Design Thinking Models and Innovation Processes

Introduction

Entrepreneurial opportunities rarely arrive as fully defined problems with obvious solutions. They emerge from incomplete information, changing technologies, institutional constraints and the lived experiences of users. Design thinking provides a disciplined way to work with this uncertainty. It combines inquiry, interpretation, imagination, experimentation and learning so that entrepreneurs can develop solutions that are desirable to people, feasible to deliver and viable as ventures.

Design thinking is not a single universal method. It is better understood as a family of related models. The Double Diamond, Stanford d.school, IDEO human-centred design, Design Council approaches and broader human-centred design frameworks use different language and sequencing, yet share several commitments: understand people in context, frame the right problem, generate alternatives, learn through prototypes and improve solutions through iteration.

For postgraduate entrepreneurship students, the important task is not memorising stages. It is learning to select and adapt a model, recognise when to diverge or converge, reason abductively from incomplete evidence, and establish reflective practices that improve both the idea and the entrepreneurial team.

Learning Objectives

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

1. Explain the assumptions and stages of major design thinking models.
2. Compare the Double Diamond, Stanford d.school, IDEO, Design Council and human-centred design approaches.
3. Distinguish divergent thinking, convergent thinking, deductive reasoning, inductive reasoning and abductive reasoning.
4. Apply iteration and reflective practice to entrepreneurial opportunity development.
5. Select an appropriate process for a specific innovation context.
6. Evaluate ethical, organisational and commercial implications of design-led innovation.

Design Thinking as an Innovation Logic

A conventional analytical approach often begins with a defined problem, identifies alternatives and selects the best one. Design thinking is particularly useful when the problem itself is uncertain. It treats problem definition as a design activity rather than as a fixed starting point.

Three characteristics are central:

• Human orientation: Decisions are informed by people’s goals, abilities, constraints, emotions and social contexts.
• Learning through making: Prototypes make assumptions visible and enable evidence to be gathered before substantial investment.
• Iterative movement: Teams move repeatedly between understanding, framing, imagining and testing rather than following a strictly linear sequence.

A useful entrepreneurial interpretation is that a venture idea contains a set of hypotheses. These may concern a customer’s problem, the value proposition, the delivery channel, willingness to pay, operational feasibility and social consequences. Design thinking helps formulate and test these hypotheses. It does not, by itself, guarantee market demand, technical feasibility or financial viability; those require complementary market, operational and financial analysis.

Important terminology

• User or participant: A person affected by, using or contributing to a product, service or system.
• Stakeholder: Any party that influences or is affected by the innovation, including non-users.
• Insight: An evidence-based interpretation that explains a meaningful need, behaviour or tension.
• Point of view: A concise framing of a user, need and significant insight.
• Design challenge: An open-ended question that directs ideation without prescribing a solution.
• Prototype: A representation of an idea created to learn, communicate or test.
• Assumption: A proposition treated as true but not yet adequately supported by evidence.
• Iteration: A deliberate cycle of making, testing, learning and revising.
• Minimum viable product (MVP): The smallest version of a product that can generate valid learning about a specified value hypothesis. It is not necessarily the cheapest or least functional product.

A Common Cognitive Architecture

Although the named models differ, most design processes contain four recurring cognitive movements.

Discover and understand

The team investigates people, environments, existing practices, technologies and institutions. Methods include observation, contextual interviews, diary studies, journey mapping, desk research and analysis of existing service data. The purpose is not simply to collect opinions. It is to understand what people do, why they do it, what constrains them and what workarounds they have developed.

Define and frame

Raw observations are organised into patterns and implications. A team may cluster observations, identify tensions, construct personas or formulate a point of view. Framing determines what counts as relevant evidence and what solutions become imaginable. A narrow frame can exclude important stakeholders; an excessively broad frame can make action impossible.

Develop and imagine

The team generates multiple possible responses. Divergence is encouraged before evaluation. Ideas may involve products, services, business models, policies, partnerships or changes to an existing process. At this stage, quantity is useful because early ideas are often incomplete and because combinations of ideas can reveal stronger possibilities.

Deliver and learn

Ideas are represented at an appropriate level of fidelity, tested with relevant participants and revised. Delivery may mean a pilot, service trial, landing page, role-play, physical model or software experiment. Learning may invalidate the initial concept, refine it or reveal a new opportunity.

