Design Thinking Principles, Models and Mindsets

Introduction

Design thinking is a human-centred approach to addressing complex problems. It combines research, interpretation, idea generation, experimentation and learning. Although the term is strongly associated with product and service design, its principles apply equally to websites, mobile applications, public services, learning platforms, healthcare systems and organisational processes.

For UI/UX practitioners, design thinking provides a disciplined alternative to designing from personal preference or from assumptions about what users need. It encourages practitioners to understand people in context, frame problems carefully, generate alternatives, make ideas tangible and obtain evidence through testing. It is not a rigid sequence in which one stage is completed permanently before the next begins. Instead, it is an iterative way of working in which teams move between understanding, defining, creating and evaluating.

A central principle is that a design solution is not successful merely because it is visually attractive or technically possible. It should respond to a meaningful user need, support an appropriate task or goal, and create value within organisational, technical and ethical constraints.

Learning Objectives

By the end of this chapter, you should be able to:

• explain the principles and mindset of design thinking;
• distinguish divergent thinking from convergent thinking;
• apply the Empathize–Define–Ideate–Prototype–Test model;
• compare the Double Diamond and Lean UX approaches;
• use iteration and experimentation to reduce uncertainty;
• plan collaborative design activities;
• identify limitations, ethical risks and inappropriate uses of design thinking; and
• apply a design-thinking process to a practical UI/UX problem.

1. The Foundations of Design Thinking

1.1 Human-centredness

Human-centred design begins with people rather than with a preferred technology or predetermined solution. The designer seeks to understand users’ goals, behaviours, emotions, abilities and environments. This includes recognising that users are not a single homogeneous group. Differences in age, language, culture, disability, digital confidence, economic circumstances and access to devices may substantially affect experience.

Human-centredness does not mean accepting every user request literally. Users can describe difficulties and goals, but they may not be able to identify the best underlying solution. The designer therefore studies actual behaviour and context, combines several forms of evidence, and interprets findings responsibly.

1.2 Empathy and perspective-taking

Empathy in design means making a serious effort to understand another person’s experience without assuming that one’s own experience is universal. It involves listening, observing, asking non-leading questions and noticing barriers that users may consider normal. Empathy is not simply being polite or sympathetic; it is an investigative capability.

A researcher should distinguish among:

• What users say: their reported opinions, needs and explanations;
• What users do: observable actions and workarounds;
• What users think: beliefs, expectations and mental models; and
• What users feel: emotions such as confidence, anxiety, frustration or relief.

These sources may conflict. For example, a person may say that an online form is easy but repeatedly ask for help while completing it. Such discrepancies are valuable evidence rather than reasons to dismiss the participant.

1.3 Collaboration across perspectives

Design problems involve multiple stakeholders. A user, customer-support employee, developer, manager and compliance officer may each see a different aspect of the same service. Collaborative design brings these perspectives into a shared process.

Effective collaboration is not the same as allowing the most senior person to decide. Teams should create conditions in which evidence can be examined, minority viewpoints can be heard and assumptions can be challenged. Workshops, critique sessions, co-design activities and shared research repositories can support this process.

1.4 Iteration and learning

Iteration means repeating activities while improving understanding or the proposed solution. A prototype may reveal that the team misunderstood the problem. A usability session may expose an unexpected workflow. A technical constraint may require the team to revise the concept.

Iteration is valuable because uncertainty is unavoidable. However, iteration should not mean making random changes indefinitely. Each cycle should have a purpose, a question or hypothesis, an appropriate method, and a decision about what will happen next.

1.5 Making ideas tangible

Abstract discussion often hides disagreement. A sketch, storyboard, wireframe, clickable prototype or role-play makes an idea visible and testable. Prototypes are not only demonstrations of finished work; they are learning instruments. Their fidelity should match the question being investigated. A paper sketch may be sufficient for exploring screen structure, whereas a high-fidelity interactive prototype may be appropriate for testing navigation or animation.

1.6 Comfort with ambiguity

At the beginning of a project, the problem may be unclear and evidence may be incomplete. Design thinking encourages teams to delay premature commitment while still moving forward through small, purposeful experiments. This balance is important: excessive certainty produces solutions to the wrong problem, while excessive openness prevents decisions.

