Foundations of Design Thinking

Introduction and context

Design thinking is a human-centred approach to understanding situations, framing opportunities, generating possibilities, making ideas tangible, and learning through evidence. In product management, it helps teams decide not only what to build, but also whether a problem is worth solving, for whom, under what conditions, and with what consequences.

The term is used in several ways. It may refer to a professional way of working associated with designers; a set of methods such as interviewing, observation, prototyping and testing; a model of innovation; or a mindset that values inquiry, experimentation and constructive collaboration. These meanings overlap but are not identical. Design thinking is not a linear recipe, a brainstorming session, or a guarantee of innovation. It is a disciplined response to uncertainty.

Product managers operate between customers, business leaders, designers, engineers, operations, regulators and other stakeholders. Their work therefore involves ambiguity, competing objectives and incomplete evidence. Design thinking provides a useful foundation for making progress in such conditions while keeping human needs and system consequences visible.

Learning objectives

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

1. Explain the historical origins and intellectual foundations of design thinking.
2. Distinguish design thinking from human-centred design and systems thinking.
3. Define major terms, including empathy, framing, ideation, prototyping, wicked problem and abductive reasoning.
4. Apply divergent and convergent thinking to product discovery and delivery.
5. Compare design thinking with Lean Startup, Agile and conventional product management.
6. Analyse a complex product situation and select suitable discovery activities.
7. Evaluate the limitations, ethical risks and organisational conditions affecting design-thinking practice.

1. Origins and intellectual foundations

Design has always involved shaping artefacts, services and environments for human purposes. Modern design thinking, however, emerged from several intellectual and professional traditions rather than from a single inventor.

Design as a distinct way of knowing

Herbert Simon’s The Sciences of the Artificial described design as concerned with changing existing situations into preferred ones. This distinguished design from the natural sciences, which primarily seek to explain how the world is. Design asks what could be made or changed and how an intended future might be realised.

Design researchers such as Nigel Cross argued that design has characteristic forms of knowledge, including dealing with ill-defined problems, using constructive modelling and reasoning from proposed solutions. Donald Schön described professional design as “reflective practice”: practitioners act, observe consequences, reinterpret the situation and revise their understanding. This explains why design work rarely follows a fully specified plan.

In the late twentieth and early twenty-first centuries, consultancies and universities popularised design thinking as a collaborative approach to innovation. IDEO and the Stanford d.school contributed influential practical models, while the Design Council’s Double Diamond communicated a related pattern of discovering, defining, developing and delivering. These models are useful teaching devices, but they should not be treated as universal laws or fixed sequences.

A pattern rather than a pipeline

Design thinking is often illustrated as a sequence of empathise, define, ideate, prototype and test. In practice, teams move backwards and forwards. Testing can reveal that the original problem was poorly framed; a prototype can expose a new user need; and stakeholder research can invalidate an attractive solution. The process is therefore iterative, evidence-seeking and partly non-linear.

2. Core principles

Human-centredness

Human-centred practice begins with people’s goals, behaviours, constraints and lived contexts rather than with a predetermined feature. It considers direct users, non-users, employees, affected communities and other stakeholders. Human-centredness does not mean satisfying every stated request. People may describe a solution while actually revealing a deeper need. The product team must investigate the underlying situation.

Problem framing before solution commitment

A problem statement directs attention. “How might we help first-time patients feel prepared before a consultation?” is more generative than “How might we add a reminder feature?” Good framing is specific enough to guide inquiry but open enough to permit alternatives. It identifies a meaningful outcome, a relevant population and important constraints without prematurely prescribing the answer.

Learning through making

A prototype is a deliberately incomplete representation used to learn. It may be a sketch, storyboard, clickable interface, role-play, service blueprint, landing page or manual simulation. Its value depends on the question it is designed to answer. A low-fidelity prototype is often preferable when the team is testing a concept or workflow; a higher-fidelity prototype may be necessary to assess interaction details or technical feasibility.

Collaboration across disciplines

Complex products require multiple forms of expertise. Designers may contribute research and interaction knowledge, engineers technical insight, commercial teams market understanding, legal specialists risk analysis, and domain experts contextual knowledge. Collaboration is not merely consensus-building. Productive disagreement can reveal assumptions and improve decisions when it is managed respectfully and supported by evidence.

Evidence and judgment

Design thinking values evidence but does not eliminate judgment. Interviews may reveal motivations, analytics may reveal behaviour, experiments may estimate causal effects, and prototypes may expose usability problems. Each method has limits. Product decisions require interpretation of evidence in relation to strategy, feasibility, viability, desirability, accessibility and ethics.

