UX medium Flashcards
What do you need to do after the briefing with stakeholders?
Create shared document to make sure everybody is on the same level if understanding
Never do during interviews and meetings with stakeholder
Never offer solutions right during the meeting/interview
The tool for transcription audio notes
Otter (English), oTranscribe
What is Business Model Canvas
It is a strategical management template which is used for documenting brief product’s / company’s business model. Consists of nine boxes: key partners, key activities, key resources, value propositions, customer relationships, channels, customer segments, cost structure, revenue streams
Business model canvas scheme
What is the difference between Business model canvas and Lean canvas
We use Business model canvas for existing businesses and Lean canvas for startups
Lean canvas scheme
Value proposition
Describes the benefits customers can expect from your products and services.
Value (Proposition) Map
Describes the features of a specific value proposition in your business model in a more structured and detailed way. It breaks your value proposition down into products and services, pain relievers, and gain creators.
Value Proposition Map consists of
Products&Services, Gain creators, Pain relievers
Customer Profile
The Customer (Segment) Profile describes a specific customer segment in your business model in a more structured and detailed way. It breaks the customer down into its jobs, pains, and gains.
What is a fit between Value map and Customer profile
You achieve Fit when your value map meets your customer profile—when your products and services produce pain relievers and gain creators that match one or more of the jobs, pains, and gains that are important to your customer.
What customer profile consists of
It breaks the customer down into its jobs, pains, and gains.
How do we get to know our customer
- Assumptions
- Secondary research
- Surveys
- Interviews
- Customer Support
- Marketing team
- Database queries
What the story of the user may consist of
• the way of thinking• background• current scenario of problem solving
Persona map
- Goals
- Needs
- Wants
- Fears
- Pain points
- Story
- Picture, name
- Optional demographic
Ideal persona canvas
How to conduct internal stakeholders interview
- Clarify your own research goals
2. Identify stakeholders
Questions to clarify research goals for the interview with stakeholders
- What do you need to learn in order to move forward with this research project?
- Does everyone agree about project objectives, or do they have conflicting goals and ideas?
- How do they perceive their own role in the success of the project?
- What work has already been done, and what needs to be started from scratch?
- What does long-term success look like to each stakeholder, in the context of this project?
- Why are we building this product?
- What do we know about our user’s preferences around this product, and likewise, what are we not yet sure about?
- Are there competitive examples of what we’re building that we should take a look at?
Who are internal stakeholders
Anyone whose job will be impacted by your research is a stakeholder, regardless of seniority or job title.
Hidden stakeholders
People whose input and approval you need, but whose relevance isn’t obvious. These could be people like customer support specialists who will end up shouldering a lot of the burden of bad product design, or they could be silent partners who might weigh in heavily at the tail end of a project.
Potential UX stakeholders
Have organizational influence
Make decisions about time, money, and resources. Are involved in the UX and product design process
Have information relevant to your project. Will be expected to act on research insights
Key scenarios analysis. Как начать
- Взять все юзер роли
- Выписать список задач, которые все эти пользователи могут делать в интерфейсе (это все юзкейсы сценариев)
- После этого начинаем прорабатывать все по одному по схеме.
- Рядом с каждым проработанным сценарием прорабатываем ветку userflowСкорее всего, абсолютно все предусмотреть не удастся. Корнер кейсы, которые всплывут потом, пока не учитываем. Поэтому мы и называем эту методику именно key сценарии
Что такое юзкейс
Решение конкретной задачи через интерфейс (то, что пользователь хочет или может делать в интерфейсе)
Схема проработки каждого Key scenario analysis
- Who am I?
- What is my goal?
- Why it’s my goal?
- What’s the trigger?
- What actions do I need to do to achieve the goal?
- What is my desired outcome?На выходе получаем job story:
When situation, I want to motivation, so that I expected outcome
Структура Job story
When situation, I want to motivation, so that I expected outcome
Здесь цель (motivation) — четкая примитивная цель в рамках интерфейса (не абстрактная, не результат). На примере инстаграмма — запостить фотографию. На примере обучающей платформы — написать ментору.
Пример:
When I faced some problems with the platform or can’t join the private lesson, I want to text my mentor, So that I can inform him and get some help as soon as possible.
Пример заполненного Key scenario analysis
Job story:
When I faced some problems with the platform or can’t join the private lesson, I want to text my mentor, So that I can inform him and get some help as soon as possible.
Как выглядит Userflow, который мы создаем рядом с разобранным Key scenario analysis
Этапы Discovery по Паше Билащуку
Основное: * Business model canvas * Persona или protopersona * Value proposition canvas * Key scenario analysis (юзкейсы) + userflow Опционально: * Информационная архитектура (appmap/sitemap — страницы, разделы, меню и его подразделы ) * Competitors research * Visual research Не только наша зона ответственности: * [создание беклога и эстимейты]
Определение visual approach
Делается в конце дискавери или в начале работы.
- Следование гайдлайнам и существующим продуктам — самый простой путь
- Чтобы сделать что-то новое, нужен еще и визуальный research
Value proposition canvas
- Матчится ли то, что мы знаем о бизнесе с тем, что мы знаем о пользователе
Создание протоперсоны со стейкхолдерами
На воркшопе создаем shared документ с лэйаутом для заполнения персоны каждым стейкхолдером. Даем в тишине мин 20 на заполнение. А потом каждый презентует (желательно от первого лица). На основе повторяющихся элементов создается финальная персона.
Чим відрізняється персона від протоперсони
Протоперсона базується на припущеннях команди й ще не підтверджена інформацією від ЦА. Персона – це набір уже провалідованої інформації про ЦА. Тобто, з часом під час досліджень побудована на припущеннях протоперсона перетворюється на персону.
Types of customers jobs (value proposition canvas)
- Functional
- Social
- Emotional / personal
Supportive
Functional customer jobs
To perform a specific task or solve a specific problem, for example, mow the lawn, eat healthy as a consumer, write a report, or help clients as
a professional.
Social customer jobs
To look good or gain power or status. These jobs describe how custom- ers want to be perceived by others, for example, look trendy as a consumer or be perceived as competent as a professional.
Personal/emotional customer jobs
To seek a specific emotional state, such as feeling good or secure, for example, seeking peace of mind regarding one’s invest- ments as a consumer or achieving the feeling of job security at one’s workplace.
