
焦点小组探索用户态度与偏好每组6-10人每组60-90分钟中等

Evaluative Methods

评估式研究方法

MethodWhen to UseParticipantsDurationCost
Usability TestingValidating designs against tasks5-830-60 min eachMedium
A/B TestingComparing two design variants1000+ per variant1-4 weeksLow
Card SortingOrganizing information architecture15-3020-30 minLow
Tree TestingValidating navigation structure50+10-15 minLow
First Click TestingEvaluating initial user instincts30+5-10 minLow
Heuristic EvaluationExpert review without users3-5 evaluators1-2 hoursLow
方法适用场景参与者数量耗时成本
可用性测试验证设计是否能支持用户完成任务5-8人每人30-60分钟中等
A/B测试对比两个设计版本的表现每个版本1000人以上1-4周低
卡片分类梳理信息架构逻辑15-30人每人20-30分钟低
树测试验证导航结构合理性50人以上每人10-15分钟低
首次点击测试评估用户初始操作直觉30人以上每人5-10分钟低
启发式评估无需用户参与的专家评审3-5名评估者1-2小时低

Method Selection Decision Table

方法选择决策表

SituationRecommended MethodWhy
No idea who users areInterviews + contextual inquiryDeep understanding needed
Have assumptions to validateSurveys + usability testingQuantitative confirmation
Redesigning navigationCard sorting + tree testingStructure-focused
Evaluating existing productHeuristic evaluation + usability testFind problems fast
Comparing two designsA/B testingStatistical comparison
Limited budget/timeHeuristic evaluationNo participants needed
Long-term behavior understandingDiary studyCaptures patterns over time
场景推荐方法原因
完全不了解目标用户用户访谈+上下文访谈需要深度认知用户
有假设需要验证问卷调研+可用性测试可获得定量验证结论
重构导航设计卡片分类+树测试聚焦信息结构优化
评估现有产品问题启发式评估+可用性测试可快速定位问题
对比两个设计方案A/B测试可获得统计层面的对比结论
预算/时间有限启发式评估无需招募用户
了解长期行为特征日记研究可捕捉长时间维度的行为模式

Interview Guide Template

访谈指南模板

1. Introduction (5 min)
   - Thank participant, explain purpose
   - Get consent for recording
   - "There are no wrong answers"

2. Warm-up (5 min)
   - Background questions about role/context
   - Current tools and workflows

3. Core Questions (30 min)
   - Open-ended questions about behaviors
   - Follow-up probes: "Tell me more about..."
   - Critical incident: "Describe a time when..."
   - Avoid leading questions

4. Wrap-up (5 min)
   - "Is there anything I didn't ask that you think is important?"
   - Thank and explain next steps
1. Introduction (5 min)
   - Thank participant, explain purpose
   - Get consent for recording
   - "There are no wrong answers"

2. Warm-up (5 min)
   - Background questions about role/context
   - Current tools and workflows

3. Core Questions (30 min)
   - Open-ended questions about behaviors
   - Follow-up probes: "Tell me more about..."
   - Critical incident: "Describe a time when..."
   - Avoid leading questions

4. Wrap-up (5 min)
   - "Is there anything I didn't ask that you think is important?"
   - Thank and explain next steps

Phase 2: Analysis and Modeling

第二阶段：分析与建模

  1. Create user personas from research data
  2. Map user journey for key scenarios
  3. Define information architecture
  4. Identify pain points and opportunities
  5. Prioritize using impact/effort matrix
STOP — Present personas and journey map for review before design validation.
  1. 基于研究数据创建用户persona
  2. 绘制核心场景的用户旅程地图
  3. 定义信息架构
  4. 识别痛点与优化机会
  5. 使用影响/ effort矩阵排序优先级
暂停 — 先向用户展示persona与旅程地图，待评审通过后再进入设计验证环节。

Persona Template

Persona模板

markdown
undefined
markdown
undefined

[Persona Name]

[Persona Name]

Demographics

Demographics

  • Age: [range]
  • Occupation: [role]
  • Technical proficiency: [low/medium/high]
  • Usage frequency: [daily/weekly/monthly]
  • Age: [range]
  • Occupation: [role]
  • Technical proficiency: [low/medium/high]
  • Usage frequency: [daily/weekly/monthly]