2. Divergent and Convergent Thinking

Design work alternates between expansion and selection.

Divergent thinking expands the field of possibilities. The team gathers many observations, identifies multiple interpretations and generates numerous ideas. Judgement is temporarily deferred so that familiar or dominant solutions do not eliminate promising alternatives too early.

Convergent thinking narrows the field. The team groups evidence, prioritises needs, evaluates concepts and chooses a direction. Convergence should be based on explicit criteria such as user value, feasibility, accessibility, risk, cost and alignment with the service goal.

The two modes are complementary. Divergence without convergence produces an unmanageable list of ideas. Convergence without divergence locks the team into the first plausible answer.

A useful pattern is:

1. expand understanding through research;
2. narrow the evidence into a problem frame;
3. expand possible solutions through ideation;
4. narrow concepts through criteria and prototyping; and
5. expand learning through testing before making final decisions.

The stages do not always occur once or in a perfectly linear order. A team may return to research after testing reveals a false assumption.

3. The Empathize–Define–Ideate–Prototype–Test Model

The five-part model is widely used as an accessible introduction to design thinking. The stages are best understood as connected activities rather than mandatory gates.

3.1 Empathize

The purpose of empathize activities is to understand relevant people, situations and needs. Common methods include semi-structured interviews, contextual observation, diary studies, journey mapping, analysis of support requests and accessibility inquiries, and review of existing research.

A strong interview guide uses open questions such as “Tell me about the last time you…,” “What happened next?” and “How did you decide what to do?” Questions should invite concrete examples rather than idealised opinions. The researcher should avoid leading questions such as “Would you like a faster button?” because they suggest a solution.

Observation is particularly useful when users find it difficult to explain routine behaviour. The researcher records actions, interruptions, tools, environmental constraints and workarounds. Ethical observation requires informed consent, respect for privacy and careful handling of personal data.

Outputs may include evidence statements, quotations, behavioural patterns, personas, empathy maps and journey maps. These outputs are interpretations of evidence, not substitutes for the evidence itself.

3.2 Define

The define activity converts a broad situation into a focused, evidence-informed problem frame. A useful problem statement identifies a target group, a meaningful need and relevant context without prescribing a solution.

A weak statement is: “We need a dashboard with more features.” This assumes the answer and provides no evidence about the problem. A stronger statement is: “New volunteers need a clear way to understand which tasks are urgent because they currently rely on scattered messages and are uncertain about priorities.”

A useful format is:

“[User group] needs a way to [need or goal] in [context] because [insight or barrier].”

The team should distinguish a problem statement from a design challenge. A problem statement describes the situation. A design challenge reframes it as an invitation to explore possibilities, often beginning with “How might we…?” For example: “How might we help new volunteers recognise and accept urgent tasks without requiring them to search across multiple channels?”

Definition also involves identifying assumptions. Teams can record assumptions about users, value, behaviour, technology and organisational capability. These assumptions later become candidates for research or experimentation.

3.3 Ideate

Ideation aims to produce alternatives before selecting one. Techniques include brainstorming, brainwriting, sketching, worst-possible-idea exercises, analogies, storyboards, mind maps and co-design. The facilitator should establish a clear challenge, time limit and participation rules.

Useful ideation principles include:

• defer evaluation during initial idea generation;
• build on others’ contributions rather than defending ownership;
• make ideas visible;
• encourage a range of concepts, including deliberately ambitious ones; and
• include people with different knowledge and experiences.

Ideation does not mean that every idea is equally valuable. After divergence, concepts should be grouped and evaluated. Criteria may include desirability, usability, accessibility, feasibility, viability, sustainability and risk. A concept that is attractive but excludes users with disabilities is not a strong candidate.

3.4 Prototype

A prototype is a representation of a proposed experience. It may be a paper interface, wireframe, storyboard, service role-play, clickable screen flow or coded experiment. The appropriate level of detail depends on the learning goal.

A low-fidelity prototype is quick and inexpensive to change. It is useful for exploring information architecture, rough layout and alternative task flows. A higher-fidelity prototype can investigate interaction details, content hierarchy, visual confidence and realistic task completion, but it may cause participants to focus on cosmetic details or assume that unfinished elements are fixed.