What are supporting customer jobs
- Buyer of value
- Co-creator of value
- Transferrer of value
Buyer of value supportive job
jobs related to buying value, such as comparing offers, deciding which products to buy, standing in a checkout line, completing a purchase, or taking delivery of
a product or service.
Co-creator of value supportive job
jobs related to cocreating value with your organization, such as posting product reviews and feedback or even participating in the design of a product or service.
Transferrer of value supportive job
jobs related to the end of a value proposition’s life cycle, such as canceling a subscription, disposing of a product, transferring it to others, or reselling it.
Customer pains
Pains describe anything that annoys your customers before, during, and after trying to get a job 14 done or simply prevents them from getting a job done. Pains also describe risks, that is, potential bad outcomes, related to getting a job done badly or not at all.
Types of customer pains
- Undesired outcomes
- Obstacles
- Risks
Questions to find our customer pains
- How do your customers define too costly? Takes a lot of time, costs too much money, or requires substantial efforts?
- What makes your customers feel bad? What are their frustrations, annoyances, or things
that give them a headache?
3.
Types of customer gains
- Required
- Expected
- Desired
- Unexpected
Questions to find out customers gains
- Which savings would make your customers happy? Which savings in terms of time, money, and effort would they value?
- What quality levels do they expect, and what 17 would they wish for more or less of?
- How do current value propositions delight your customers? Which specific features do they enjoy? What performance and quality do they expect?
- What would make your customers’ jobs or lives easier? Could there be a flatter learning curve, more services, or lower costs of ownership?
- What positive social consequences do your customers desire? What makes them look good? What increases their power or their status?
- What are customers looking for most? Are they searching for good design, guarantees, specific or more features?
- What do customers dream about? What do they aspire to achieve, or what would be a big relief to them?
- How do your customers measure success and failure? How do they gauge performance or cost?
- What would increase your customers’ likelihood of adopting a value proposition? Do they desire lower cost, less investment, lower risk, or better quality?
Common mistakes for mapping jobs, pains and gains
- Mixing some customer segments in one profile
- Mixing jobs and outcomes
- Focusing on functional jobs only and forgetting about social and emotional jobs
- Listing jobs, pains, and gains with your value proposition in mind
- Identifying too few jobs, pains, and gains
- Being too vague in descriptions of pains and gains
The difference between jobs and gains in value proposition canvas
Jobs are the tasks customers are trying to perform, the problems they are trying to solve, or the needs they are trying to satisfy.
Gains are the concrete outcomes they want to achieve—or avoid and eliminate in the case of pains.
What is the difference between persona’s goals and wants?
Goals are desired outcomes and wants are what will make the process of attaining goals more comfortable
What is the difference between persona’s fears and pains?
Pain is existing or recurrent inconvenience or problem. Fear might not happen at all or be very unlikely
What is needs in persona’s framework?
It is something what does the person needs to achieve goal
What is a design process
The Design Process is an approach for breaking down a large project into manageable chunks.
Stages of scientific method
- Make an observation
- Ask a question
- Form a hypothesis, or testable explanation
- Make a prediction based on the hypothesis
- Test the prediction
- Repeat/Iterate
Stages of any design process (design thinking)
- Empathize
- Define
- Ideate
- Prototype
- Test
On which stages of design process mistakes are normal
Here are normal: * Empathize * Define * Ideate Here shows not good level of professionalism: * Prototype * Test
Круги Эйлера
геометрическая схема, с помощью которой можно изобразить отношения между подмножествами, для наглядного представления
Human Centered Design
Табуретка Нормана
About the first stage of the design process
Empathise (with people)
Aim of the stage
• Understand who are the people you are designing for;• Empathise to them and understand their goals, pains, gains and way of thinking/living.
What to do at this stage• Find and meet your TA;• Talking to people you are designing for. Conduct interviews;
Empathy maps, interview repots, personas…
with people
About second stage of design process
Define (a problem)
Aim of the stage
• Extract, sort and prioritise people’s pains from the previous step; • Identity problems to solve.
What to do at this stage• Convert pains and fears into design challenge.
How might we
About third stage of design process
Ideate (a solution)
Aim of the stage
• Based on the problem/s identified on the previous stage come up with testable solution;
What to do at this stage• Create a considerable number of potential solutions; • Vote and rate it. Determine the best solution/s to prototype.
Brainstorm sessions, conceptualising, user flows, IA, CJM…
About forth stage of design process
Prototype (a solution)
Aim of the stage
• Convert ideas into something testable.
What to do at this stage• Choose the best type of the prototype for the certain ideas• Build a prototype.
Diagrams, paper prototypes, wireframes…
About 5th stage of design process
Test (a solution)
Aim of the stage
• Check whether a solution works or not; • If works, how well?• If not, why?
What to do at this stage• Organise a test sessions;• Gather and analyse feedback.• Go back to the stage where problems occurs.
Test report, test analysis…
Which company department can give some information about persona’s
Marketing department
What is maximum reasonable amount of Personas?
Around 10. If more, then very hard to keep them up to date
What is the difference between goals and wants in persona’s profile
Ціль — це те, без чого в продукті не буде сенсу.
Wants — nice to have. Це параметри, які дають нам змогу розвивати бізнес далі
What is the difference between pain points and fears?
Pain point is a recurring or permanent problem.
Fear is something that has some probability
What is the difference between primary and secondary research
Primary research
- качественные исследования (глубинные интервью, юзабилити тестирование и тд)
- количественные исследования (аналитика, аб тестирование, опросы и тп)Secondary research
- анализ рынка и конкурентов
- анализ продукта и личный опыт использования
- анализ данных из открытых источников
What is primary research
- качественные исследования (глубинные интервью, юзабилити тестирование и тд)
- количественные исследования (аналитика, аб тестирование, опросы и тп)
What is secondary research
- анализ рынка и конкурентов
- анализ продукта и личный опыт использования
- анализ данных из открытых источников
Типы конкурентов
- Direct competitors
- Secondary (indirect) competitors
- Indirect (replacement) competitors
Direct competitors
They do the same job in the same way (McDonalds and BUrger King)
Secondary competitors
They do the same job in a different way (Skype vs Business class travel)
Indirect (replacement) competitors
They do a different job with a conflicting outcome (McDonalds and Weight Watchers, coffee and kale)
Где искать цифровых конкурентов
- Product hunt
* Alternative tool
Points of parity
Обязательный функционал, который обязан быть в конкурирующих продуктах. Без него пользователи не будут воспринимать этот продукт
Анализ ценностного предложения
- Points of parity
* Points of difference
Points of difference
То, что выделяет продукт на фоне остальных на рынке
A typical competitor analysis matrix consists of
- Competitor’s name
- URL
- About (main value proposition) from Crunchbase
- The number of users/downloads (monthly)
- The main features being offered
- Strengths (optional)
- Weaknesses/areas of improvement (optional)
- Monetization model / revenue streams
- Notes
Questions for in-depth competitors analysis
- what are competitors doing right, and what they might be struggling with, leaving opportunities available.