These movements should not be interpreted as four rigid phases. A prototype can reveal a new user need; a stakeholder interview can expose a flawed business assumption; and a delivery pilot can require reframing the original problem.

Divergent and Convergent Thinking

Divergent thinking expands the field of possibilities. It asks: “What might be possible?” It is supported by brainstorming, brainwriting, analogy, extreme-user exploration, bodystorming and combination exercises. Psychological safety matters because premature criticism suppresses unusual but potentially valuable ideas.

Convergent thinking narrows possibilities using explicit criteria. It asks: “Which direction deserves further investment?” Criteria may include user value, strategic fit, technical feasibility, regulatory acceptability, environmental effect, accessibility and economic viability.

Effective innovation alternates between the two. Divergence without convergence produces an unmanageable collection of ideas. Convergence without sufficient divergence merely optimises familiar solutions.

A practical method is to separate stages of an ideation session:

1. Clarify the design challenge and evidence.
2. Generate ideas individually and collectively.
3. Cluster ideas into themes.
4. Select promising concepts using agreed criteria.
5. Build low-cost prototypes.
6. Test and reopen divergence if evidence challenges the frame.

Voting is not a substitute for judgement. A popular idea may be attractive but infeasible, while a less popular idea may address a critical unmet need. Selection should therefore combine structured criteria with informed discussion.

Abductive Reasoning in Entrepreneurship

Deduction applies a general rule to a particular case. Induction generalises from observations. Abduction proposes the most plausible explanation or course of action from incomplete evidence. It is often described as reasoning toward the best available explanation.

For example:

• Observation: commuters repeatedly carry food from home despite nearby cafés.
• Possible explanation: the cafés are too expensive, too slow, perceived as unhealthy or poorly matched to dietary needs.
• Abductive design question: what explanation best accounts for the behaviour, and what small experiment could distinguish among explanations?

Abduction is essential because entrepreneurs act before certainty is available. However, an abductive hypothesis is not a fact. It must be expressed as a proposition that can be investigated. Teams should record the evidence, the interpretation, competing explanations and the proposed test. This prevents a compelling story from being mistaken for validated knowledge.

The Double Diamond

The Double Diamond was popularised by the UK Design Council. It represents innovation as two connected diamonds. The first concerns doing the right problem; the second concerns developing the right response. Each diamond expands and contracts: Discover, Define, Develop and Deliver.

• Discover: Explore the situation broadly through research and engagement.
• Define: Synthesise evidence and articulate a focused design brief.
• Develop: Generate and develop multiple concepts.
• Deliver: Test, refine and implement selected solutions.

The two diamonds are significant because they distinguish problem framing from solution development. A team may discover that its initial brief was wrong and return to discovery. The model therefore depicts a pattern of activity rather than a mandatory linear workflow.

The Double Diamond is valuable for communicating a project to sponsors and multidisciplinary teams. Its limitation is that its broad labels do not prescribe specific research methods, decision rules or commercial tests. Teams must add these themselves.

The Stanford d.school Model

The Stanford d.school commonly presents five modes: Empathize, Define, Ideate, Prototype and Test. These modes are often shown sequentially, but the d.school emphasises that design is iterative and that teams can move backward or revisit earlier modes.

• Empathize: Seek an informed understanding of people and their experiences.
• Define: Develop a meaningful and actionable problem statement.
• Ideate: Create a wide range of possible solutions.
• Prototype: Build representations that enable learning.
• Test: Obtain feedback and refine the understanding or solution.

This model is especially accessible for teaching and workshop facilitation. It makes the relationship between empathy and problem framing explicit. Its limitation is that “empathy” can be interpreted superficially as simply being nice or asking users what they want. Rigorous empathy requires observation, contextual interpretation, attention to power and triangulation with other evidence.

IDEO and Human-Centred Design

IDEO’s methods emphasise designing with people rather than merely for them. IDEO’s human-centred design resources have often organised work around inspiration, ideation and implementation, while the organisation’s broader practice uses research, synthesis, prototyping and implementation in flexible combinations.