Inclusive and responsible design

A product can work for its intended user while harming people excluded from research or affected indirectly. Inclusive practice considers disability, language, culture, income, age, gender, digital access and power differences. Responsible design also examines privacy, safety, manipulation, environmental effects and the distribution of benefits and burdens.

3. Important terminology

User or participant: A person who uses, encounters or is affected by a product. “Participant” is often preferable in research because not everyone studied is a current user.

Stakeholder: A person or organisation that affects, is affected by, or has an interest in the product.

Empathy: An effort to understand another person’s perspective, experience and context. It is not the claim that the researcher has identical feelings.

Observation: Systematic attention to behaviour and context, including workarounds and environmental conditions.

Insight: An interpreted and evidence-supported understanding that helps explain behaviour or an unmet need. A quotation alone is not necessarily an insight.

Need: A desired outcome, capability or condition. Needs should be distinguished from proposed features.

Assumption: A belief treated as true for planning purposes but not yet adequately validated.

Problem framing: The formulation of a situation and opportunity that guides investigation and solution development.

How might we question: An open but bounded prompt used to turn an observation into a design opportunity.

Ideation: The generation and development of possible approaches.

Prototype: A representation made to learn, communicate or test an idea.

Usability: The extent to which specified users can use a product effectively, efficiently and satisfactorily in a specified context.

Desirability: The extent to which people value or want an outcome or solution.

Feasibility: The extent to which the organisation can build, operate and support a solution with available capabilities.

Viability: The extent to which a solution can create sustainable organisational or economic value.

Divergence: Expanding the range of perspectives, questions or possible solutions.

Convergence: Narrowing possibilities through synthesis, prioritisation, evaluation or selection.

Wicked problem: A problem with contested definitions, interconnected causes, no definitive formulation and no objectively final solution.

Abduction: Reasoning from observations toward a plausible explanation or course of action.

4. Divergent and convergent thinking

Divergent thinking expands the search space. During discovery, a team may interview different user groups, map multiple journeys and identify several interpretations of the same complaint. During ideation, it may generate many possible concepts before judging them. The purpose is not uncontrolled creativity; it is to avoid premature commitment to the first plausible answer.

Convergent thinking synthesises and selects. Teams cluster research observations, identify patterns, define criteria, compare concepts and choose what to investigate next. Convergence should be based on explicit criteria rather than the loudest opinion or the most polished presentation.

A useful rhythm is:

1. Diverge to gather perspectives and possibilities.
2. Converge to identify patterns and a provisional focus.
3. Diverge to create alternative responses.
4. Converge to select prototypes or experiments.
5. Diverge again when evidence reveals an important unknown.

This rhythm is not a mechanical algorithm. Some activities combine both modes. For example, affinity mapping generates interpretations while also grouping them; concept selection expands understanding when rejected options reveal overlooked criteria.

Product managers should protect both modes. Excessive divergence produces an unmanageable backlog and weak commitment. Excessive convergence creates confirmation bias, feature fixation and solution lock-in.

5. Abductive reasoning and design decisions

Deduction applies a general rule to a case. Induction generalises from cases to a broader pattern. Abduction proposes the most plausible explanation or response for incomplete information. Design work relies heavily on abduction because teams rarely possess complete evidence.

For example:

• Observation: customers repeatedly abandon a form near the payment step.
• Possible explanations: the form is confusing, trust is low, payment methods are unsuitable, or the price is unexpected.
• Abductive proposal: unclear total cost may be causing uncertainty; test a transparent price summary and compare behaviour and feedback.

An abductive conclusion is provisional. It should be expressed as a hypothesis rather than a fact. Product teams can improve abductive reasoning by separating observations from interpretations, recording competing explanations, identifying disconfirming evidence and designing the smallest useful test.

A practical hypothesis structure is: “We believe that [user/context] experiences [problem] because [possible mechanism]. If we provide [intervention], we expect [observable outcome].” The statement is not a scientific law; it is a decision aid that makes assumptions inspectable.

6. Wicked problems and complexity

Wicked problems differ from well-defined technical problems. In a technical problem, the objective, constraints and success criteria may be relatively stable. In a wicked problem, stakeholders disagree about what the problem is, interventions change the situation, and success is judged from multiple perspectives.

Examples include improving urban mobility, reducing health inequality, increasing trust in online information and designing public-service access. A product may address only one part of such a system. Its apparent success can produce second-order effects elsewhere.

Common characteristics include:

• no single agreed problem definition;
• causes that interact rather than form a simple chain;
• solutions that are judged as better or worse, not finally right or wrong;
• consequences that emerge over time;
• stakeholders with unequal power and conflicting values; and
• constraints that shift as the intervention changes the system.