- What is the precise nature of the problem they’re solving?
- What are the specific features of a competitor’s product designed to solve the problem?
- How well are they solving a problem? Could it be improved?
- Examine how easy or frustrating it is to complete a task.
Competitor functional analysis matrix
Feature list and if every competitor has the specific feature from the list
Amount of competitors to analyze deeply
2-4
And the more the better for brief general analysis (around 10)
Possible deliverables after competitors research
- The goal of research
- The list of direct (among them key), indirect, and replacement competitors
- The general competitors analysis matrix
- The functional competitors analysis matrix
- A link to the screenshots (Miro/Figma boards)
- Notable findings
High-level take-aways/conclusions
An extensions for taking webpage screenshots and screencasts
- Awesome screenshot https://chrome.google.com/webstore/detail/awesome-screenshot-and-sc/nlipoenfbbikpbjkfpfillcgkoblgpmj
2.
Before the competitor research
- To understand exactly what information you are looking for
Goals of competitor analysis
Why are you doing this competitor analysis? What do you hope to achieve?
- To see how many competitors are there
- To know the strengths and weaknesses of your competition
- To understand own areas to growth
The tools for analyzing websites traffic
- Similar web
- Compete.com
- Quantcast
- Alexa
Tools for iPhone stats
- App Annie
- AppFigures
- Mopapp
- Distimo
What is UGC
User-generated content
Types of personas
- Marketing
- Design
- Protopersona
Marketing personas
Some basic characteristics such demography, habits, preferences etc. show facts but not explain ‘why’
Design personas
Focus on user goals, current behavior, and pain points as opposed to their buying or media preferences and behaviors. They are based on field research and real people. They tell a story and describe why people do what they do in attempt to help everyone involved in designing and building a product or service understand, relate to, and remember the end user throughout the entire product development process.
Why it is good to have a primary persona
When you design for your primary persona, you end up delighting your primary persona and satisfying your secondary persona(s). If you design for everyone, you delight no one. That is the recipe for a mediocre product.
Reasons to make personas
- Personas can be used to validate or disprove design decisions.
- Personas allow us to set and prioritize feature requests.
- Personas are an inspiration in ideation.
Usability heuristics
- Visibility of system status
- Match between system and the real world
- User control and freedom
- Consistency and standards
- Error prevention
- Recognition rather than recall
- Flexibility and efficiency of use
- Aesthetic and minimalist design
- Help users recognize, diagnose, and recover from errors
- Help and documentation
Виды исследований
Количественные (что пользователи делают) и качественные (почему)
Какой тип исследований важнее дизайнеру
Качественные исследования (не количественные). Потому что там находятся инсайты (но лучше комбинировать)
Что такое парситипативный дизайн
Когда пользователь создаёт что-то вместе с нами
What do we need before interview
- Research questions (why we can’t start working on project right now)
What is result of the stage of user interviews?
Clearly formulated answers to research questions
Where to find respondents for the interviews
- Application / webpage itself
- Own Facebook page
- Custom Facebook groups
When to stop conduct interviews
When you have reached the point of saturation — with each next interview you don’t get any new information
Про интро на пользовательских интервью
Респондентам лучше не знать в деталях, что мы исследуем. Минимум информации. Иначе они начинают размышлять, что нам нужно сказать, а что по их мнению не будет полезным
Цель пользовательских интервью
Собирать истории. Не должно быть блиц-опроса
Вредные привычки на интервью
Избавиться от привычки после вопроса добавлять варианты (где вы создаёте накладные — на ходу, дома и тд). От привычки спрашивать про обычно, как правило и тд. Это плохие вопросы, потому что ответы сильно рационализированы. Надо про конкретные случаи.
Service designer
Обеспечивает бесшовное прохождение всего сценария, независимо от точек контакта
Userflow
Отображает все возможные разветвления клиентских сценариев и того, как сервис реагирует на разные действия пользователя
UX assessment
Research and testing without users
Research without users methods
- Cognitive walkthrough
- Heuristic evaluation
- GOMS
- Databases (SQL)
Cognitive walkthrough
A method which allows to understand learnability and ease-of-use of the interface.
We take a colleague and see how he interacts with the interface the first time. It helps to find the most rough mistakes in the interface before user testing.
Better with triangulation (some people who are nearby)
❗️Is not enough alone, needs other methods to complement
Preparing to cognitive walkthrough
- Prepare description of the product for the person
- To understand persona or protopersona and to choose more appropriate colleague
- To give scenario
During cognitive walkthrough
We ask the person to comment all his actions
Before every action concentrate on: Does the user know what action to take to proceed towards the goal?
After every action concentrate on:
Did the user understand if that step succeeded?
Cognitive walkthrough questions
Before each action
- Does the user try to achieve the right effect?
(Например, в Airbnb надо сначала указать место, а он сразу ищет фильтр по цене) - Does the user notice that the correct action is available?
(Например он знает, что надо ввести место, но не может найти поле для ввода) - Does the user associates the correct action with the effect trying to be achieved?
(Имеет ли он представление о том, что должно произойти дальше?)
И после действия:
4. If the correct action was performed does the user understand that he achieved progress towards his goal?
(Он увидел варианты жилья — понял ли он, что это такое)
Example of cognitive walkthrough deliverable
Simplified question for cognitive walkthrough
Before:
1. To make progress toward their goal, does the user know what action to take next?
After
2. Based on the system response, does the user see that progress is being made towards their goal completion?
Specifics of cognitive walkthrough
Better if it is not finalized interface (with hovers/animation etc) but more rough wireframes. Because interface should be clear on that stage. Hovers and animations then will make it just a little more understandable
For checking what type of users we usually use cognitive walkthrough?