• Inspiration: Learn from the people and contexts involved.
• Ideation: Identify patterns, frame opportunities and create possibilities.
• Implementation: Move promising ideas into pilots, operational plans and sustainable delivery.

IDEO’s approach is notable for its practical orientation, cross-disciplinary teamwork, visualisation and rapid prototyping. It treats implementation as part of design rather than an afterthought. In entrepreneurship, this is important because a desirable concept that cannot be delivered through a viable operating model is incomplete.

A limitation is the risk of importing workshop techniques without the deeper research discipline. Methods such as brainstorming or personas are not automatically human-centred; their quality depends on evidence, inclusion and how decisions are made.

Design Council and the Framework for Innovation

The Design Council’s Framework for Innovation develops the Double Diamond and identifies four principles: put people first, communicate visually and inclusively, collaborate and co-create, and iterate, iterate, iterate. It also describes activities associated with each diamond and stresses that the framework is not a rigid prescription.

The framework is particularly useful for organisational innovation because it links design activity with collaboration, systems awareness and implementation. It encourages teams to consider the wider system, including policy, supply chains, environmental impact, staff experience and unintended consequences.

Human-Centred Design as an Orientation

Human-centred design is broader than any single five- or four-stage process. It is an orientation in which the needs, capabilities, dignity and contexts of affected people shape design decisions throughout the lifecycle. It includes accessibility, participation, cultural awareness, safety and accountability.

Human-centred design does not mean that every user request should be implemented. Users may describe an immediate preference without identifying the underlying need; different groups may have conflicting interests; and a popular feature may create harm for non-users. The designer’s responsibility is to interpret evidence carefully and consider the whole system.

Related approaches include participatory design, inclusive design, service design and systems thinking. Participatory design gives affected people greater influence over decisions. Inclusive design seeks to accommodate diverse abilities and circumstances. Service design examines interactions across frontstage and backstage processes. Systems thinking considers relationships, feedback loops and consequences beyond a single transaction.

Comparing the Models

The models are complementary rather than mutually exclusive. An entrepreneurial team might use the Double Diamond to communicate the overall programme, d.school modes to structure a workshop, IDEO-style methods to conduct field research and prototyping, and human-centred design principles to govern inclusion and ethics.

Figure: The Double Diamond shows divergence and convergence in problem and solution spaces.

Iteration and Reflective Practice

Iteration is not repeated activity for its own sake. Each cycle should have a learning purpose. Before an experiment, the team should state:

• the assumption being examined;
• the evidence that would support or challenge it;
• the prototype or research activity;
• the participants and context;
• the decision that will follow from the result.

A useful reflection occurs at three levels. Reflection-in-action means noticing and adjusting while an interview, workshop or prototype test is occurring. Reflection-on-action means reviewing what happened after the activity. Reflection-for-action means using that learning to redesign the next cycle.

Teams should maintain an assumption-and-learning log. A simple classification is:

Reflection should include the team’s own behaviour. Did dominant voices control the interpretation? Were participants representative of those affected? Did the prototype invite useful criticism or merely seek approval? Were negative findings dismissed because of sunk-cost pressure?

A Practical Innovation Process for Entrepreneurs

The following integrated process combines the models without treating them as a rigid recipe.

1. Establish the opportunity area

Define the broad context, affected groups, strategic intent and constraints. Avoid assuming that the initial request is the real problem.

2. Conduct contextual inquiry

Use several sources of evidence. Observe behaviour where possible, interview different stakeholders and examine existing data. Record direct observations separately from interpretations.

3. Synthesise and frame

Cluster evidence, identify tensions and formulate a point of view or design challenge. A strong challenge is open enough to permit alternatives but bounded enough to guide action. “How might we help first-time users complete a complex task with confidence?” is more productive than “How might we build an app?”

4. Generate alternatives

Use divergent techniques and include concepts at different system levels: a feature, a service redesign, a partnership, a policy change and a new revenue model. Make constraints explicit without allowing them to eliminate all imaginative exploration.

5. Select learning priorities

Prioritise assumptions by their uncertainty and consequence. A technically difficult feature may be less urgent than an untested belief that customers will pay. Select experiments that produce decision-relevant evidence quickly and ethically.