A prototype should have a learning objective. For example:

• “Can users identify the next action from the task screen?”
• “Do users understand the difference between saving and submitting?”
• “Can a screen-reader user reach the confirmation status?”

The team should prototype the riskiest or most uncertain part first, rather than polishing the entire product.

3.5 Test

Testing obtains evidence about how people interpret and use a prototype. A moderated usability session may ask participants to complete realistic tasks while the facilitator observes. The facilitator should avoid teaching the interface during the task and should distinguish a participant’s difficulty from a facilitator’s failure to explain the study.

A test plan should specify participants, recruitment criteria, tasks, context, measures and ethical procedures. Possible measures include task completion, observable errors, time-on-task, requests for assistance and participant confidence. Qualitative observations often explain why a difficulty occurred.

Testing is not a referendum in which participants vote on whether they like a design. A participant may prefer a familiar but inefficient pattern, while another may misunderstand a concept because of unclear content. The designer interprets findings in context and considers whether the problem is isolated or systematic.

After testing, the team decides whether to refine the concept, test an alternative, revisit the problem definition or stop pursuing the idea. A negative result is useful when it prevents investment in an unsuitable direction.

4. The Double Diamond

The Double Diamond, associated with the Design Council, represents design as two linked diamonds. The first diamond concerns identifying and defining the problem; the second concerns developing and delivering a response. Each diamond contains a period of divergence followed by convergence.

The four commonly described activities are:

The model corrects a common misconception that design begins with solution generation. It highlights the importance of finding the right problem before developing an answer. It also shows that delivery is not necessarily the end: live services generate new evidence and may require further discovery.

The Double Diamond is a communication model rather than a complete method. It does not prescribe recruitment numbers, interview questions, prioritisation rules or usability metrics. Teams must select methods suitable for the problem and context.

5. Lean UX

Lean UX adapts user-centred design to environments where products are developed incrementally, often by cross-functional Agile teams. It emphasises shared understanding, outcomes, rapid learning and reducing the amount of documentation that does not support a decision.

A typical Lean UX cycle is:

1. identify a business or user outcome;
2. state assumptions about users, needs and behaviours;
3. formulate a hypothesis;
4. create the smallest experiment or prototype that can test it;
5. obtain evidence; and
6. decide whether to persevere, modify or abandon the direction.

A hypothesis can be expressed as:

“We believe that [change] will help [user group] achieve [outcome]. We will know this is supported when [observable evidence].”

For example: “We believe that showing a single recommended next step will help first-time applicants complete the form. We will know this is supported when more participants begin the correct next step without assistance.”

Lean UX encourages the team to focus on outcomes rather than output. “Release three screens” is an output. “Reduce uncertainty about the next step” is an outcome. The distinction helps teams avoid measuring productivity only by the volume of features delivered.

Lean UX does not justify skipping research, accessibility work or documentation that is needed for safety, legal compliance or organisational memory. Speed is valuable only when the learning is reliable enough to inform action.

6. Design Thinking as an Experimentation System

An experiment is a deliberate investigation of an assumption. It should identify what is uncertain, what evidence would support or challenge the assumption, and what decision will follow.

A simple experiment structure is:

• Assumption: what the team currently believes;
• Risk: why being wrong matters;
• Method: how evidence will be obtained;
• Indicator: what will be observed or measured; and
• Decision rule: what the team will do after the result.

For instance, a team may assume that users understand a particular icon. A quick comprehension test with several candidate icons may reveal that the icon is ambiguous. The team can then replace it before building the full interface.

Experiments should be proportionate. A high-risk assumption about medication dosage requires stronger evidence and governance than a low-risk assumption about the location of a decorative control. Teams should also avoid treating a small usability study as statistically representative of an entire population. Small studies can reveal severe interaction problems, but they do not establish population prevalence with precision.

7. Collaboration and Facilitation

Design thinking depends on the quality of group interaction. A facilitator establishes the purpose, time boundaries, activity instructions and decision process. Before a workshop, the facilitator should identify participants, prepare accessible materials, define the evidence base and clarify whether the session is for exploration, prioritisation or decision-making.

Several practices improve collaboration:

• separate idea generation from idea evaluation;
• use silent writing before discussion so that confident speakers do not dominate;
• ask participants to explain evidence rather than merely assert preferences;
• record decisions, unresolved questions and owners;
• invite people with operational, technical and accessibility knowledge; and
• make disagreement explicit and constructive.