For new users which see interface the first time
Heuristics evaluation
Checking the product by all heuristics of Nilsen
❗️Requires triangulation (one person — up to 35%, 3-5 people — 50-75%. 15-25 people — up to 95%)
Process of triangulated heuristic evaluation
3-5 experts do heuristic evaluation separately, without discussing which each other. Discussion only after evaluation was finished
Severity of usability problems
0 = I do not agree that it is a usability problem 1 = only cosmetic problem: no need to fix if the project does not have enough time for this 2 = minor usability issue: fixing this should be planned with low priority 3 = big usability problem: it’s important to fix it, so you should give high priority 4 = usability catastrophe: you need to fix it before the product can be released
Factors of severity problems
- Frequency of the problem — is it common or rare
- The impact of the problem — is it easy or difficult to overcome the problem
- Persistence of the problem — is this a one time problem that users can overcome when they already know about it, or will users constantly worry about the problem?
Heuristic evaluation report structure
- Title, Presentation of experts
- Outputs (first)
- Summary table
- Methodology presentation (in general, heuristic description, severity description, what impacts severity). Why do we need this methodology in our case
- Screenshots and explanations (one page — one problem)
Structure of heuristic problem description
Better number of severity
Example of table for heuristic analysis
What is GOMS
It is a family of predictive models of human performance that can be used to improve the efficiency of human-computer interaction by identifying and eliminating unnecessary user actions
GOMS acronym
Goals — a state to be achieved
Operators — elementary motor or cognitive gestures
Methods — procedure for accomplishing goal
Selection rules — how do we choose between available methods
GOMS scheme
Cons of GOMS
- We can not predict the system response
- Fits only experienced users
- Not all operators are described
Pros of GOMS
- Applicable to any interface (more complicated with AR/voice)
- Can be used at any design stage
- One of the ways to calculate the ideal time
- Instrument to predict user errors
Conception of GOMS
The task execution time is the sum of time periods of all gestures from which this task consists
Arrangements of the operator M (mental) at GOMS
- Insert M before all K (keystrokes) and in front of all P (pointings). Do not place M in front of any P that points to any arguments of those commands
- Removing the expected operators M (if we were thinking before pointing we should paste M before clicking)
- Removal of M inside cognitive units (4656)
- Delete M before successive delimiters (, (), —)
- Removing M which are overloaded with R (system response)
SQL
Structured query language
Why SQL
- For recruiting for interviews and surveys
Analytics systems
- Grafana
- Dynatrace
- Datadog
- AppDynamics
- Splunk Signal FX
- Sumo logic
- Logic monitor
- Prometheus
- Centreon
GOMS KLM
Keystroke level model
Натиснення клавіші клавіатури
Альтернативи для GOMS KLM
TLM (tapstroke) or FLM (fingerstroke)
- Visibility of system status
The system should always keep users informed about what is going on, through appropriate feedback within reasonable time.
Communicating the current state allows users to feel in control of the system, take appropriate actions to reach their goal, and ultimately trust the brand.
Understanding that a system is working on the request. Open and continuous communication.
Examples: persistent signs like wifi, battery indicators, car speed, or response to user’s action like elevator button and floor indication, or web button color changing. ‘You are here’ maps. How many items left or how many money more you should spend for the free delivery, information about of out of stock. Multi step processes.
- Match Between the System and the Real World
The design should speak the users’ language. Use words, phrases, and concepts familiar to the user, rather than internal jargon. Follow real-world conventions, making information appear in a natural and logical order.
Systems should speak the users’ language with familiar words, phrases, and concepts rather than system-oriented terms. Interfaces that follow real-world conventions and make information appear in a natural and logical order demonstrate empathy and acknowledgement for users.
Examples: skeuomorphic web design, increase volume above the decrease volume, metaphors (highlighting text in articles with the same green color as marker has), stovetop controls, shopping cart icon
- User Control and Freedom
Users often make mistakes or change their minds. Allow them to exit a flow or undo their last action and go back to the system’s previous state.
Users often choose system functions by mistake and will need a clearly marked “emergency exit” to leave the unwanted state without having to go through an extended dialogue. Support undo and redo.
Users should be able to quickly correct mistakes or backtrack on choices made.
There are several UI controls that typically allow people to go back to the previous state of the system:
— A Back link which returns users to a previous page or screen
— A Cancel link which allows the user to quit a task or multi-step process
— A Close link which allows users to close a new view
— An Undo option (and a corresponding redo option) to allow users to backtrack on a change to a UI element
Examples: opening links in new tabs breaks this severity
- Consistency and standards
Users should not have to wonder whether different words, situations, or actions mean the same thing. Follow platform and industry conventions.
Tip: Improve learnability by maintaining both types of consistency: internal and external.
∙ Tip: Maintain consistency within a single product or a family of products (internal consistency).
∙ Tip: Follow established industry conventions (external consistency).
Examples: blue hyperlinks, logo as a home button
- Preventing User Errors
Users are often distracted from the task at hand, so prevent unconscious errors by offering suggestions, utilizing constraints, and being flexible.
Types of user errors
Slips and mistakes
Slips errors
Slips occur when users intend to perform one action, but end up doing another (often similar) action. For example, typing an “i” instead of an “o” counts as a slip; accidentally putting liquid hand soap on one’s toothbrush instead of toothpaste is also a slip. Slips are typically made when users are on autopilot, and when they do not fully devote their attention resources to the task at hand.
Mistakes errors
Mistakes are made when users have goals that are inappropriate for the current problem or task; even if they take the right steps to complete their goals, the steps will result in an error. For example, if I misunderstood the meaning of the oil-pressure warning light in my car, and thought it was the tire-pressure monitor, no matter how carefully I added air to my tires, it would not fix the issue with my oil pressure. This would be a mistake, since the goal that I was attempting to accomplish was inappropriate for the situation, even though I made no errors in executing my plan. Mistakes are conscious errors, and often (though not exclusively) arise when a user has incomplete or incorrect information about the task, and develops a mental model that doesn’t match how the interface actually works.
General Guidelines for Preventing Slips
- Includes helpful constraints
(You cannot choose arrival date before the departure one) - Offer suggestions
(Search suggestions) - Choose Good Defaults
(In a reminder app Today, Tomorrow, in one hour) - Use Forgiving Formatting
(Smaller digits chunks that are easier to scan (777) 555-1212)
Preventing User Errors: Avoiding Conscious Mistakes
Mistakes occur when a user has developed a mental model of the interface that isn’t correct, and forms a goal that doesn’t suit the situation well.