Goals

Goals

  1. Primary goal: [what they are trying to achieve]
  2. Secondary goal: [supporting objective]
  3. Tertiary goal: [nice-to-have]
  1. Primary goal: [what they are trying to achieve]
  2. Secondary goal: [supporting objective]
  3. Tertiary goal: [nice-to-have]

Pain Points

Pain Points

  1. [Frustration with current process]
  2. [Unmet need]
  3. [Workaround they have created]
  1. [Frustration with current process]
  2. [Unmet need]
  3. [Workaround they have created]

Behaviors

Behaviors

  • [How they currently solve the problem]
  • [Tools and methods they use]
  • [Decision-making patterns]
  • [How they currently solve the problem]
  • [Tools and methods they use]
  • [Decision-making patterns]

Quotes (from research)

Quotes (from research)

  • "[Verbatim quote that captures their perspective]"
  • "[Another representative quote]"
  • "[Verbatim quote that captures their perspective]"
  • "[Another representative quote]"

Scenario

Scenario

[A paragraph describing a typical day/task where they would use the product]
undefined
[A paragraph describing a typical day/task where they would use the product]
undefined

Persona Quality Decision Table

Persona质量校验表

CheckPassFail
Based on real research dataQuotes and behaviors from interviewsInvented or assumed behaviors
Actionable for designSpecific goals and pain pointsVague "wants to be productive"
Distinct from other personasDifferent goals, behaviors, constraintsOverlapping with another persona
Number of personas2-4 primaryMore than 5 (too many to design for)
校验项合格标准不合格表现
是否基于真实研究数据包含访谈中获得的用户原话与行为描述虚构或假设的用户行为
是否对设计有指导意义有明确的目标与痛点模糊的描述如「想要提升效率」
是否与其他persona有区分度目标、行为、约束均有差异与其他persona特征重叠
persona数量2-4个核心persona超过5个（无法同时适配所有角色）

Journey Map Structure

旅程地图结构

Stages:     Awareness -> Consideration -> Onboarding -> Usage -> Advocacy
                |              |             |          |          |
Actions:   [What they do at each stage]
                |              |             |          |          |
Thoughts:  [What they are thinking]
                |              |             |          |          |
Emotions:  [Frustration/neutral/delight mapped to each stage]
                |              |             |          |          |
Pain Points: [Friction and frustration points]
                |              |             |          |          |
Opportunities: [Design opportunities to improve]
                |              |             |          |          |
Touchpoints: [Channels and interfaces involved]
Stages:     Awareness -> Consideration -> Onboarding -> Usage -> Advocacy
                |              |             |          |          |
Actions:   [What they do at each stage]
                |              |             |          |          |
Thoughts:  [What they are thinking]
                |              |             |          |          |
Emotions:  [Frustration/neutral/delight mapped to each stage]
                |              |             |          |          |
Pain Points: [Friction and frustration points]
                |              |             |          |          |
Opportunities: [Design opportunities to improve]
                |              |             |          |          |
Touchpoints: [Channels and interfaces involved]

Journey Map Elements

旅程地图核心元素

  • Moments of Truth: Critical points where users form lasting impressions
  • Service Blueprints: Front-stage actions mapped to back-stage processes
  • Emotion Curve: Visual line showing emotional highs and lows
  • Gap Analysis: Difference between current and desired experience
  • 关键时刻：用户形成持久印象的关键节点
  • 服务蓝图：前台用户行为与后台支撑流程的对应关系
  • 情绪曲线：展示用户情绪高低变化的可视化曲线
  • 差距分析：当前体验与期望体验的差异

Heuristic Evaluation (Nielsen's 10)