Co-design involves users or representatives directly in creating and evaluating ideas. It can produce valuable knowledge and increase relevance, but participation must be meaningful. Users should not be asked to perform unpaid specialist labour without appropriate consideration, and organisations should not create the impression that every participant suggestion will be implemented.

8. Applied Case: Improving a Community Clinic Appointment Service

The following is an illustrative case designed for learning purposes.

Background

A community clinic receives appointments through telephone calls and an online booking service. The clinic wants to encourage digital booking because staff spend substantial time answering routine calls. The initial proposal is to build a mobile application.

Problem

The team first defines the problem as “patients need an app.” During discovery, researchers interview patients, observe reception workflows and review call reasons. They find that many patients share devices with family members, some have limited data access, and several are uncertain whether an appointment request has been successfully submitted. Older patients are not necessarily opposed to digital booking; their main concern is lack of confirmation and difficulty understanding medical service categories.

The evidence changes the problem frame. The central issue is not the absence of an app but uncertainty and complexity in booking, especially for people with limited digital confidence or access.

Analysis

The team maps the current journey: finding the clinic number, waiting on the telephone, explaining the request, selecting a service, receiving an appointment time and remembering the details. Breakdowns occur when service names are unfamiliar, confirmation is verbal only, or callers must repeat information.

Important assumptions include:

• patients can distinguish the clinic’s service categories;
• patients have reliable access to a personal device;
• a digital confirmation will reduce anxiety; and
• staff can support a revised workflow.

The team ranks the assumption about service-category comprehension as high risk because misunderstanding could lead to inappropriate bookings.

Approach and solution

During ideation, the team considers an application, a responsive website, a telephone-assisted digital service and improved confirmation messages. It prototypes a responsive website with plain-language service descriptions, a prominent “check your appointment” function, optional text confirmation and a clear route to telephone support.

The prototype is tested with patients who vary in age, language background, disability and digital experience. Participants complete tasks such as booking a routine consultation, changing an appointment and checking confirmation. Testing reveals that participants interpret “request appointment” as an immediate booking, even when the clinic must approve the request. The team changes the language to “send booking request,” explains the expected response time and provides a reference number.

Accessibility review identifies insufficient colour contrast and a confirmation message that is difficult to reach with keyboard navigation. These issues are corrected before further testing. Staff also test the administrative workflow to ensure that digital requests do not create duplicate records.

Outcome

The revised service is released as a responsive web service rather than as an app-only solution. Early monitoring shows fewer avoidable clarification calls and fewer incomplete requests. The clinic continues to offer telephone support because digital access is not universal. The result is therefore not simply a new interface but a more inclusive service arrangement.

Learning points

1. The initial solution request concealed a broader service problem.
2. Discovery challenged assumptions about users and device access.
3. A small prototype exposed a critical interpretation problem before implementation.
4. Accessibility and operational feasibility were considered alongside visual design.
5. The final response supported multiple channels rather than forcing every user into one channel.
6. Testing produced changes to content, interaction and organisational workflow.

9. Ethical, Inclusive and Responsible Practice

Design thinking can reproduce harm if teams treat user research as extraction or if they optimise a service for organisational convenience alone. Ethical practice requires informed consent, voluntary participation, confidentiality, secure data handling and sensitivity to vulnerable groups. Participants should understand the purpose of a study, what will happen to their information and how they can withdraw where appropriate.

Inclusion should be considered from the beginning, not added as a final compliance check. The team should ask who is absent from research and who might be disproportionately affected by the design. Interfaces should support accessible text, sufficient contrast, keyboard operation, understandable language, alternative input methods and compatibility with assistive technologies where relevant.

Designers should also examine dark patterns: interface choices that manipulate users into actions they did not intend, such as difficult cancellation, disguised advertising, preselected consent or confusing privacy controls. A design may improve a short-term conversion measure while damaging trust and autonomy.

AI tools can assist with clustering notes, generating alternative wording, creating synthetic scenarios or producing early visual concepts. They do not replace direct engagement with users or professional judgement. Sensitive research data should not be entered into an AI system without appropriate permission and security review. AI-generated content may contain bias, fabricated claims or inaccessible language and must be checked by a responsible human.

