Ideation, Concept Development and Selection

Introduction

Product prototyping does not begin with materials, software, or a workshop. It begins with a decision about what is worth making. Ideation is the disciplined creation of possible solutions; concept development transforms promising ideas into sufficiently defined product concepts; concept selection determines which concepts deserve further investment.

For entrepreneurs, this sequence is particularly important because resources are constrained and early decisions shape later costs. A weak concept may consume months of development, while a premature decision may eliminate a valuable opportunity. Effective product teams therefore alternate between expansion and reduction: they deliberately widen the solution space before narrowing it through evidence, strategic judgement, and explicit criteria.

Ideation is not simply “being creative.” It combines opportunity framing, user understanding, technical reasoning, collaboration, representation, experimentation, and decision governance. This chapter presents a structured approach using divergent and convergent thinking, co-creation, morphological analysis, concept mapping, and selection matrices.

Learning Objectives

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

1. Distinguish divergent ideation from convergent evaluation and explain why both are necessary.
2. Frame a product opportunity as a user-centred design challenge.
3. Facilitate individual and collaborative ideation, including co-creation with users and stakeholders.
4. Use morphological analysis and concept mapping to explore solution spaces systematically.
5. Develop, represent, compare, and refine product concepts.
6. Construct and interpret a weighted concept-selection matrix.
7. Recognise cognitive, ethical, strategic, and organisational risks in concept selection.
8. Recommend a defensible concept-development decision using evidence and assumptions.

1. From Opportunity to Product Concept

An opportunity is a potentially valuable problem, unmet need, or change in the environment. It is not yet a product. A product concept is a coherent proposition describing:

• the target user or customer;
• the problem or job to be addressed;
• the proposed value or benefit;
• the principal functions and user experience;
• the enabling technology or operating logic;
• the delivery, revenue, or service model where relevant; and
• the assumptions that must be tested.

A useful distinction is the following:

A concept should be specific enough to evaluate but not so detailed that untested assumptions appear to be facts. At the early stage, teams should record uncertainty explicitly. For example, “users will trust a colour indicator” is an assumption, not a validated requirement.

The design challenge

A poorly framed challenge restricts ideation. “How can we build a better refrigerator?” directs attention toward appliances and mechanisms. “How might we help households recognise and use food before it becomes waste?” permits behavioural, service, packaging, and technological responses.

A strong challenge statement generally identifies a user, a meaningful outcome, and a context while avoiding commitment to one solution. It should be neither so broad that almost anything qualifies nor so narrow that only one answer is possible.

2. Divergent and Convergent Thinking

Divergent thinking generates variety. It suspends premature judgement and encourages multiple interpretations, user groups, mechanisms, business models, and experiences. Convergent thinking evaluates, combines, improves, and selects alternatives according to explicit criteria.

These modes are complementary rather than sequential in a simple one-time pattern. A team may diverge, converge provisionally, identify a missing assumption, diverge again, and then converge with better information. The process is therefore iterative.

Figure: Divergent and convergent movement from opportunity exploration to concept selection.

Divergent methods

Brainstorming is a rapid verbal generation method. Its effectiveness depends on a clear prompt, time limits, visible recording, and a temporary separation between idea generation and evaluation. Rules such as “one conversation at a time,” “build on others’ ideas,” and “defer criticism” protect participation.

Brainwriting replaces verbal turn-taking with silent written generation. Participants write ideas individually and then exchange or develop them. It reduces the influence of hierarchy and helps less vocal participants contribute.

SCAMPER prompts modification through Substitute, Combine, Adapt, Modify, Put to another use, Eliminate, and Rearrange or Reverse. It is useful when an existing product or process provides a starting point.

Analogy and forced association transfer principles from another domain. A logistics team might ask what a hospital triage system, a beehive, or a streaming service suggests about prioritising deliveries. Analogies stimulate possibilities but must later be tested for technical and contextual fit.

Worst-possible-idea exercises deliberately generate bad solutions and then reverse their characteristics. They can reveal hidden assumptions, such as the assumption that a product must be owned rather than shared.