Design thinking helps by making perspectives visible, reframing assumptions, visualising journeys and systems, and testing interventions incrementally. It does not make wicked problems simple. Systems thinking is needed to examine relationships, feedback loops, incentives and unintended consequences.

A product manager should therefore ask not only “Does this feature improve the target metric?” but also “Who benefits, who bears the cost, what behaviour might it encourage, and what could happen at scale?”

7. Relationship to related approaches

Human-centred design

Human-centred design (HCD) is the broader commitment to designing with attention to human needs, capabilities, contexts and rights. Design thinking is one influential way of practising HCD, especially in innovation and ambiguous problem spaces. HCD may include formal standards, accessibility practice, participatory design and detailed usability engineering. Design thinking often emphasises reframing, ideation and rapid prototyping. The two should not be treated as competitors.

Systems thinking

Systems thinking focuses on interdependence, structure, feedback and patterns over time. Design thinking focuses more directly on human experience, opportunity framing and intervention. They complement one another. A service blueprint can connect front-stage user experience with back-stage processes, while a causal-loop or stakeholder map can expose systemic effects. Combining them reduces the risk of optimising a local interaction while damaging the wider service.

Lean Startup

Lean Startup emphasises experimentation under uncertainty, validated learning, the minimum viable product (MVP), and decisions to persevere or pivot. Design thinking contributes rich problem understanding and concept exploration before or alongside experiments. Lean methods can test whether behaviour supports a value proposition; design research can explain why people behave as they do and whether the proposed problem matters.

An MVP should not automatically mean a poor-quality or unsafe product. Its scope should be the smallest responsible intervention that tests a critical assumption. In regulated or high-risk contexts, learning experiments may need simulations, expert review or staged deployment rather than exposure of real users to avoidable harm.

Agile

Agile is a family of software-development approaches that values working increments, collaboration, feedback and adaptation. Design thinking and Agile operate at different but overlapping levels. Design thinking often informs discovery and problem framing; Agile structures iterative delivery. “Dual-track” practice uses a discovery stream to investigate problems and a delivery stream to build validated solutions. The streams must communicate continuously; otherwise teams may efficiently build an invalid concept.

Product management

Product management integrates customer value, organisational strategy, delivery capability and market or service outcomes. Design thinking is not a substitute for product strategy, prioritisation, financial analysis, operations or governance. It is a way to improve the quality of product decisions, particularly when needs and solutions are uncertain.

8. A practical product-discovery cycle

A product team can adapt the following cycle:

Explore the situation

Review existing evidence, interview and observe relevant participants, examine support contacts and analytics, and map stakeholders. Recruitment should include people with different abilities and levels of access. Obtain informed consent where appropriate, protect personal data and avoid leading questions.

Synthesize and frame

Separate raw observations, interpretations and decisions. Cluster evidence, identify tensions and write several possible problem frames. Select a provisional frame using criteria such as importance, evidence strength, strategic relevance and ability to influence.

Generate alternatives

Use silent ideation, analogy, sketching, role-play or structured workshops. Invite ideas from technical, operational and domain specialists. Delay evaluation briefly, then apply explicit criteria. Avoid treating an idea count as proof of creativity or quality.

Prototype and test

Match prototype fidelity to the learning objective. A storyboard can test sequence and emotional response; a clickable prototype can test navigation; a service rehearsal can test handoffs. Define what would count as useful evidence before testing. Record not only success rates but also confusion, workarounds, accessibility barriers and unanticipated effects.

Decide and learn

Convert learning into a decision: proceed, revise, investigate further, stop, or transfer the insight to another opportunity. Document assumptions, evidence, confidence and unresolved risks. After launch, monitor outcomes and unintended consequences rather than declaring success based solely on delivery.

9. Applied case: improving appointment preparation

The following is an illustrative case, not a report of a real organisation.

Background

A regional healthcare service observes that many patients arrive unprepared for specialist appointments. Clinicians spend time collecting missing information, appointments run late, and some patients must return for follow-up.

Problem

The initial request is “build a reminder app.” A product manager resists immediate solution commitment and commissions discovery involving patients, carers, reception staff, clinicians and accessibility advisers.

Analysis

Observation reveals several distinct situations. Some patients do not understand which documents are needed; some lack reliable internet access; some have language or literacy barriers; and some receive reminders but cannot act because transport or work arrangements are unresolved. The apparent reminder problem is partly a coordination and preparation problem.