10. Challenges and Limitations

Design thinking is useful but not universally sufficient.

Oversimplification: Five-stage diagrams may imply a neat process, whereas real projects involve political, technical and organisational complexity.

Participation bias: Research participants may be easier to recruit than the people most affected by the service. Findings should therefore be interpreted in relation to recruitment and context.

Consensus pressure: Collaboration can become a search for agreement rather than a search for truth. Evidence and dissent must remain visible.

Prototype realism: Participants may behave differently in a prototype than in a real service, particularly when consequences, time pressure or personal stakes are absent.

Resource constraints: Meaningful research, accessibility testing and iteration require time and skill. Organisations that demand instant solutions may misuse the language of design thinking while avoiding genuine learning.

Power and implementation: A well-researched concept may fail because of procurement, policy, architecture or organisational incentives. Design thinking must be connected to implementation planning.

False certainty from metrics: A single metric rarely explains user experience. Completion rates, for example, may improve because users are forced through a flow rather than because the service is genuinely better.

These limitations do not invalidate design thinking. They indicate the need for appropriate research methods, critical reflection, domain knowledge and responsible decision-making.

11. Best Practices for UI/UX Students and Practitioners

• Begin with a real user or service problem, not a preferred feature.
• Make assumptions explicit and rank them by risk.
• Use research methods that match the question.
• Include diverse participants and examine who is missing.
• Alternate deliberate divergence with evidence-based convergence.
• Prototype the riskiest interaction early.
• Test realistic tasks rather than asking only whether users like a design.
• Record observations, interpretations and decisions separately.
• Treat accessibility, privacy and inclusion as design requirements.
• Measure outcomes, not merely the number of screens or features delivered.
• Share findings with the whole team so that research is not isolated.
• Continue learning after launch through support data, analytics, feedback and further research.

Chapter Summary

Design thinking is a human-centred, iterative and collaborative approach to solving complex problems. Its core principles include empathy, contextual understanding, experimentation, making ideas tangible, collaboration and willingness to revise assumptions. Divergent thinking expands possible interpretations and solutions, while convergent thinking selects and focuses them.

The Empathize–Define–Ideate–Prototype–Test model provides a practical introduction to the design process. The Double Diamond emphasises the distinction between exploring and defining a problem and developing and delivering a response. Lean UX connects design thinking with incremental product development through assumptions, hypotheses, experiments and outcome-focused learning.

Effective practice treats prototypes as learning tools, tests risky assumptions early and uses evidence rather than personal preference. Ethical and inclusive design requires attention to consent, privacy, accessibility, representation, autonomy and the possible misuse of technology. Design thinking is not a universal recipe; it must be adapted to context and combined with domain expertise, implementation knowledge and appropriate evaluation.

Key Terms

Design thinking: A human-centred approach that combines inquiry, framing, ideation, prototyping and evaluation.

Human-centred design: Designing around people’s goals, contexts, capabilities and experiences.

Empathy: Informed understanding of another person’s situation and perspective.

Divergent thinking: Expanding the range of evidence, interpretations or ideas.

Convergent thinking: Narrowing options through analysis, criteria and decisions.

Iteration: Repeating a process to improve understanding or a proposed solution.

Problem frame: A focused description of the situation, user need and context being addressed.

How might we question: An open design challenge that invites possible responses without prescribing one.

Ideation: Deliberate generation of alternative concepts or approaches.

Prototype: A representation used to explore or test an idea.

Fidelity: The degree to which a prototype resembles the proposed final experience.

Usability testing: Evaluation of an interface or service through representative tasks and observation.

Double Diamond: A model of discover, define, develop and deliver, organised around divergence and convergence.

Lean UX: An approach that combines user-centred design, cross-functional collaboration and rapid outcome-focused experiments.

Assumption: A belief accepted temporarily without sufficient evidence.

Hypothesis: A testable statement linking a proposed change to an expected user or business outcome.

Co-design: Designing with users or stakeholders as active contributors.

Dark pattern: A manipulative interface practice that interferes with informed user choice.

Accessibility: The extent to which a product or service can be used by people with varied abilities and needs.