- Follow design conventions
- Communicate Affordances (shadows outside the buttons and inside fields)
- Preview results
Preventing Both Mistakes and Slips
- Remove Memory Burdens (Whenever possible, remove conditions that require users to keep information in their own memory while they move from one step to another in complex, multistep procedures. )
- Confirm Before Destructive Actions
- Support Undo
- Recognition rather than recall
Showing users things they can recognize improves usability over needing to recall items from scratch because the extra context helps users retrieve information from memory.
Minimize the user’s memory load by making elements, actions, and options visible. The user should not have to remember information from one part of the interface to another. Information required to use the design should be visible or easily retrievable when needed.
Examples: login&password at a website is a recall, menu is a recognition; providing past histories in search engines is a recognition, Amazon list of things you have watched before is recognition
- Flexibility and efficiency of use
Shortcuts— unseen by the novice user — speed up the interaction for the expert users such that the system can cater to both inexperienced and experienced users.
The 7th usability heuristic (flexibility and efficiency of use) is about allowing users to approach tasks in a variety of ways. New users may require guidance in performing their tasks, whereas experienced users can take advantage of accelerators and other secondary features designed to speed up commonly performed actions.
There are two different aspects to making a flexible and efficient system:
- Multiple methods to accomplish the same task according to one’s preferences
- Accelerators that don’t slow down inexperienced users, but speed up advanced users
Examples: tap twice to like in instagram, cmd C cmd V
- Aesthetic and minimalist design
Interfaces should not contain information which is irrelevant or rarely needed. Every extra unit of information in an interface competes with the relevant units of information and diminishes their relative visibility.
- Keep the content and visual design of UI focus on the essentials.
- Don’t let unnecessary elements distract users from the information they really need.
- Prioritize the content and features to support primary goals.
Examples: ornate and minimal teapot
- Help users recognize, diagnose, and recover from errors
Error messages should be expressed in plain language (no error codes), precisely indicate the problem, and constructively suggest a solution.
— Use traditional error message visuals, like bold, red text.
— Tell users what went wrong in language they will understand — avoid technical jargon.
— Offer users a solution, like a shortcut that can solve the error immediately.
- Help and documentation
It’s best if the system doesn’t need any additional explanation. However, it may be necessary to provide documentation to help users understand how to complete their tasks.
Специфіка питань про те, які ще є проблеми
Це не те щоб погане питання, але краще робити такі висновки через історії респондента
Коли проводити інтерв’ю
- Коли виникають питання «чому»
- Коли потрібно скласти реальну картину, як люди користуються сервісом (але щоб отримати дозвол і тп, краще мати додаткові питання)
Сервіс сафарі
Користування сервісами
How many times per survey does it make sense to use open questions?
Up to 2, in very seldom cases 3.
Never required.
(What is your main objective of this course)
Population in statistics
The total number of people in the group you are trying to study. If it’s all people of US, then it is 317 million. If you study people of your company then it will be the headcount of the company.
Margin of error (похибка)
How sure you need to be that the answers reflect the views of your population.
The larger is the margin of error, the less confidence one should have that a poll result would reflect the result of the survey for the whole population.
Confidence level
How sure you need to be that the sample accurately samples your population
p-value in statistics
Стандартное отклонение (среднеквадратическое). Standard deviation
Мера разброса значений от среднего. Чем меньше стандартное отклонение, тем более кучно все значения расположены вокруг конкретного значения .
Z score
Shows the difference of standard deviations between sample data and raw data.
(Average height is 173 cm, my height is 185 cm, and how many average deviations is the difference )
What do you need to calculate a sample size for survey
- Population amount
- Confidence level (usually 95%)
- Margin of error (usually 5%)
Typical response rate for the survey
Optimistic — 20-30%
Realistic — 10-15%
Why we might need a UX survey?
- Collecting quantitative data to back-up your qualitative research findings: unfortunately business leaders tend to believe you more, when you can show them big numbers. So after doing your qualitative research (user tests, field research, interviews) on a small scale, you can validate your findings with a survey.
- Recruiting test participants for moderated user tests with experienced users
- The main advantage of a UX survey is that you can ask real visitors during their actual visit. People forget things easily and later they can give different explanations. With a survey in your product you can gather valuable information about who they are and why they’re there.
Что насчёт персон
Это неплохо, но в первую очередь это инструмент эмпатии. Но для серьёзных исследовательских решений больше подходят работы.
Types of jobs to be done
1 Functional
- Social
- Emotional
Интро на интервью
Обязательно сделать, чтобы расслабить человека. Но не сильно детальное, чтобы не раскрыть все карты. Иначе он начнёт фильтровать мысленно, что по его мнению нам нужно сказать, а что нет. Нормальная практика поговорить сначала ни о чем, на отвлечённые темы, и плавно скатиться в интервью.
Что нас не должно интересовать на интервью
- Мнение респондента о нашем продукте, что он о нем думает
- То, как человек ~обычно~ что-то делает
- Как ~бы~ он поступил в определенной ситуации
Что нас интересует на интервью
Реальные факты и истории из жизни человека
Метафора для интервью
Поиск ценного металла под землей. Задаём общие вопросы, сканируем в целом. Если натолкнулись на что-то интересное, то углубляемся. Точечные вопросы — то же самое, что копать лопатой в конкретных местах с надеждой, что повезёт. Но мы должны знать, что именно ищем (какой металл)
The Forces of progress
Pushes, pulls (increase progress)
Anxieties, habits (decrease progress)
What is NPS?
Net Promoter Score
Widely used market research metric that typically takes the form of a single survey question asking respondents to rate the likelihood that they would recommend a company, product, or a service to a friend or colleague.
SUM method
Completion rate metric
Binary metric, 1 or 0.
(Can be 0.5 in very complicated cases — to book a flight, to the city, near the window, with a lunch. But maybe better to divide big tasks into some smaller ones?)
Measuring usability metrics
- Completion rate
Average completion rate
78%
If the problem breaks some heuristics what will be with severity
Each heuristic will have its severity
How often do we need interviews
- When we have why questions
2. When we need a realistic view of using our product
SEQ
Single Ease Question (SEQ) is a 7-point rating scale to assess how difficult users find a task.
After users attempt a task, ask them this simple question: Overall, how difficult or easy was the task to complete?
SEQ exactly
Overall, how difficult or easy was the task to complete?
What if SEQ less than 5?