启发式评估（尼尔森十大可用性原则）

#HeuristicWhat to Look For
1Visibility of system statusLoading indicators, progress bars, save confirmations
2Match with real worldNatural language, familiar metaphors, logical order
3User control and freedomUndo, cancel, back, escape hatches
4Consistency and standardsSame action = same result, platform conventions
5Error preventionConfirmation dialogs, constraints, smart defaults
6Recognition over recallVisible options, contextual help, recent history
7Flexibility and efficiencyShortcuts, customization, bulk actions
8Aesthetic and minimalist designNo unnecessary information, clear hierarchy
9Help users with errorsPlain language errors, specific cause, suggest fix
10Help and documentationSearchable, task-oriented, concise
序号可用性原则检查要点
1系统状态可见性加载指示器、进度条、保存成功提示
2贴合真实世界场景自然语言、用户熟悉的隐喻、符合逻辑的顺序
3用户控制权与自由度撤销、取消、返回、退出路径
4一致性与标准化相同操作产生相同结果、符合平台规范
5错误预防确认弹窗、操作约束、智能默认值
6识别优于回忆可见的操作选项、上下文帮助、最近操作历史
7灵活性与效率快捷键、自定义设置、批量操作
8简约美学设计无冗余信息、清晰的信息层级
9帮助用户识别与解决错误通俗易懂的错误提示、明确的错误原因、修复建议
10帮助与文档可搜索、面向任务、简洁明了

Severity Rating Scale

严重程度评分标准

RatingDescriptionAction
0Not a usability problemNo action
1Cosmetic onlyFix if time allows
2Minor problemLow priority fix
3Major problemHigh priority, fix before launch
4Usability catastropheMust fix immediately
评分描述处理建议
0不是可用性问题无需处理
1仅视觉层面问题时间允许的情况下修复
2次要问题低优先级修复
3主要问题高优先级，上线前必须修复
4可用性灾难必须立即修复

Phase 3: Design Validation

第三阶段：设计验证

  1. Create testable prototypes (low or high fidelity)
  2. Plan usability testing sessions
  3. Conduct tests with 5+ participants
  4. Analyze results and iterate
  5. Document findings and recommendations
STOP — Present usability test results and recommendations for review.
  1. 创建可测试的原型（低/高保真均可）
  2. 规划可用性测试流程
  3. 邀请5名以上用户参与测试
  4. 分析结果并迭代设计
  5. 整理发现与优化建议
暂停 — 先向用户展示可用性测试结果与优化建议，待评审通过后再推进。

Prototype Fidelity Decision Table

原型保真度选择决策表

SituationFidelityToolWhy
Early concept validationLow (paper/wireframe)Balsamiq, paperFast iteration, low commitment
Navigation testingMedium (clickable)Figma prototypeTest flow without visual polish
Visual design validationHigh (pixel-perfect)Figma, coded prototypeTest actual look and feel
Interaction validationHigh (coded)HTML/CSS/JS prototypeTest real interactions
场景保真度工具原因
早期概念验证低（手绘/线框图）Balsamiq、手绘迭代速度快、成本低
导航流程测试中（可点击）Figma原型无需视觉设计即可测试流程
视觉设计验证高（像素级还原）Figma、可编码原型测试真实视觉表现
交互效果验证高（可编码）HTML/CSS/JS原型测试真实交互体验

A/B Testing Methodology

A/B测试方法规范

StepDetails
Hypothesis"Changing [X] will [improve/decrease] [metric] because [reason]"
Sample sizePower analysis (95% confidence, 80% power)
DurationMinimum 2 full business cycles (2+ weeks)
Variable controlTest one change at a time
AnalysisStatistical significance (p < 0.05)
步骤细节要求
假设「调整[X]将[提升/降低][指标]，因为[原因]」
样本量功效分析（95%置信度、80%统计功效）
测试时长至少覆盖2个完整业务周期（2周以上）
变量控制每次仅测试一个变量
分析要求统计显著性（p < 0.05）

Common UX Metrics

常见UX指标

MetricWhat It MeasuresBenchmark
Task success rate% completing target task> 78% (acceptable)
Time on taskDuration to complete actionVaries by task
Error rateMistakes per task< 10%
System Usability Scale (SUS)Overall usability score68 = average
Net Promoter Score (NPS)Likelihood to recommend> 0 = good, > 50 = excellent
Customer Effort Score (CES)Ease of experience> 5/7
指标衡量维度基准值
任务成功率完成目标任务的用户占比>78%（合格）
任务耗时完成操作的时长不同任务有不同基准
错误率单次任务中的操作错误次数<10%
系统可用性量表（SUS）整体可用性评分68为平均水平
净推荐值（NPS）用户推荐意愿>0为良好，>50为优秀
客户努力度评分（CES）体验流畅度>5/7