6. Prototype and test

Match fidelity to the question. A storyboard can test desirability and sequence; a role-play can test service interactions; a clickable interface can test navigation; a pilot can test operational coordination. Do not use a polished prototype to conceal unresolved assumptions.

7. Integrate venture analysis

Examine the value proposition, customer segments, channels, resources, partners, costs, revenues and risks. Design evidence should be connected to market and operational evidence. A positive user reaction is not equivalent to willingness to pay.

8. Decide and scale cautiously

Possible decisions include persevere, pivot, pause, partner, narrow the segment or stop. Scaling should follow evidence about repeatability, safety, economics and capacity, not enthusiasm alone.

Applied Case: FreshRoute Community Food Service

The following is a fictional illustrative case designed for educational use.

Background

FreshRoute is a proposed venture intended to improve access to affordable, nutritious meals for postgraduate students living near a metropolitan campus. The founding team initially planned a mobile application through which students could order discounted meals from local restaurants.

Problem and discovery

Using a Double Diamond structure, the team began with discovery rather than building the application. It observed food-purchasing behaviour at three times of day, interviewed students, restaurant staff and campus welfare personnel, and reviewed anonymised campus survey data. Students reported wanting affordability and convenience, but interviews revealed additional issues: unpredictable class schedules, dietary requirements, limited refrigeration in shared housing and reluctance to wait in queues. Restaurants were concerned about unpredictable demand and packaging costs.

The initial assumption—“students need another ordering app”—was challenged. The more plausible opportunity was a coordinated, flexible meal service that reduced uncertainty for both students and restaurants.

Framing and ideation

The team formulated the challenge: “How might we enable students with irregular schedules and varied dietary needs to obtain dependable, affordable meals while giving local kitchens more predictable demand?”

During ideation, the team generated concepts including pre-order collection lockers, subscription meals, a shared menu with dietary filters, surplus-food bundles and scheduled group delivery. Convergence using user value, operational feasibility, inclusion, environmental impact and revenue potential produced two concepts for prototyping: a weekly pre-order menu and a campus collection point.

Prototyping and testing

The first prototype was a paper menu and booking form. It tested whether students would select meals several days in advance. The result showed interest but also exposed a need for late changes and transparent allergen information. The second prototype was a one-week manual service using a spreadsheet, text messages and a staffed collection table. This low-fidelity prototype tested operations without the cost of software development.

The pilot revealed that collection worked for students after evening classes but not for those attending daytime sessions. It also showed that restaurant preparation improved when orders were closed at a fixed time. However, one dietary label was ambiguous, creating a safety concern. The team paused expansion, consulted qualified food-safety advisers and redesigned its information and verification process.

Outcome

FreshRoute did not immediately launch the proposed app. It narrowed the initial segment to evening postgraduate students, established a limited menu, introduced a clear allergen protocol and negotiated a revenue-sharing arrangement with two restaurants. A later pilot generated stronger evidence about repeat use and operational cost. The venture’s first innovation was therefore not a software product but a tested service system.

Learning points

1. Discovery changed the problem definition.
2. Abductive reasoning generated competing explanations rather than one premature answer.
3. Low-fidelity prototypes reduced financial and technical risk.
4. Convergence required ethical and operational criteria, not popularity alone.
5. Testing exposed a safety issue that would have been more costly after scaling.
6. Human-centred design included restaurant workers and people with dietary risks, not only the paying customer.

Professional, Ethical and Governance Considerations

Design decisions distribute benefits, burdens and risks. Teams should ask who is represented in research, who is absent, who controls data and who bears the consequences of failure. In digital ventures, relevant issues include informed consent, privacy, data minimisation, accessibility, algorithmic bias, explainability and cybersecurity. In health, finance, education, food and mobility, regulatory obligations may apply before a pilot is launched.

Inclusive recruitment should avoid treating a convenient sample as universal. Participants should understand the purpose of research and how recordings, quotations or personal data will be used. Compensation should be fair and should not create undue pressure. Sensitive groups may require additional safeguards and specialist review.

AI can support design by helping organise interview transcripts, generate alternative concepts, simulate scenarios or create rapid interface prototypes. It can also reproduce bias, hallucinate evidence, expose confidential data or create a false impression of understanding. AI-generated summaries should be checked against source material, and confidential research data should not be entered into systems without appropriate governance. Human accountability remains essential, especially where decisions affect access, safety or rights.