Sketching and storyboarding externalise ideas. A rough drawing, sequence of use, or service blueprint often reveals interaction problems that remain invisible in discussion.

The output of divergence is not a long list of vague slogans. Each idea should be captured in a consistent form: user, situation, proposed action, expected benefit, and notable assumption.

Convergent methods

Convergence begins with clarification, not immediate scoring. Similar ideas are grouped, duplicates are removed, and incomplete ideas are developed. Teams then apply criteria such as desirability, feasibility, viability, sustainability, safety, strategic fit, and learning value.

A concept should not be rejected solely because it is unfamiliar. Nor should it be retained merely because it was proposed by a senior person. Teams should distinguish:

• evidence, such as observed behaviour or test results;
• assumptions, which require validation;
• preferences, reflecting personal taste; and
• constraints, which are externally imposed or genuinely binding.

A useful practice is to keep several concepts alive until the most critical uncertainties have been investigated. Early convergence should reduce noise, not eliminate productive uncertainty.

3. Co-Creation and Collaborative Ideation

Co-creation involves users, customers, employees, suppliers, domain experts, or other stakeholders in creating or refining value propositions. It differs from simply asking users what product they want. Participants may be unable to articulate latent needs, may describe idealised behaviour, or may optimise for their individual interests. Co-creation is therefore most effective when participants respond to concrete activities, scenarios, prototypes, and trade-offs.

Designing a co-creation activity

A professional session requires more than inviting people to a workshop. The team should:

1. Define the decision or learning objective.
2. Recruit participants who represent relevant behaviours and contexts, not merely convenient demographics.
3. Provide accessible stimulus material, such as journey maps, photographs, role cards, or rough prototypes.
4. Use activities that reveal workarounds, frustrations, priorities, and compromises.
5. Record contributions systematically and obtain consent for recording or reuse.
6. Separate participant statements from the research team’s interpretation.
7. Close the loop by explaining how input affected decisions.

A co-creation session might begin with participants mapping their current experience, continue with “How might we?” prompts, and conclude with concept cards that they rank and improve. The facilitator should avoid leading questions such as “Would you like an app that does X?” Instead, ask “How do you currently handle X?” and “What makes that difficult?”

Benefits and risks

Co-creation can uncover tacit knowledge, increase adoption, reveal accessibility requirements, and identify implementation barriers. However, it can also reproduce group power differences, overrepresent articulate users, or turn participation into unpaid idea extraction. Participants should understand the purpose of the activity, how data will be used, and whether compensation or attribution is appropriate.

For vulnerable groups, the team must apply heightened safeguards. Participation should be voluntary, consent should be informed and ongoing, and sensitive information should be minimised. A participant’s enthusiasm is not evidence of market demand; it is one input into a broader evidence base.

4. Morphological Analysis

Morphological analysis explores combinations of functions and solution principles. The team first decomposes a product opportunity into essential functions, identifies alternative ways to fulfil each function, and combines alternatives into complete concepts.

Consider a personal emergency-alert product. Its functions might include detecting an event, interpreting severity, notifying someone, and supplying power. A simplified morphological chart could be:

One concept could combine manual button, rule-based interpretation, cellular message, and rechargeable power. Another could combine motion sensing, remote interpretation, application notification, and solar assistance.

If a chart has n₁ alternatives for the first function, n₂ for the second, through n_k for k functions, the theoretical number of combinations is:

Here, N is the number of possible combinations, n_i is the number of alternatives for function i, and k is the number of functions. This is a combinatorial upper bound, not a recommendation to build every combination. It assumes each choice can be paired with every other choice; technical incompatibilities, regulations, cost, and user experience may eliminate many combinations.

Morphological analysis is valuable because it prevents fixation on the first familiar architecture. It also makes trade-offs visible. Its limitation is that function decomposition is interpretive: an incomplete or biased list produces an incomplete solution space. Teams should therefore review the chart with technical and user experts.

5. Concept Mapping

A concept map represents relationships among needs, actors, functions, technologies, constraints, and outcomes. Unlike a simple mind map, it should communicate the nature of relationships. For example, “low battery” may constrain “continuous sensing,” while “continuous sensing” may enable “early warning.”

