Frameworks

Ask 'why' repeatedly to move from a symptom to a root cause worth designing for. Use when: Framing the problem before jumping to solutions.

Analyse, Design, Develop, Implement, Evaluate — the classical ID lifecycle. Use when: Structuring a full curriculum project end to end.

Activities, Environments, Interactions, Objects, Users — a lens for structured field observation. Use when: Running contextual inquiry or ethnographic visits.

Apple's guidance on setting expectations, handling mistakes and communicating model behaviour. Use when: Shipping on-device or generative features on Apple platforms.

Plot assumptions by importance and evidence to decide what to test first. Use when: Choosing what to research before building anything.

Atoms, molecules, organisms, templates, pages — a mental model for design systems that scale. Use when: Structuring a component library or documentation.

Break findings into experiments, facts, insights and conclusions to make research reusable. Use when: Building a research repository the team actually queries.

Start from desired results, then evidence of learning, then learning experiences. Use when: Designing a new module or reworking one that isn't landing.

Remember, understand, apply, analyse, evaluate, create — leveled learning objectives. Use when: Writing course outcomes and assignment rubrics.

Six-phase reflexive method for coding qualitative data into defensible themes. Use when: Synthesising interviews for a rigorous, sharable report.

Reduce intrinsic, extraneous and germane load so users can focus on the task, not the interface. Use when: Simplifying dense enterprise or data-heavy screens.

Align outcomes, teaching activities and assessment so what you test is what you teach. Use when: Reviewing an existing syllabus for coherence.

Sketch 8 ideas in 8 minutes to break out of the first-idea trap. Use when: Warming up before a concept sketch or workshop.

Sort situations into clear, complicated, complex, chaotic — pick the right decision style. Use when: Deciding how to lead a decision with unclear cause and effect.

Understand, sketch, decide, prototype, test — a compressed project template. Use when: Running a mini project in a week.

How, what and who of DesignOps — orchestrate people, process and craft at scale. Use when: Building or restructuring a design ops function.

Stages from ad-hoc to embedded — assess where your ops sit and what to invest in next. Use when: Planning DesignOps hires and tooling.

Discover, Define, Develop, Deliver — a shared vocabulary for divergent and convergent design. Use when: Planning a project phase or explaining process to stakeholders.

Sort work by urgency vs importance so the important-not-urgent gets protected time. Use when: Weekly planning when the backlog is bigger than the calendar.

Says, thinks, does, feels — a canvas to consolidate what you learn about users. Use when: Turning interview notes into a shareable artefact.

Cagan's model for autonomy, coaching and clarity that lets teams solve problems, not ship features. Use when: Diagnosing why a team is executing but not moving the metric.

Unacceptable, high, limited, minimal — the risk classification driving EU AI compliance. Use when: Scoping AI features against European regulation.

Behaviour happens when motivation, ability and a prompt converge at the same moment. Use when: Designing nudges, onboarding, or habit-forming flows.

Proximity, similarity, closure and continuity — how the eye groups elements into meaning. Use when: Fixing visual hierarchy or grouping in a dense screen.

Happiness, Engagement, Adoption, Retention, Task success — connect UX to metrics. Use when: Defining product-level UX KPIs with PMs and data.

Patterns for mental models, explainability, feedback, errors and trust in AI-powered UI. Use when: Designing an LLM feature, agent, or AI-assisted flow.

Trigger, action, variable reward, investment — the loop behind habit-forming products. Use when: Auditing whether your engagement loop is ethical and complete.

Inspiration, Ideation, Implementation — a field guide for designing with people, not for them. Use when: Kicking off a discovery with real user immersion.

Frame every feature as a job a user hires the product to do in a specific situation. Use when: Prioritising features against real user progress, not personas.

Reconstruct the timeline of a customer switching to your product to surface real forces of progress. Use when: Understanding acquisition, churn or category switching.

Classify features as must-be, performance, attractive, indifferent or reverse based on satisfaction. Use when: Prioritising a feature list against user delight vs. table stakes.

Concrete experience, reflection, conceptualisation, experimentation — a loop for studio work. Use when: Designing studio, lab or project-based learning.

A compact reference of cognitive and behavioural laws (Fitts, Hick, Miller, Jakob) applied to UI. Use when: Justifying a layout, spacing or interaction decision.

Practical patterns for colour, touch targets, focus and motion that meet WCAG in real UI. Use when: Turning WCAG into concrete component decisions.