The team creates a service map showing patient actions, reminder channels, reception processes, clinical preparation and data handoffs. It identifies risks: sensitive health information, exclusion of people without smartphones, and the possibility that reminders may create anxiety or imply blame.

Approach

The team frames an opportunity: “How might we help patients understand and complete essential preparation in a way that works across digital and non-digital channels?” It prototypes three approaches: a plain-language preparation checklist, a text-message sequence with a call-back option, and a staff-assisted preparation workflow. Patients and staff test paper and digital versions.

Testing indicates that a single long checklist is confusing. Participants prefer a short sequence with clear reasons, translation support and a way to report barriers. The team conducts a small operational pilot with explicit consent and privacy controls. It measures preparation completeness, staff time, missed appointments, accessibility issues and patient-reported confidence.

Outcome

The pilot does not produce a single app. It produces a coordinated service: concise messages, accessible printed material, staff escalation and improved internal instructions. The product team continues to monitor whether the intervention shifts workload to reception staff or disadvantages people who cannot respond by text.

Learning points

First, the requested feature was not the same as the underlying need. Second, stakeholders outside the nominal user group shaped the service outcome. Third, prototyping the service, not only the interface, exposed operational constraints. Finally, success required balancing desirability, feasibility, viability, accessibility, privacy and clinical safety.

10. Limitations, risks and governance

Design thinking can become superficial when organisations use fashionable language without changing decision practices. Workshops may privilege articulate participants, reward novelty over usefulness or create a false sense of certainty. Interviews can be affected by social desirability bias; prototypes can receive polite approval; and teams can selectively report evidence supporting a preferred concept.

Common failure modes include:

• conducting empathy activities without authority to address discovered problems;
• confusing user preference with demonstrated value;
• treating a persona as a stereotype rather than an evidence-based representation;
• using “fail fast” to excuse preventable harm;
• excluding marginalised people because they are harder to recruit;
• measuring workshop outputs instead of product outcomes; and
• stopping discovery when a senior stakeholder becomes enthusiastic.

Ethical governance should include informed participation, data minimisation, secure handling of research materials, accessibility, non-discrimination, transparency about experimentation and appropriate review for vulnerable populations. In AI-enabled products, teams should additionally examine data provenance, model error across groups, explainability appropriate to the context, human oversight, security and the possibility of automation amplifying existing inequities. AI tools can assist with transcription, synthesis or concept generation, but they do not replace participant consent, contextual interpretation or accountability for decisions.

11. Best practices for product managers

1. Begin with outcomes and contexts, not feature inventories.
2. Make assumptions explicit and rank them by consequence and uncertainty.
3. Use multiple evidence sources rather than relying on a single method.
4. Recruit participants who reflect both intended users and potentially affected non-users.
5. Alternate divergence and convergence deliberately.
6. Prototype the riskiest assumption at the lowest responsible cost.
7. Include engineers, operations, legal, commercial and accessibility expertise early.
8. Define decision criteria before reviewing concepts where practical.
9. Record rejected alternatives and the evidence behind decisions.
10. Continue learning after release through outcome measures, qualitative feedback and monitoring of harms.

Chapter Summary

Design thinking is a disciplined, human-centred approach to making progress under uncertainty. Its origins lie in design research, reflective professional practice and theories of design as a way of shaping preferred futures. It is characterised by problem framing, empathy, collaboration, prototyping, iteration and evidence-informed judgment.

Divergent thinking expands perspectives and possibilities; convergent thinking synthesises and selects. Abductive reasoning supports provisional explanations and interventions when evidence is incomplete. Wicked problems require attention to contested definitions, interconnected systems and unintended consequences.

Design thinking overlaps with human-centred design but is not identical to it. It complements systems thinking, Lean Startup, Agile and product management: systems thinking broadens the view, Lean Startup tests critical assumptions, Agile structures incremental delivery, and product management integrates value and strategy. Effective practice requires ethical governance, inclusion, explicit assumptions and continuous learning. It is not a linear recipe or a substitute for research, engineering, business judgment or accountability.

Key Terms

Design thinking: A human-centred, iterative approach to framing opportunities, generating possibilities, prototyping and learning.

Human-centred design: Design practice that attends to human needs, capabilities, contexts, inclusion and rights.

Systems thinking: Analysis of relationships, feedback, structures and consequences within a wider system.

Empathy: An effort to understand another person’s perspective and context.

Problem framing: Formulating a situation and opportunity to guide investigation and action.

Insight: An interpreted, evidence-supported understanding that informs a decision.

Assumption: A belief treated as true but not yet adequately validated.

Ideation: Generation and development of possible approaches.