A practical concept map can be built in five stages:

1. Place the user outcome or central problem at the centre.
2. Add actors and contexts around it.
3. Add functions and value mechanisms.
4. Add enabling technologies, dependencies, and constraints.
5. Label relationships such as enables, causes, depends on, conflicts with, or measures.

Concept maps support ideation by showing where a concept depends on an untested assumption. They are also useful for identifying system-level opportunities. For example, a reusable packaging concept may depend not only on container design but also on reverse logistics, cleaning standards, retailer incentives, and customer deposits.

A map is an analytical aid, not proof of causality. Arrows express a proposition that must be investigated. Teams should distinguish direct evidence from hypotheses and mark uncertainty visibly.

6. Developing and Representing Concepts

Ideas become comparable when they are expressed at a similar level of detail. A concept card should usually include:

• concept name and one-sentence proposition;
• target user and use context;
• user problem or desired outcome;
• main interaction or operating sequence;
• essential functions;
• expected benefits and possible harms;
• differentiating attribute;
• key assumptions;
• initial evidence;
• unresolved feasibility or compliance questions; and
• proposed next experiment.

The appropriate representation depends on the question. A sketch tests form and spatial arrangement. A storyboard tests sequence and context. A role-play tests service interaction. A technical mock-up tests mechanism. A clickable interface tests navigation. A spreadsheet or simulation may test economics or capacity. A prototype should be judged by the learning it enables, not by its polish.

Concept development often includes combination and refinement. Two weak concepts may contain complementary strengths; a team can combine them, provided the resulting concept remains coherent. Refinement should not become uncontrolled feature accumulation. Each added feature creates cost, complexity, potential failure, and sometimes regulatory exposure.

7. Structured Concept Selection

Concept selection is a decision under uncertainty. A selection matrix makes criteria and judgements visible, but it does not eliminate subjectivity. It should support discussion and learning rather than create false precision.

Establishing criteria

Criteria should arise from the product strategy, user research, technical realities, and stakeholder obligations. Common criteria include:

• user value or desirability;
• technical feasibility;
• economic viability;
• time to prototype or launch;
• strategic fit;
• scalability;
• environmental performance;
• accessibility and inclusion;
• safety, privacy, and regulatory acceptability; and
• learning potential.

Criteria must be operationalised. “Innovative” is too vague unless the team defines what innovation means in the decision. “Low cost” requires a relevant cost boundary and time horizon. Criteria should be independent where possible; otherwise the same advantage may be counted twice.

Weighted decision matrix

Let concept j receive a score s_ij on criterion i, and let criterion i have weight w_i. If scores are normalised to a common scale and the weights sum to 1, the weighted total is:

Here, T_j is the total score for concept j, m is the number of criteria, w_i is the importance weight of criterion i, and s_ij is concept j’s score on criterion i. The formula assumes that scores are comparable, higher scores consistently indicate better performance, and additive trade-offs are acceptable. If a concept fails a non-negotiable safety or legal requirement, it should normally be screened out rather than rescued by a high total elsewhere.

A five-point scale may be defined as 1 = very weak, 3 = acceptable, and 5 = very strong, with evidence notes for every score. The scale is ordinal in practice, even though weighted addition treats differences as sufficiently comparable. Teams should therefore use sensitivity analysis and discussion, not treat the result as an objective measurement.

Worked example

An entrepreneurial team compares three concepts for a low-cost household food-waste solution. It assigns the following weights:

Scores from 1 to 5 are:

For Concept A:

For Concept B:

For Concept C:

Concept B ranks highest, but the result does not prove that it is the correct product. The team should test whether the reminder service can change behaviour and whether users will provide the required data. It should also examine whether the sustainability score is understated or whether the viability score assumes unrealistic customer acquisition costs.

Reference concepts and selection gates

A Pugh-style comparison evaluates concepts relative to a reference, marking each criterion as better, similar, or worse. It is useful when absolute scores are difficult to justify, but it can anchor the team to the reference concept.