We briefly ask why
Specification limit for the metric ‘time on task’ (more than 25 respondents)
Median
Specification limit for the metric ‘time on task’ (less than 25 respondents)
Average geometric
An interesting specification limit for the metric ‘time on task’
Median or geometric average only of those people who filed positive satisfaction test (more than 4 or 5)
Specification limit in measuring usability
Threshold value which defines ‘goodness and evil’
Tasks order for measuring usability
Not to put nearly similar tasks one by one. Otherwise second task will be much quicker because of learnability
Formula for counting error rate
Specification limit for SUS
68%
The most popular method in the world to measure usability
SUS (system usability scale), 10 statements
Amount of people for SUS
5 people for 50%
10 people for 75%
17 people for 95%
SUM measurement
Single Usability Metric
What metrics SUM consists of
Completion, time, errors, satisfaction
Usability
Competition rate usability metric
Completion rates are the fundamental usability metric.
A binary measure of pass and fail (coded as 1 or 0) provides a simple metric of success. If users cannot complete a task, not much else matters with respect to usability or utility.
What is an average completion rate?
78%
SEQ
Single Ease Question
Overall, how difficult or easy did you find this task?
How to count error rate while measuring usability
What is SUS
The System Usability Scale (SUS) is a ten-item questionnaire administered to users for measuring the perceived ease of use of software, hardware, cell phones and websites.
SUS questions
- I think that I would like to use this system frequently.
- I found the system unnecessarily complex.
- I thought the system was easy to use.
- I found the system very cumbersome to use.
- I felt very confident using the system.
- I think that I would need the support of a technical person to be able to use this system.
- I found the various functions in this system were well integrated.
- I thought there was too much inconsistency in this system.
- I would imagine that most people would learn to use this system very quickly.
- I needed to learn a lot of things before I could get going with this system
SUS answers
1-5,
strongly disagree — strongly agree
Calculating SUS answers
- For odd items: subtract one from the user response.
- For even-numbered items: subtract the user responses from 5
- This scales all values from 0 to 4 (with four being the most positive response).
- Add up the converted responses for each user and multiply that total by 2.5. This converts the range of possible values from 0 to 100 instead of from 0 to 40.
SUS scores
SUS scores can also be translated into letter grades, which may be helpful for communicating results to stakeholders.
The scale also indicates a SUS score of 78.9 or above would constitute an “A-” or above, while a SUS score of 51.6 or below would constitute an “F”.
Learnability and usability at SUS
While SUS was only intended to measure perceived ease-of-use (a single dimension), recent research shows that it provides a global measure of system satisfaction and sub-scales of usability and learnability.
Items 4 and 10 provide the learnability dimension and the other 8 items provide the usability dimension. This means you can track and report on both subscales and the global SUS score.
SUS bias
Measuring U found that repeat users rated the websites with SUS scores 11% higher than those of first-time users for websites. The same pattern held for software. Users with five or more years of experience with software-generated SUS scores 11% higher than users with 0-3 years of experience.
Sample size for SUS
What is usability
Usability is the intersection of effectiveness, efficiency and satisfaction
What efficiency can mean
Time on task
What effectiveness can mean
Completion rate and errors rate
Light version of SUM
Completion, time and satisfaction
Without errors
Time specification limit
- Time of completing product
- Time from the most satisfied users
- Time of expert user
- Last resort — multiplying KLM or the fastest result with any multiplier
Weakness of bootstrapping method of taking specification time limit
The most obvious shortcoming to this method is that it relies on the sample of data to build a specification limit for the sample. This means the spec limit is product dependent [1 p.195]. That is, while users may be providing a sufficient satisfaction level after completing the tasks, the interface may still be forcing users to take too long to complete a task.
Perceived usefulness of mobile apps
For measuring perceived usefulness of mobile apps, we use the item “The application’s capabilities meet my requirements” which has a five-point rating scale (Measuring U).
For measuring website usability
For measuring website usability, we use the 13 item SUPR-Q. Four of the items can generate a reliable SUS equivalent score. The other items provide measures of credibility/trust, appearance and loyalty.
Learnability
Learnability considers how easy it is for users to accomplish a task the first time they encounter the interface and how many repetitions it takes for them to become efficient at that task.
Learning curve
With the data from the learning curve, we can identify how long it takes users to reach saturation — a plateau in our charted data which tells us that users have learned the interface as much as possible.
Aspects of learnability
First-use learnability: How easy is it to use the design the first time you try? This aspect of learnability is of interest to those users who will only perform the task once. These users won’t progress up the learning curve, so they don’t care how it looks.
Steepness of the learning curve: How quickly do people get better with repeated use of the design? This facet of learnability is particularly important for users who will use the design multiple times, even though they won’t use it excessively. If people feel that they are progressing and getting better and better at using your system, they’ll be motivated to stick with it. (And conversely, if people feel that it’s hardly getting better, no matter how hard they try, they’ll start looking for a better solution.)
Efficiency of the ultimate plateau: How high is the productivity that users can reach with this interface, once they have fully learned how to use it? This aspect is particularly important for people with a frequent and long-lasting need to use the system — for example, when it’s the main tool for important everyday tasks.
Learnability testing
Gather participants with little to no experience using the system that they’ll be testing.
When applicable, recruit participants with no similar system experience and participants with some similar-system experience, and plan to compare corresponding data from both groups.
As for any quantitative study, we recommend that you recruit a fairly large number of participants (usually at least 30–40).
Most popular metric of learnability
- Time on task
Sometimes Number of errors metric also can work.
Trials at learnability
The next step consists of deciding how often to collect these metrics — each instance of data collection is known as a trial.
Remember, we’re trying to plot this metric over time, so we need to have the same participants complete the same task multiple times. We recommend you repeat the trials until a plateau is reached. A flattened curve indicates our participants have learned the system (specific to this task) as much as possible.
Amount of trials for learnability study
As a starting point, consider 5–10 trials but when in doubt, plan for more trials than you think you need, for two reasons: (1) you want to be sure that you’ve reached stable performance and (2) once you’ve reached a point of stable performance, it’s generally easier to cancel usability sessions than to schedule more.
Що таке parking lot
Це місце, куди під час брейнсторминга ми складаємо нерелевантні ідеі
How to count time of task completion when measuring usability?
Use the geometric mean for the small samples (less than 25) and median for large samples. Never use arithmetic mean.
Bootstrapping a specification limit
Derived from the phrase “pulling oneself up by one’s bootstraps” since we are building the spec limit from the very data which we will then apply the spec limit to.