Challenges and Limitations

Design thinking is sometimes criticised for being presented as universally applicable, culturally neutral or inherently benevolent. Several limitations deserve attention:

• Method worship: Following a named process can replace judgement and domain expertise.
• Token participation: Inviting users to a workshop does not necessarily give them influence.
• Small-sample overconfidence: Qualitative research can reveal meaning but cannot alone estimate market size or prevalence.
• Novelty bias: Teams may favour visibly innovative concepts over effective incremental improvements.
• Implementation gap: A compelling prototype may fail under procurement, regulation, staffing or cost constraints.
• Power and exclusion: The people with authority to approve a project may not be those most affected by it.
• Iteration fatigue: Endless experimentation can delay necessary decisions unless learning thresholds and stopping rules are defined.

Strong practice responds by combining design research with domain knowledge, quantitative analysis, financial modelling, technical assessment, legal review and responsible governance.

Best Practices for Postgraduate Entrepreneurs

1. Treat every early idea as a hypothesis, not a commitment.
2. Separate observations, interpretations and decisions in project records.
3. Include non-users and indirectly affected stakeholders.
4. Alternate divergence and convergence deliberately.
5. Choose prototype fidelity according to the question being tested.
6. Test behaviour and trade-offs, not only stated enthusiasm.
7. Define decision criteria before results are known where feasible.
8. Document disconfirming evidence and rejected alternatives.
9. Connect user value to feasibility, viability, sustainability and ethics.
10. Make reflection a scheduled team practice rather than an informal afterthought.

Chapter Summary

Design thinking is a family of human-oriented, iterative approaches for addressing uncertain problems and developing innovations. The Double Diamond distinguishes expansion and contraction in problem and solution spaces. The Stanford d.school model highlights empathize, define, ideate, prototype and test. IDEO approaches connect inspiration, ideation and implementation, while the Design Council emphasises principles such as collaboration, visual communication, inclusion and repeated iteration. Human-centred design is the wider orientation that places people, context, dignity and consequences at the centre.

Divergent thinking expands possibilities; convergent thinking selects and develops directions. Abductive reasoning helps entrepreneurs form plausible explanations from incomplete evidence, but those explanations must be tested. Iteration is valuable when each cycle has a clear learning purpose, and reflective practice improves both the innovation and the team’s judgement. No model replaces market, technical, financial, ethical or regulatory analysis. Effective entrepreneurial design integrates all of these forms of reasoning while remaining open to reframing.

Key Terms

Design thinking: An iterative approach to innovation that combines human understanding, framing, ideation, prototyping and learning.

Double Diamond: A Design Council model comprising discover, define, develop and deliver.

Empathy: An informed understanding of people’s experiences, perspectives and contexts.

Human-centred design: An orientation that places affected people, their needs, capabilities and dignity at the centre of design decisions.

Divergent thinking: Expanding the range of possible interpretations or solutions.

Convergent thinking: Narrowing alternatives using evidence and explicit criteria.

Abductive reasoning: Developing the most plausible explanation or action from incomplete evidence.

Inductive reasoning: Generalising patterns from particular observations.

Deductive reasoning: Applying a general rule or proposition to a particular case.

Design challenge: An open-ended question that directs design activity without prescribing a solution.

Insight: An evidence-based interpretation of a meaningful need, behaviour or tension.

Prototype: A representation of an idea used for learning, communication or testing.

Iteration: A cycle of making, testing, learning and revising.

Assumption: A proposition treated as true but not adequately verified.

Stakeholder: A person or organisation affected by or able to influence an innovation.

Participatory design: An approach that gives affected people an active role in design decisions.

Service design: Design that examines user interactions and the organisational processes supporting them.

Minimum viable product: The smallest product version capable of generating valid learning about a specified value hypothesis.

Reflective practice: Deliberate examination of action and experience to improve future judgement and activity.

Review Questions

5 Short Answer Questions

1. What is the primary distinction between the two diamonds in the Double Diamond?
2. Define divergent thinking and give one example of a divergent technique.
3. Why is abductive reasoning important in entrepreneurship?
4. What makes a prototype useful even when it is not a finished product?
5. State two ethical considerations in human-centred design.