Information Architecture

信息架构

Card Sort Analysis Decision Table

卡片分类分析决策表

Sort TypeWhen to UseAnalysis Method
Open sortDiscovery — users create categoriesSimilarity matrix, dendrogram
Closed sortValidation — sort into predefined categoriesCategory agreement percentage
Hybrid sortBoth — predefined with ability to add newCombined analysis
分类类型适用场景分析方法
开放式分类探索阶段：由用户自主创建分类相似矩阵、树状图
封闭式分类验证阶段：将内容归入预设分类分类匹配度百分比
混合式分类兼顾两者：预设分类+允许用户新增分类组合分析

Navigation Patterns

导航模式

PatternUse Case
Global navigationPersistent across all pages
Local navigationWithin a section
Contextual navigationRelated content links
Utility navigationSettings, account, help
BreadcrumbsLocation within hierarchy
模式适用场景
全局导航所有页面常驻显示
局部导航某个板块内部使用
上下文导航相关内容链接
实用工具导航设置、账户、帮助入口
面包屑导航展示用户在层级结构中的位置

Deliverables Checklist

交付物检查清单

  • Research plan with objectives and methods
  • Participant recruitment screener
  • Interview/test script
  • Affinity map of findings
  • Personas (2-4 primary)
  • Journey map for key scenario
  • Information architecture diagram
  • Usability test report with severity ratings
  • Prioritized recommendations with evidence
  • 包含目标与方法的研究计划
  • 参与者招募筛选问卷
  • 访谈/测试脚本
  • 研究发现亲和图
  • Persona（2-4个核心）
  • 核心场景旅程地图
  • 信息架构图
  • 带严重程度评分的可用性测试报告
  • 有实证支撑的优先级优化建议

Anti-Patterns / Common Mistakes

反模式/常见错误

Anti-PatternWhy It Is WrongWhat to Do Instead
Designing without researchAssumptions lead to wrong designsStart with discovery research
Testing with colleaguesBiased, know too much about productRecruit external participants
Asking users what they wantUsers cannot predict behaviorObserve what they do instead
Confirmation biasOnly seeing what supports beliefsUse structured analysis, multiple evaluators
Too many personas (5+)Cannot design for everyoneKeep to 2-4 primary personas
Skipping synthesisRaw data is not insightsAlways do affinity mapping
Underpowered A/B testsResults are meaningless noiseCalculate sample size before starting
Presenting findings without recommendationsResearch without action is wastedAlways include prioritized next steps
反模式问题所在正确做法
不做研究直接设计主观假设会导致设计不符合用户需求从探索性研究开始
找内部同事做测试存在偏见，对产品过于熟悉招募外部真实用户参与测试
直接问用户想要什么用户无法准确预测自身行为观察用户的实际操作
确认偏见只关注符合自身预设的结论采用结构化分析方法、多人交叉评估
Persona数量过多（5个以上）无法同时适配所有用户需求仅保留2-4个核心persona
跳过信息整合环节原始数据不等于洞察必须完成亲和图整理环节
样本量不足的A/B测试结果无统计意义，属于随机噪声测试前先计算所需最小样本量
只展示研究发现不给出建议无法落地的研究没有价值必须同步给出优先级排序的后续行动建议

Integration Points

集成关联能力

SkillIntegration
ui-ux-pro-max
UX guidelines and design patterns
mobile-design
Mobile usability testing patterns
planning
Research plan is part of the implementation plan
spec-writing
User research informs JTBD specifications
prd-generation
Personas and journey maps feed into PRDs
llm-as-judge
Evaluate design quality with rubrics
能力集成方式
ui-ux-pro-max
参考UX设计规范与模式
mobile-design
参考移动端可用性测试模式
planning
研究计划属于项目执行计划的一部分
spec-writing
用户研究结论为JTBD需求说明提供支撑
prd-generation
Persona与旅程地图是PRD的输入内容
llm-as-judge
使用评分标准评估设计质量

Skill Type

技能类型

FLEXIBLE — Select and combine research methods based on project constraints (budget, timeline, access to users). Lightweight methods (heuristic evaluation, guerrilla testing) are acceptable when full research is impractical.
灵活适配型 — 可根据项目约束（预算、 timeline、用户触达条件）选择组合研究方法。当无法开展完整研究时，可采用轻量化方法（启发式评估、游击测试）替代。