A selection gate asks whether a concept satisfies a minimum condition, such as legal compliance, safe operation, or access to a critical capability. Gates are appropriate for non-compensatory constraints. A high commercial score cannot ethically compensate for unacceptable user risk.

Sensitivity and robustness

A decision is more robust when reasonable changes in assumptions do not reverse it. Teams can vary weights, replace uncertain scores with ranges, or calculate the result under optimistic and conservative scenarios. If small changes reverse the ranking, the appropriate conclusion is not “the matrix failed”; it is that the team needs better evidence or should preserve multiple options.

8. An Integrated Workflow

A practical end-to-end workflow is:

1. Frame the opportunity: define the user, context, outcome, and boundaries.
2. Gather evidence: combine observation, interviews, market knowledge, technical review, and regulatory analysis.
3. Diverge individually and collectively: generate alternatives through several methods.
4. Organise the space: cluster ideas, build a concept map, and construct a morphological chart.
5. Create concept cards: make alternatives comparable and record assumptions.
6. Conduct an initial screen: remove duplicates and concepts that violate non-negotiable constraints.
7. Co-create and prototype: expose leading concepts to users and relevant experts.
8. Evaluate systematically: apply weighted criteria, evidence notes, and sensitivity analysis.
9. Select a portfolio: choose a lead concept, a backup or adjacent concept where justified, and explicit kill criteria.
10. Plan the next experiment: state the riskiest assumption, test method, threshold, owner, and decision date.

This workflow is not a rigid linear recipe. New evidence may change the challenge, criteria, or concept architecture. Good governance records why a decision changed.

9. Illustrative Case: FreshLoop Food-Waste Venture

Background

FreshLoop is an illustrative early-stage venture addressing household food waste. Its founders observed that users often buy food with good intentions but lose track of opened packages and leftovers. The team initially proposed a sensor-equipped storage container.

Problem and analysis

Interviews and home observations suggested three distinct problems: users forgot when food was opened, freshness judgements were uncertain, and reminders often arrived at inconvenient times. The team mapped the system and found dependencies involving container cleaning, battery life, food-safety communication, notification fatigue, and household sharing.

The original “smart container” idea was placed in a morphological chart. Detection options included manual date entry, a printed indicator, and a sensor. Reminder options included a mobile notification, a kitchen display, and a weekly planning message. The chart revealed lower-technology combinations that had not been considered.

Approach

The team generated twelve ideas through brainwriting and a co-creation session with households, a food-safety specialist, and a retailer. It developed four concept cards:

• a sensor-equipped container;
• reusable colour-coded labels with a planning service;
• a household reminder application using manual input; and
• a retailer-linked return and discount programme.

A selection matrix ranked the reminder service highest because it was feasible and inexpensive to test. However, participants expressed low enthusiasm for manual data entry. The team therefore created a hybrid prototype: reusable labels plus rapid voice or barcode input, with optional reminders rather than automatic alerts.

Outcome

A two-week pilot tested whether users would label leftovers and act on reminders. The prototype was intentionally low fidelity. Its purpose was to test behaviour, not validate the final visual design. Results showed that users valued a visible kitchen cue but ignored frequent phone notifications. The team revised the concept toward fewer, user-controlled reminders and clearer food-safety language.

Learning points

1. The first technical solution was not the same as the underlying opportunity.
2. Morphological analysis expanded the design space beyond sensors.
3. Co-creation exposed workflow and trust issues that interviews alone had not revealed.
4. The matrix supported prioritisation but did not replace behavioural testing.
5. A low-fidelity prototype generated more valuable learning than an expensive technical demonstrator at that stage.

10. Challenges, Ethics, and Contemporary Practice

Cognitive and organisational bias

Anchoring, confirmation bias, availability bias, status effects, and groupthink can distort selection. Teams can reduce these effects through independent scoring before discussion, anonymous idea submission, rotating facilitation, explicit evidence logs, and deliberate inclusion of disconfirming evidence. Diversity is valuable not as a symbolic goal but because different experiences expose different assumptions and risks.