5 Long Answer Questions

1. Compare the Double Diamond and Stanford d.school models in terms of stages, strengths and limitations.
2. Explain how IDEO’s inspiration–ideation–implementation logic can support venture development.
3. Discuss why human-centred design is an orientation rather than merely a sequence of steps.
4. Explain how divergent and convergent thinking should be balanced during opportunity development.
5. Analyse the role of reflection in improving the quality of iterative innovation.

3 Analytical / Application Questions

1. A start-up wants to develop a digital platform for independent caregivers but has only a preliminary feature list. Design a discovery, framing, ideation and testing process, and identify three assumptions that should be tested first.
2. A prototype receives enthusiastic comments from workshop participants, but few agree to use it repeatedly or pay for it. Analyse what this evidence does and does not establish, and recommend the next experiments.
3. Apply the Double Diamond and Stanford d.school models to a campus sustainability problem. Explain where the models would overlap, where they would differ and how you would incorporate inclusion and regulatory review.

Multiple Choice Questions with Answers

Question 1

Which activity belongs primarily to the Define stage of the Double Diamond?

A. Synthesising research into a focused design brief

B. Generating as many solution concepts as possible

C. Launching the final product at scale

D. Conducting a technical maintenance review

Correct Answer: A. Synthesising research into a focused design brief

Question 2

What is the principal purpose of divergent thinking?

A. To select one final idea immediately

B. To expand the range of possible interpretations or solutions

C. To prove that a business model is profitable

D. To eliminate stakeholder participation

Correct Answer: B. To expand the range of possible interpretations or solutions

Question 3

Which sequence represents the commonly taught Stanford d.school modes?

A. Discover, Define, Develop, Deliver

B. Inspire, Ideate, Implement, Institutionalise

C. Empathize, Define, Ideate, Prototype, Test

D. Research, Finance, Manufacture, Distribute, Audit

Correct Answer: C. Empathize, Define, Ideate, Prototype, Test

Question 4

An entrepreneur observes that customers avoid a service and proposes that confusing pricing may explain the behaviour. This is primarily an example of which reasoning type?

A. Deductive reasoning

B. Abductive reasoning

C. Algorithmic reasoning

D. Descriptive accounting

Correct Answer: B. Abductive reasoning

Question 5

Why should prototype fidelity be matched to the learning question?

A. High fidelity is always ethically required

B. Low fidelity prevents any useful feedback

C. The appropriate representation can test the assumption without unnecessary investment

D. Fidelity determines the legal ownership of an idea

Correct Answer: C. The appropriate representation can test the assumption without unnecessary investment

Question 6

Which statement best describes convergence?

A. It narrows alternatives using evidence and explicit criteria

B. It prohibits evaluation of ideas

C. It requires building the most expensive prototype

D. It means asking only existing customers for ideas

Correct Answer: A. It narrows alternatives using evidence and explicit criteria

Question 7

In the FreshRoute case, what changed as a result of discovery?

A. The team confirmed that a new app was the only solution

B. The team reframed the opportunity as a coordinated and flexible meal service

C. The team abandoned all stakeholder research

D. The team scaled nationally before conducting a pilot

Correct Answer: B. The team reframed the opportunity as a coordinated and flexible meal service

Question 8

Which statement about human-centred design is most accurate?

A. It means implementing every request made by a user

B. It focuses only on the paying customer

C. It considers people, context, capabilities, dignity and wider consequences

D. It eliminates the need for technical and financial analysis

Correct Answer: C. It considers people, context, capabilities, dignity and wider consequences

Question 9

What is the main risk of treating a design model as a rigid linear recipe?

A. It may prevent teams from reframing the problem when new evidence appears

B. It guarantees too much stakeholder participation

C. It makes prototypes impossible to create

D. It automatically increases regulatory compliance

Correct Answer: A. It may prevent teams from reframing the problem when new evidence appears

Question 10

Which practice most directly supports reflective innovation?

A. Recording only successful experiments

B. Separating observations, interpretations and decisions in project records

C. Avoiding negative participant feedback

D. Choosing criteria after seeing the results

Correct Answer: B. Separating observations, interpretations and decisions in project records