18 validated guidelines for human-AI interaction across initial, during, when-wrong and over-time. Use when: Reviewing an AI feature for trust and recovery.

Recognise exclusion, learn from diversity, solve for one and extend to many. Use when: Expanding scope beyond a narrow 'average user'.

Practical requirements for fairness, reliability, safety, privacy, inclusion and accountability. Use when: Turning AI principles into shippable engineering practice.

Age-appropriate AI concepts, ethics and hands-on activities across levels. Use when: Bringing AI into a design curriculum responsibly.

Must, Should, Could, Won't — a simple way to scope a project realistically. Use when: Scoping a bootcamp or capstone project.

Ten general principles for interaction design — the most-cited critique lens in the field. Use when: Reviewing a flow or auditing an interface for quick wins.

Govern, Map, Measure, Manage — the reference to make AI programs trustworthy at scale. Use when: Standing up AI governance in a regulated org.

Small-sample moderated testing — five users find ~85% of issues if the tasks are right. Use when: Validating a specific flow before or after release.

Six stages from absent to user-driven — a language to diagnose and level up an org. Use when: Making the case for investment or a 12-month UX roadmap.

What hiring managers actually look for — process, impact, honest trade-offs. Use when: Building or refreshing your portfolio.

Pick one leading metric that captures value delivered and connect inputs that move it. Use when: Aligning research on the metric that matters most.

Human-centred values, transparency, robustness and accountability for trustworthy AI. Use when: Framing a company-wide responsible AI policy.

Objectives and key results applied to design outcomes rather than output. Use when: Setting quarterly goals that connect design to business.

Link outcomes to opportunities, solutions and assumption tests for continuous discovery. Use when: Structuring discovery and connecting research to shipped work.

PM + Design + Engineering discover and deliver together — the modern operating unit. Use when: Setting up an empowered team that owns outcomes end-to-end.

Responsible, Accountable, Consulted, Informed — make decision rights explicit across PM/Design/Eng. Use when: Untangling ownership on a cross-functional initiative.

One tab, participants as columns, observations as rows — spot patterns as they emerge. Use when: Synthesising qualitative interviews across a small sample.

Governance, tools, recruitment, admin, comms, knowledge, competency and advocacy for scaled research. Use when: Standing up or maturing a research practice.

Reach, Impact, Confidence, Effort — a transparent score for portfolio-level trade-offs. Use when: Comparing very different initiatives in a roadmap review.

Rapid Iterative Testing and Evaluation — fix issues between sessions, not after the study. Use when: Shipping a fast-moving prototype with a tight loop.

Successive Approximation Model — iterative, prototype-driven course design. Use when: Building a course like you'd build a product.

Big questions, small groups, shared screens — students drive the inquiry. Use when: Facilitating exploratory studio sessions.

One column of criteria with space either side for evidence of concerns and strengths. Use when: Grading design work where narrative feedback matters more than levels.

Situation, Task, Action, Result — a clear narrative shape for portfolios and interviews. Use when: Writing a project case study that hiring managers can skim.

Set intent, describe, interpret, judge — turn crits from opinion parades into learning. Use when: Running a studio critique that students actually learn from.

A 10-item survey producing a single comparable usability score from 0-100. Use when: Benchmarking usability across releases or against industry.

Stream-aligned, platform, enabling and complicated-subsystem teams with clear interaction modes. Use when: Redesigning how design, research and product teams cooperate.

Camille Fournier's ladder-by-ladder guide to what changes as you grow from lead to VP. Use when: Coaching yourself or a new lead into their next level.

Rank a long list of user tasks by vote to focus design on what actually matters most. Use when: Prioritising content or IA on a large site.

Multiple means of engagement, representation and action to include every learner. Use when: Making a class accessible and inclusive by design.

A visual model of the steps a user takes to complete a task — decisions, screens, errors. Use when: Explaining a feature to peers or reviewers.

Fit customer jobs, pains and gains to the products, pain relievers and gain creators you ship. Use when: Aligning a portfolio around real customer value.

Map user needs to components on an evolution axis to spot strategy and platform bets. Use when: Deciding where to build, buy or partner.

The web accessibility baseline — perceivable, operable, understandable, robust — with 9 new SC. Use when: Committing to a concrete accessibility target for a release.

A filterable checklist of success criteria — the standard reference for accessible design. Use when: Making sure a class project meets baseline accessibility.

Cost of delay divided by job size — a lean prioritisation for flow-based delivery. Use when: Ordering a backlog against economic impact.