Feasibility, viability, and desirability tensions

A desirable concept may be infeasible; a feasible concept may lack a compelling user benefit; a viable concept may create unacceptable social or environmental harm. These dimensions should be examined together. A business case that externalises pollution, surveillance, or unpaid labour is not a complete viability assessment.

Safety, accessibility, privacy, and sustainability

Concept decisions can create foreseeable harms. Teams should consider physical safety, cybersecurity, data minimisation, accessibility, repairability, material sourcing, end-of-life recovery, and unequal effects on different groups. Privacy should be designed into the concept rather than added after architecture decisions have been made. Claims about sustainability should identify the system boundary and evidence; “recyclable” does not necessarily mean that collection and processing are available in the target market.

Generative AI in ideation

Generative AI can rapidly produce variants, analogies, personas, interface drafts, and alternative framings. It may be useful for divergence, especially when prompts specify constraints and ask for contrasting approaches. However, generated ideas can be derivative, factually unreliable, culturally biased, or based on copyrighted or confidential material. AI output should be treated as unverified stimulus. Teams remain responsible for originality checks, privacy, attribution, human participation, and decisions affecting safety or rights. Sensitive customer data should not be entered into an unapproved system.

Best practices

Professional teams should:

• define the decision before choosing a method;
• separate generation from evaluation while allowing iterative return;
• use several ideation techniques rather than one workshop ritual;
• represent concepts at comparable levels of detail;
• record assumptions and evidence with every score;
• apply gates to legal, safety, and ethical constraints;
• use prototypes to answer specific questions;
• perform sensitivity analysis on important decisions;
• preserve valuable alternatives when uncertainty is high; and
• communicate decisions, dissent, and next tests transparently.

Chapter Summary

Ideation, concept development, and selection form a central reasoning system in product prototyping. Divergent methods enlarge the opportunity space, while convergent methods organise and reduce it. Co-creation contributes contextual knowledge and can improve adoption, but it requires careful facilitation, inclusion, consent, and interpretation. Morphological analysis decomposes functions and recombines solution principles; concept mapping exposes relationships, dependencies, and assumptions. Concept cards and fit-for-purpose prototypes make alternatives comparable and testable.

Structured selection matrices clarify criteria and priorities. A weighted total can support comparison when scores are comparable and trade-offs are acceptable, but it must not create false certainty. Non-negotiable safety, legal, and ethical requirements should operate as gates. Sensitivity analysis reveals whether a decision is robust, while explicit evidence logs reduce bias. Contemporary practice increasingly combines low-fidelity experimentation, stakeholder participation, sustainability thinking, and carefully governed AI assistance. The strongest concept is not necessarily the most technically impressive; it is the concept whose value, feasibility, viability, responsibility, and learning path are best supported by evidence.

Key Terms

Opportunity: A potentially valuable problem, need, or change that may support a product or venture.

Product concept: A coherent proposition describing a target user, value, functions, and operating logic.

Divergent thinking: Expansion of the solution space through generation of varied possibilities.

Convergent thinking: Evaluation, combination, refinement, and reduction of alternatives.

Brainstorming: Rapid group idea generation with deferred judgement.

Brainwriting: Written or silent idea-generation method that distributes participation.

Co-creation: Collaborative creation or refinement of value with users or stakeholders.

Morphological analysis: Systematic exploration of combinations of functions and solution principles.

Morphological chart: A table listing alternative principles for each product function.

Concept map: A labelled representation of relationships among needs, functions, actors, and constraints.

Concept card: A structured description used to make an early concept understandable and comparable.

Assumption: An unverified belief on which a concept or decision depends.

Prototype: A representation created to learn about a product, service, or assumption.

Selection criterion: A dimension used to compare product concepts.

Weighted decision matrix: A comparison tool that combines criterion weights and concept scores.

Selection gate: A minimum requirement that a concept must satisfy.

Sensitivity analysis: Examination of how a decision changes when assumptions or inputs vary.

Desirability: The extent to which users value and are willing to adopt a solution.

Feasibility: The extent to which a solution can be technically and operationally delivered.