To identify the maximum acceptable time (spec limit) using the bootstrap method
- Remove times from failed tasks.
- Remove times where satisfaction scores are less than 4 (5 point scale).
- Find the 95th percentile of the remaining times to arrive at the specification limit.
How many users do you need to ensure that task completion rate at least 70%
8 users. If all 8 users complete the task, you can conclude with 95% confidence that the completion rate exceeds 70%
How many users do you need to ensure that the completion level is at least 90%
A minimum of 30 users
If you have 8 users, what maximum completion rate you can measure
70%. If all 8 users complete the task, you can conclude with 95% confidence that the completion rate exceeds 70%
Who has described slips
Don Norman has explained slips in Chapter 5 of The Design of everyday things
What is a mistake
When the goal is wrong it’s a mistake, even if that goal was accomplished.
Should we record every time an error occurs even if it is the same error by the same user on the same task
Yes. ‘I’ve seen the same user try unsuccessfully to click on the same heading that wasn’t clickable 5 times over a 2 minute period.
The user was confused about the navigation and really wanted that heading as a way to reorient themselves. Even though it was the same error, seeing 5 errors versus 1 error better describes the experience (which was poor).’
Alternative way to count errors while measuring usability
If there were errors then 1, if there were no, then 0
How we count errors with errors opportunities?
For example, if a task has 5 opportunities for an error and 10 users attempt the task there are 50 opportunities. If you observe 5 errors across the users the error rate is 5/50 = 10%.
Specification limit
Acceptable level of quality
6 sigma
Advanced statistical analysis can push efficiencies closer to only 3.4 defects per million opportunities or 6-sigma.
What is z-score
A normal deviate for the quality level
What is HEP
Human error probability
Determining HEP
The general approach for determining HEP is to divide human behavior in a system into small behavioral units, find data for these subdivisions and then recombine them to estimate the error probabilities for the task.
Example HEP
Example Task: Add a new customer record to the Customer List
Opportunity 1: Locate access point for adding a new customer record and launch
new customer record form
Opportunity 2: Enter new customer record ID information
Opportunity 3: Enter account opening balance information correctly
Opportunity 4: Enter customer address information
Opportunity 5: Enter customer contact information
Opportunity 6: Submit record successfully
For example, Opportunity #1 can have the following error instances associated with it:
User can’t find access point;
User launches an existing customer record instead of adding a new one;
User launches a new vendor record instead of a new customer record.
How to calculate defective rate (errors)
Say 20 participants completed a task that has 6 opportunities for error. A total of 36 errors were identified across all participants.
Total Defects / Total Opportunities = Defective Rate
36 / (6 x 20) = .3 = 30% Defective
How to transfer defective rate into Sigma
The following steps then calculate the process sigma:
1 - .30 = .70 = 70% Quality Level
The corresponding z-score to 70% on a standardized z-table is .524
.524 + 1.5 sigma shift = 2.02 sigma
List of soft skills
- Communication
- Self-control
- Productivity (ability to make jobs in some not too big terms)
- Wiseness (intuition)
- Perception
- Influence
Application for the iPhone for finding fonts by photo
What Font
Navigation testing
Card sorting / Tree testing
Progress bar in the presentation
Small sample sizes and problems detecting
The problem with small sample sizes is that we’re only able to reliably detect major issues (issues that affect a lot of users). The good news about small sample sizes is that we’re detecting issues that matter! So when you see a problem occur repeatedly with a small sample test, it means a problem is probably affecting a lot of your users. Small sample sizes don’t do a good job of finding problems that only affect a small portion of the users.
Why do we need confidence intervals
Because we almost always sample a fraction of the users from a larger population, there is uncertainty in our estimates.
Sample sizes and confidence intervals
Smaller sample sizes generate wider intervals.
Measurements and observations
It’s not choosing between the measurements and observations. It’s choosing the right measures for your observations.
Task metrics VS study metrics
Task metrics are making some specific jobs and study metrics are driven by attitudes.
Task metrics
- quantify the behavior of the task experience
- tend to be more diagnostic
Study metrics
- provide a more holistic attitude towards the product
- are likely more generalizing than the task metrics
How can be effectiveness measured
- Completion
- Findability
- Errors
How can be efficiency measured
- Time
- Clicks
How to choose 0 or 1 when measuring completing tasks
Measuring U doesn’t recommend using 0.5.
Better divide the task in advance for necessary and optional sub tasks. And based on them to put 1 or 0.
Types of time measuring
- time to complete
- time on task
- time until failure
NPS explanation
UX is made of
Attitudes and action metrics
Workshops facilitation basics
- Perfect timing (20 mins is not more than 20 mins). Minute by minute plan
- People. 5-7 people, max 9. Only those, who are truly connected
Ground rules workshop
Stone rules which are working exactly in this room
Examples
- No phones checking in the room
- Precise timing
To suggest adding any ground rules.
To take signatures or to ask everybody agree aloud with these rules
How many people do you need for the moderated testing
4-6 optimal
3 okay if you test on regular base
> 8 too much
< 3 too less
How many users do you need on moderated testing to find more than 75% interface problems?
4-6
What do we need to tell a person before the moderated user testing
We are testing not you, we are testing the interface
Any your actions will be useful for us
Anatomy of the sticker after the user testing
How to organize the results of user testing with users
What is really important to tell a person before the user testing?
- We don’t test you, we test the interface
2. You are not able to do smth wrong
Що таке АБ тест
Це коли є контрольна група, яка бачить старе рішення, і цільова група, яка бачить змінене рішення. Ми міряємо не поведінку, а результат.
Чи завжди АБ тест проводиться діленням авдиторії навпіл
Необов’язково, але класичний варіант так.
Workaround якщо ми боїмось (нові ліки або сміливе рішення) — взяти 20% авдиторії і її розділити навпіл.
Ще можна і саму авдиторію поділити нерівномірно, але тоді включається дуже інша математика
Minimum detectable effect
Статистично значний результат АБ тесту. Вимірюється у відсотках.
Дає нам кількість семплу для тестування.
What is statistical significance (статистична значущість)
More precisely, a study’s defined significance level, denoted by α
(alpha) , is the probability of the study rejecting the null hypothesis, given that the null hypothesis is true;and the p-value of a result.
p, is the probability of obtaining a result at least as extreme, given that the null hypothesis is true. The result is statistically significant, by the standards of the study, when p≤α.