Prototype: An incomplete representation created to learn, communicate or test.

Divergence: Expansion of perspectives, questions or possibilities.

Convergence: Synthesis, evaluation and narrowing of possibilities.

Abduction: Reasoning toward a plausible explanation or action from incomplete evidence.

Wicked problem: A complex problem with contested definitions and no final objectively correct solution.

Desirability: The extent to which people value an outcome or solution.

Feasibility: The extent to which a solution can be built and operated.

Viability: The extent to which a solution can create sustainable value.

Lean Startup: An approach emphasising validated learning and experimentation under uncertainty.

Agile: Iterative approaches to collaborative and adaptive product delivery.

Service blueprint: A visual representation connecting user actions with front-stage and back-stage service processes.

Minimum viable product: The smallest responsible product or intervention used to test a critical assumption.

Review Questions

Short Answer Questions

1. What is meant by human-centredness in product discovery?
2. Distinguish divergence from convergence.
3. Why is abductive reasoning important in design work?
4. What makes a problem wicked?
5. How does a prototype support learning?

Long Answer Questions

1. Explain the intellectual and professional origins of design thinking and discuss why it should not be treated as a linear process.
2. Compare design thinking, human-centred design and systems thinking.
3. Explain how design thinking complements Lean Startup and Agile in product management.
4. Discuss the relationship between desirability, feasibility and viability in product decisions.
5. Evaluate the ethical and organisational limitations of design-thinking practice.

Analytical / Application Questions

1. A banking team is asked to build a budgeting dashboard because customers frequently miss loan payments. Use divergent and convergent thinking to propose a discovery plan before approving the dashboard.
2. A prototype receives enthusiastic approval in a workshop but actual adoption is low after launch. Analyse at least four possible explanations and propose evidence-gathering activities.
3. Design an approach for improving access to a public digital service used by people with limited internet access, disabilities and low literacy. Explain how you would combine design thinking with systems thinking and governance safeguards.

Multiple Choice Questions with Answers

Question 1

Which statement best describes design thinking in product management?

A. A fixed sequence that guarantees successful innovation

B. A human-centred and iterative approach to learning and making decisions under uncertainty

C. A software coding standard for Agile teams

D. A financial model for estimating product profitability

Correct Answer: B. A human-centred and iterative approach to learning and making decisions under uncertainty

Question 2

What is the primary purpose of divergent thinking?

A. To select one solution immediately

B. To reduce all stakeholder participation

C. To expand perspectives and possible solutions

D. To measure post-launch revenue

Correct Answer: C. To expand perspectives and possible solutions

Question 3

Which reasoning form proposes a plausible explanation from incomplete evidence?

A. Deduction

B. Abduction

C. Calculation

D. Replication

Correct Answer: B. Abduction

Question 4

Which characteristic is associated with a wicked problem?

A. A universally agreed definition and final solution

B. Stable requirements and a single success criterion

C. Contested definitions and interconnected causes

D. A problem that can only be solved through coding

Correct Answer: C. Contested definitions and interconnected causes

Question 5

How does systems thinking complement design thinking?

A. It eliminates the need for user research

B. It examines relationships, feedback and wider consequences

C. It requires teams to avoid prototyping

D. It focuses only on interface aesthetics

Correct Answer: B. It examines relationships, feedback and wider consequences

Question 6

What is the most appropriate description of a prototype?

A. A fully finished product released to all customers

B. An incomplete representation created to support learning or communication

C. A legal contract with stakeholders

D. A forecast of annual revenue

Correct Answer: B. An incomplete representation created to support learning or communication

Question 7

Which approach primarily emphasises validated learning through experimentation?

A. Lean Startup

B. Systems engineering only

C. Traditional accounting

D. Graphic composition

Correct Answer: A. Lean Startup

Question 8

A product team groups interview observations, identifies patterns and selects one opportunity for further investigation. This is mainly an example of:

A. Convergence

B. Randomisation

C. Deployment

D. Automation

Correct Answer: A. Convergence

Question 9

Why should a product manager distinguish an observation from an interpretation?

A. To make assumptions and alternative explanations visible

B. To prevent any use of qualitative evidence

C. To ensure every product has more features

D. To replace testing with opinion

Correct Answer: A. To make assumptions and alternative explanations visible

Question 10

Which practice is most consistent with responsible design?

A. Recruiting only the easiest participants to reach

B. Ignoring indirect effects after launch

C. Testing only whether a feature is technically possible

D. Considering inclusion, privacy, safety and unintended consequences

Correct Answer: D. Considering inclusion, privacy, safety and unintended consequences