Review Questions

Short Answer Questions

1. What is meant by human-centredness in UI/UX design?
2. Distinguish divergent thinking from convergent thinking.
3. Why is a prototype considered a learning instrument rather than only a demonstration?
4. What is the purpose of a problem statement in the define activity?
5. Identify two ethical responsibilities involved in user research.

Long Answer Questions

1. Explain the Empathize–Define–Ideate–Prototype–Test model and show how the stages may interact iteratively.
2. Compare the Double Diamond with Lean UX in terms of purpose, activities and outputs.
3. Discuss how collaboration and facilitation influence the quality of design-thinking work.
4. Explain why accessibility and inclusion should be addressed throughout the design process rather than at the end.
5. Evaluate the major limitations of design thinking as a general problem-solving approach.

Analytical / Application Questions

1. A university wants to create an app for reporting classroom problems, but students currently use email, messaging groups and in-person requests. Apply the Double Diamond to plan an initial investigation and identify what the team should learn before choosing a solution.
2. A team assumes that adding a progress bar will help users complete a long application form. Formulate a Lean UX hypothesis, propose a low-cost experiment and specify what evidence would cause the team to revise the idea.
3. During usability testing, most participants complete a task but several misunderstand the meaning of the final confirmation message. Analyse whether the team should continue implementation, change the content, revisit the problem definition or conduct further research. Justify your decision.

Multiple Choice Questions with Answers

Question 1

Which principle most directly requires designers to understand users in their real contexts?

A. Human-centredness

B. Visual consistency

C. Feature maximisation

D. Technical optimisation

Correct Answer: A. Human-centredness

Question 2

What is the primary purpose of divergent thinking?

A. To select one final design immediately

B. To expand possible interpretations and solutions

C. To verify legal compliance

D. To measure implementation cost

Correct Answer: B. To expand possible interpretations and solutions

Question 3

Which statement is the strongest problem frame?

A. We need a modern mobile app.

B. Add more features to the dashboard.

C. New volunteers need to recognise urgent tasks because priorities are currently scattered across messages.

D. Users want better technology.

Correct Answer: C. New volunteers need to recognise urgent tasks because priorities are currently scattered across messages.

Question 4

In the Empathize–Define–Ideate–Prototype–Test model, what is the main purpose of testing?

A. To prove that the designer’s first idea is correct

B. To obtain evidence about how people interpret and use a proposed experience

C. To replace all research activities

D. To complete visual styling before development

Correct Answer: B. To obtain evidence about how people interpret and use a proposed experience

Question 5

Which sequence correctly represents the four activities commonly associated with the Double Diamond?

A. Discover, Define, Develop, Deliver

B. Define, Discover, Deliver, Develop

C. Ideate, Build, Advertise, Measure

D. Research, Code, Launch, Replace

Correct Answer: A. Discover, Define, Develop, Deliver

Question 6

Lean UX primarily encourages teams to focus on which of the following?

A. Producing the largest possible number of features

B. Outcomes, assumptions and rapid learning

C. Completing documentation before research

D. Avoiding prototypes until development begins

Correct Answer: B. Outcomes, assumptions and rapid learning

Question 7

Which prototype is generally most appropriate for quickly exploring alternative page structures?

A. A fully implemented production system

B. A low-fidelity sketch or wireframe

C. A final marketing campaign

D. A database migration

Correct Answer: B. A low-fidelity sketch or wireframe

Question 8

A team believes users understand an unfamiliar icon. What is the most appropriate first step?

A. Assume the icon is understood because it is visually attractive

B. Run a small comprehension or usability experiment

C. Add unrelated features to the screen

D. Remove all user testing

Correct Answer: B. Run a small comprehension or usability experiment

Question 9

Which practice is an example of a dark pattern?

A. Providing clear cancellation instructions

B. Using sufficient colour contrast

C. Preselecting consent and making withdrawal deliberately difficult

D. Offering keyboard navigation

Correct Answer: C. Preselecting consent and making withdrawal deliberately difficult

Question 10

Why should a design team avoid treating a small usability study as representative of an entire population?

A. Small studies can never reveal any usability problem

B. Small studies usually provide insight but do not precisely establish population prevalence

C. Usability studies measure only visual colour

D. Participants are not allowed to use prototypes

Correct Answer: B. Small studies usually provide insight but do not precisely establish population prevalence