The significance level for a study is chosen before data collection, and is typically set to 5%[13] or much lower—depending on the field of study.
Tools for AB testing
Google optimize
Extremely simple, to check some YouTube videos and then to add to the CV
Tools for AB testing
Google optimize, Amplitude і щось для апок
How to divide people to A/B groups for the tests at web services
- The person comes to LUN
- LUN asks for the cookie
- The person gives cookie
- LUN writes there A or B
- Then the person sees A or B section
The first question when we see the interface changes
- What was the challenge?
The approach for AB tests
Babysteps
What is the problem of Google optimize
Page blinking which affects results of the test
How to work around Google analytics restriction of 6.7% (if there are too many users)
To use raw data from Google analytics api
How to store results of AB tests
Better with Google Data studio
(Without ctrl C ctrl V)
Can take data from spreadsheets, from Google analytics
Блеф при АБ тестах
Тестувати неіснуючі фічі
- Приховати об’яву
- Нові фільтри на мапі
(Не забути додати, що фіча в розробці)
Що таке health метрика
Показник, що ми нічого не зламали
1) пульс
2) конверсія
SQL Amazon
RedShift
What are the major parts of AB tests
1) configurator (divides into groups, tests, so on)
2) stat model (counts)
What is important to check during AB tests
Triangulation of analytics
For example, Amplitude + GA + DB
Options of DB
Presto, RedShift (Amazon), Clickhouse (yandex)
Important thing about prod DB for AB tests
Analytics DB should be another one from prod DB
Do you need SQL or NoSQL db for analytics?
SQL
Tools for AB tests
- Google optimize (sampling problems)
- Optimize li (very expensive)
- VWO (Indian — very good)
- Amplitude (?)
The most popular stat models for AB testing are:
- Bayesian statistics (we wait for some period and then compare data to get the winner)
- Sequential statistics (one of the model will get ahead of another one, and it might happen much quicker
From what to start AB test
From AA test
When to conduct AA tests
- When we set up a new tool
2. When we choose a new segment
MVT
Minimal valuable test
What else can mean MVT
Multi variative test
What is a problem of ABC tests
Probability to get wrong data becomes higher
What is the maximum amount of AB tests with one control group?
ABC (three)
When AABB tests make sense
When we are testing on complicated segments of users (it helps to reduce possibility of mistake)
Incrementally test or uplift test
Help to understand how our actions influence all other.
We spent 100 usd and earned 200.
What if we didn’t spend 100 usd but warned 140?
So during tests we can cut sending offers, showing ads and so on
MVT (multi variations test)
Test 4 pictures with 5 buttons — which combination is better?
‘Fine polishing’
Takes loads of sources, makes sense only in very important knots of the UX (search results on Booking)
What if we don’t want to lose money during AB test
We can use multi-armed bandit test.
It distributes traffic in such a way that more ‘money-reach’ flow gets more traffic
How to launch our winner solution
Roll-out:
1%, 10%, 50%, 75%, 100%
The tool for getting real responses
Survey monkey audience
Specifics of jobs to be done framework
- The person has continued jobs to do
2. The person ‘hires’ (and then fires) products
The challenge of Jobs stories
To choose the level of abstraction
- Readiness of the team for innovations (the more abstract job story, the more various solutions)
- Restrictions
What to do with job story challenge
To make them a little more abstract than you could (and fit them with the team/market)
The difference between job story and user story
As a (persona) I want to (task) so that (profit)
When (situation) I want to (motivation) so that (better me)
What are the components of jobs to be done
Forces of progress
> Pushes Pulls
<
Anxieties Habits
Forces of progress
> Pushes Pulls
<
Anxieties Habits
What are pushes
Something you don’t like in current product /situation
(Around person, exist physically now)
‘I need to do something with it’
What are pulls
Imagery, fantasy of a perfect solution
(In the head, in the future)
‘I dream if I could…’
What else makes influence on forces of progress?
Catalysts
Situations
What is a switch interview
Interview with users who changed a product recently (they still remember their forces)
Jtbd interview goal
To discover forces (pushes, pulls, anxieties, habits)
What is the difference between Design thinking and human-centered design
- Design thinking looks at the bigger picture: It focused on innovation and creating products or services that solve problems.
- Human-centred design looks at the details: It is a way of improving the usability and the user experience of a particular product or service.
- Whilst Design Thinking is a Process, Human-Centred Design is a mindset.
What of human-centered design
Inspiration
Ideation
Implementation
What is a Design thinking
- Emphasize
- Define
- Ideate
- Prototype
- Test
Double diamond
Illustration of Design thinking
How to count geometric mean
- Multiply all values together to get their product.
- Find the nth root of the product (n is the number of values).
Before calculating the geometric mean, note that:
- The geometric mean can only be found for positive values.
- If any value in the data set is zero, the geometric mean is zero.
Jtbd interview structure
The Point of Purchase
- When did you purchase the product?
- Where were you?
- What time of day was it? (daytime/ nighttime?)
- What was the weather like?
- Was anyone else with you at the time?
- How did you purchase the product?
- Did you buy anything at the same time?
Finding the first thought
- When did you first realize you [needed something to solve your problem]?
- Where were you?
- Were you with someone?
- What were you doing, or trying to do when this happened?
Building the consideration set (With the two anchors set, we start exploring the space in between their first thought and their eventual purchase)
- Tell me about how you looked for a product to solve your problem.
- What kind of solutions did you try? Or not try? Why or why not?
Be curious about emotion
- Did you ask anyone else about what they thought about the purchase you were about to make?
- What was the conversation like when you talked about purchasing the product with your ?
- Before you purchased did you imagine what using the product would be like? Where were you when you were thinking this?
- Did you have any anxiety about the purchase? Did you hear something about the product that made you nervous? What was it? Why did it make you nervous?
Why we may ask about the weather during jtbd interview?
Most of our memories are made and recalled though association between places, people, things and our senses. If I asked you what the weather was like yesterday, you may struggle to remember; however, if I asked you what clothes you were wearing, you might remember how your feet got really wet because during your walk to work, it unexpectedly rained and your feet got wet because you were wearing sneakers…
What is a consideration set in terms of Jtbd interview?
it’s the alternative solutions your customer was considering. The interesting thing is, it’s rarely what you think your competition is.
The timeline for the Jtbd interview
5W1H
Why I am building this?
Who I am building it for?
When and where it will be used?
What I am building?
How could I measure it?