Viability: The extent to which a solution can sustain an economically and strategically sound venture.

Review Questions

Short Answer Questions

1. What is the difference between an opportunity and a product concept?
2. Why should divergent and convergent thinking be used together?
3. State two benefits and two risks of co-creation.
4. What is the purpose of a morphological chart?
5. Why should a selection matrix not be treated as an objective answer?

Long Answer Questions

1. Explain how a team can move from an opportunity statement to a set of comparable concept cards.
2. Describe the stages of a well-designed co-creation session and discuss how facilitator bias can be reduced.
3. Explain morphological analysis, including its strengths, assumptions, and limitations.
4. Discuss how desirability, feasibility, viability, sustainability, and safety may conflict during concept selection.
5. Explain how weighted matrices, selection gates, and sensitivity analysis can be integrated into a responsible decision process.

Analytical / Application Questions

1. Construct a morphological chart for a product that helps university students manage personal medication adherence. Identify at least four functions and three alternatives for each.
2. Using the weighted-matrix method, compare three concepts for a campus food-ordering service. Define at least five criteria, assign weights summing to 1, score each concept, and explain whether the ranking is robust.
3. A team has selected a technically advanced concept because it received the highest total score, but user testing shows low trust and high effort. Diagnose the decision failure and propose a revised ideation, prototyping, and selection plan.

Multiple Choice Questions with Answers

Question 1

Which activity is primarily divergent?

A. Ranking concepts against weighted criteria

B. Generating multiple alternative solution principles

C. Applying a legal compliance gate

D. Selecting a lead concept

Correct Answer: B. Generating multiple alternative solution principles

Question 2

What is the main purpose of a morphological chart?

A. To calculate market share

B. To document project expenses

C. To combine alternatives across product functions

D. To replace user research

Correct Answer: C. To combine alternatives across product functions

Question 3

Which statement best describes co-creation?

A. Experts make all design decisions for users

B. Users and other stakeholders help create or refine value propositions

C. A company copies a competitor's product

D. A team evaluates only technical performance

Correct Answer: B. Users and other stakeholders help create or refine value propositions

Question 4

In a weighted decision matrix, what does a criterion weight represent?

A. The number of prototypes built

B. The relative importance assigned to that criterion

C. The manufacturing mass of the product

D. The number of users interviewed

Correct Answer: B. The relative importance assigned to that criterion

Question 5

If criterion weights are 0.4 and 0.6, and a concept scores 3 and 5 respectively on a common scale, what is its weighted total?

A. 3.8

B. 4.0

C. 4.2

D. 4.6

Correct Answer: C. 4.2

Question 6

Which condition most appropriately requires a selection gate rather than compensation through a high total score?

A. Aesthetic preference

B. Minor difference in learning potential

C. Failure to meet a mandatory safety requirement

D. Uncertainty about colour choice

Correct Answer: C. Failure to meet a mandatory safety requirement

Question 7

What should a concept card primarily do?

A. Present a final manufacturing specification

B. Make an early concept understandable and comparable

C. Guarantee commercial success

D. Eliminate all assumptions

Correct Answer: B. Make an early concept understandable and comparable

Question 8

Why is sensitivity analysis useful after a concept matrix is scored?

A. It proves that all scores are objective

B. It tests whether reasonable changes in assumptions alter the decision

C. It removes the need for prototypes

D. It ensures every concept receives the same score

Correct Answer: B. It tests whether reasonable changes in assumptions alter the decision

Question 9

Which is an appropriate use of a low-fidelity prototype?

A. Testing a specific user or workflow assumption cheaply

B. Demonstrating that regulatory approval has been obtained

C. Proving long-term manufacturing reliability

D. Replacing all market evidence

Correct Answer: A. Testing a specific user or workflow assumption cheaply

Question 10

Which statement about generative AI in ideation is most responsible?

A. Its outputs can be accepted without verification

B. It should only be used after product launch

C. It can provide stimulus for divergence, but outputs require human validation and governance

D. It removes the need for user participation

Correct Answer: C. It can provide stimulus for divergence, but outputs require human validation and governance